~ubuntu-branches/ubuntu/maverick/webkit/maverick

(KJS::UString::UString): If calls to expandCapacity, expandPreCapacity or fastMalloc fail then use the null representation rather than crashing in memcpy.

366

(KJS::UString::append): Ditto.

367

(KJS::UString::operator=): Ditto.

368

* kjs/ustring.h: Change return type of expandedSize from int to size_t.

369

370

2007-08-01 Darin Adler <darin@apple.com>

371

372

Reviewed by Kevin McCullough.

373

374

- fix <rdar://problem/5375186> pointers to pieces of class definition passed to JSClassCreate should all be const

375

376

* API/JSObjectRef.h: Added const.

377

378

* API/JSClassRef.cpp:

379

(OpaqueJSClass::OpaqueJSClass): Added const.

380

(OpaqueJSClass::create): Added const.

381

* API/JSObjectRef.cpp:

382

(JSClassCreate): Added const.

383

384

2007-08-01 Steve Falkenburg <sfalken@apple.com>

385

386

Build mod: Fix sln to match configs in vcproj.

387

388

Reviewed by Adam.

389

390

* JavaScriptCore.vcproj/JavaScriptCore.make:

391

* JavaScriptCore.vcproj/JavaScriptCore.sln:

392

393

2007-07-30 Simon Hausmann <hausmann@kde.org>

394

395

Done with and reviewed by Lars.

396

397

Removed the __BUILDING_QT ifdef in JSStringRef.h and changed UChar for the Qt build to use wchar_t on Windows.

398

399

* API/JSStringRef.h:

400

* wtf/unicode/qt4/UnicodeQt4.h:

401

402

2007-07-27 Simon Hausmann <hausmann@kde.org>

403

404

Done with and reviewed by Lars and Zack.

405

406

Always define JSChar to be unsigned short for the Qt builds, to ensure compatibility with UChar.

407

408

* API/JSStringRef.h:

409

410

2007-07-27 Simon Hausmann <hausmann@kde.org>

411

412

Done with and reviewed by Lars and Zack.

413

414

Fix compilation with Qt on Windows with MingW: Implemented currentThreadStackBase() for this platform.

415

416

* kjs/collector.cpp:

417

(KJS::currentThreadStackBase):

418

419

2007-07-27 Simon Hausmann <hausmann@kde.org>

420

421

Done with and reviewed by Lars and Zack.

422

423

Fix compilation with Qt on Windows with MingW: The MingW headers do not provide a prototype for a reentrant version of localtime. But since we don't use multiple threads for the Qt build we can use the plain localtime() function.

424

425

* kjs/DateMath.cpp:

426

(KJS::getDSTOffsetSimple):

427

428

2007-07-27 Simon Hausmann <hausmann@kde.org>

429

430

Done with and reviewed by Lars and Zack.

431

432

Use $(MOVE) instead of mv to eliminated the shell dependency and replaced the long shell line to call bison and modify the css grammar file with a few lines of portable perl code.

433

434

* JavaScriptCore.pri:

435

436

2007-07-27 Simon Hausmann <hausmann@kde.org>

437

438

Done with and reviewed by Lars and Zack.

439

440

Implemented currentTime() in the interpreter by using QDateTime, so that we don't need timeGetTime() on Windows and therefore also don't need to link against Winmm.dll.

441

442

* kjs/interpreter.cpp:

443

(KJS::getCurrentTime):

444

* kjs/testkjs.cpp:

445

(StopWatch::start):

446

(StopWatch::stop):

447

448

2007-07-27 Simon Hausmann <hausmann@kde.org>

449

450

Done with and reviewed by Lars and Zack.

451

452

Replace the use of snprintf with QByteArray to compile under msvc 2005 express.

453

454

* bindings/qt/qt_instance.cpp:

455

(KJS::Bindings::QtInstance::stringValue):

456

457

2007-07-27 Simon Hausmann <hausmann@kde.org>

458

459

Done with and reviewed by Lars and Zack.

460

461

Don't use pthread.h unless thread support is enabled.

462

463

* kjs/collector.cpp:

464

(KJS::Collector::registerAsMainThread):

465

(KJS::onMainThread):

466

467

2007-07-27 Simon Hausmann <hausmann@kde.org>

468

469

Done with and reviewed by Lars and Zack.

470

471

Removed TCSystemMalloc from the Qt build, it's not necessary it seems.

472

473

* JavaScriptCore.pri:

474

475

2007-07-27 Simon Hausmann <hausmann@kde.org>

476

477

Done with and reviewed by Lars and Zack.

478

479

Added os-win32 to the include search path for the Qt windows build in order to provide the fake stdint.h header file.

480

481

* JavaScriptCore.pri:

482

483

2007-07-25 Maciej Stachowiak <mjs@apple.com>

484

485

Reviewed by Mark.

486

487

- follow-up to previous change

488

489

* kjs/ustring.cpp:

490

(KJS::UString::operator=): Make sure to reset the length when

491

replacing the buffer contents for a single-owned string.

492

493

2007-07-25 Maciej Stachowiak <mjs@apple.com>

494

495

Reviewed by Darin.

496

497

- JavaScriptCore part of fix for <rdar://problem/5300291> Optimize GC to reclaim big, temporary objects (like XMLHttpRequest.responseXML) quickly

498

499

Also, as a side effect of optimizations included in this patch:

500

- 7% speedup on JavaScript iBench

501

- 4% speedup on "Celtic Kane" JS benchmark

502

503

The basic idea is explained in a big comment in collector.cpp. When unusually

504

large objecs are allocated, we push the next GC closer on the assumption that

505

most objects are short-lived.

506

507

I also did the following two optimizations in the course of tuning

508

this not to be a performance regression:

509

510

1) Change UString::Rep to hold a self-pointer as the baseString in

511

the unshared case, instead of a null pointer; this removes a

512

number of null checks in hot code because many places already

513

wanted to use the rep itself or the baseString as appropriate.

514

515

2) Avoid creating duplicate StringImpls when creating a

516

StringInstance (the object wrapper for a JS string) or calling

517

their methods. Since a temporary wrapper object is made every time

518

a string method is called, this resulted in two useless extra

519

StringImpls being allocated for no reason whenever a String method

520

was invoked on a string value. Now we bypass those.

521

522

* kjs/collector.cpp:

523

(KJS::):

524

(KJS::Collector::recordExtraCost): Basics of the extra cost mechanism.

525

(KJS::Collector::allocate): ditto

526

(KJS::Collector::collect): ditto

527

* kjs/collector.h:

528

(KJS::Collector::reportExtraMemoryCost): ditto

529

* kjs/array_object.cpp:

530

(ArrayInstance::ArrayInstance): record extra cost

531

* kjs/internal.cpp:

532

(KJS::StringImp::toObject): don't create a whole new StringImpl just

533

to be the internal value of a StringInstance! StringImpls are immutable

534

so there's no point tot his.

535

* kjs/internal.h:

536

(KJS::StringImp::StringImp): report extra cost

537

* kjs/string_object.cpp:

538

(KJS::StringInstance::StringInstance): new version that takes a StringImp

539

(KJS::StringProtoFunc::callAsFunction): don't create a whole new StringImpl

540

just to convert self to string! we already have one in the internal value

541

* kjs/string_object.h: report extra cost

542

* kjs/ustring.cpp: All changes to handle baseString being self instead of null in the

543

unshared case.

544

(KJS::):

545

(KJS::UString::Rep::create):

546

(KJS::UString::Rep::destroy):

547

(KJS::UString::usedCapacity):

548

(KJS::UString::usedPreCapacity):

549

(KJS::UString::expandCapacity):

550

(KJS::UString::expandPreCapacity):

551

(KJS::UString::UString):

552

(KJS::UString::append):

553

(KJS::UString::operator=):

554

(KJS::UString::copyForWriting):

555

* kjs/ustring.h:

556

(KJS::UString::Rep::baseIsSelf): new method, now that baseString is

557

self instead of null in the unshared case we can't just null check.

558

(KJS::UString::Rep::data): adjusted as mentioned above

559

(KJS::UString::cost): new method to compute the cost for a UString, for

560

use by StringImpl.

561

562

* kjs/value.cpp:

563

(KJS::jsString): style fixups.

564

(KJS::jsOwnedString): new method, use this for strings allocated from UStrings

565

held by the parse tree. Tracking their cost as part of string cost is pointless,

566

because garbage collecting them will not actually free the relevant string buffer.

567

* kjs/value.h: prototyped jsOwnedString.

568

* kjs/nodes.cpp:

569

(StringNode::evaluate): use jsOwnedString as appropriate

570

(RegExpNode::evaluate): ditto

571

(PropertyNameNode::evaluate): ditto

572

(ForInNode::execute): ditto

573

574

* JavaScriptCore.exp: Exported some new symbols.

575

576

2007-07-23 Anders Carlsson <andersca@apple.com>

577

578

Reviewed by Geoff.

579

580

<rdar://problem/5121461> REGRESSION: Unable to load JigZone puzzle

581

582

* bindings/jni/jni_jsobject.cpp:

583

(JavaJSObject::createNative):

584

585

Call RootObject::gcProtect on the global object, thereby putting it in the

586

"protect count" set which is used for checking if a native handle is valid.

587

588

2007-07-23 Darin Adler <darin@apple.com>

589

590

* pcre/pcre_compile.c: Roll back a tiny accidental change in the unused !JAVASCRIPT

591

side of an #ifdef. This has no effect when using PCRE in JAVASCRIPT mode as we do,

592

but seems worth rolling back.

593

594

2007-07-23 Maciej Stachowiak <mjs@apple.com>

595

596

Reviewed by Oliver.

597

598

- fix remaining problems with Window shadowing

599

600

* kjs/nodes.cpp:

601

(VarDeclNode::evaluate): Tweak the special case a little.

602

603

2007-07-23 Maciej Stachowiak <mjs@apple.com>

604

605

Reviewed by Oliver.

606

607

- fix Window shadowing regressions caused by the previous commit.

608

609

* kjs/nodes.cpp:

610

(VarDeclNode::evaluate): Handle the case of global scope specially.

611

612

2007-07-22 Maciej Stachowiak <mjs@apple.com>

613

614

Reviewed by Darin.

615

616

-fixed <rdar://problem/5353293> REGRESSION (r24287): 1% i-Bench JS slowdown from JavaScript compatibility fix (14719)

617

http://bugs.webkit.org/show_bug.cgi?id=14719

618

619

My fix for this actually resulted in JS iBench being 1% faster than before the regression

620

and the Celtic Kane benchmark being 5% faster than before the regression.

621

622

* kjs/nodes.cpp:

623

(VarDeclNode::handleSlowCase): factored out the slow code path to be out of line.

624

(VarDeclNode::evaluate): I did a couple of things:

625

(1) Don't check if the variable is already declared by looking for the property in

626

the variable object, that code path was dead code.

627

(2) Special-case the common case where the top of the scope and the variable object

628

are the same; in that case the variable must always be in the variable object.

629

(3) Don't return a jsString() of the variable name, nothing uses the return value

630

from this node types evaluate method.

631

* kjs/nodes.h:

632

633

2007-07-22 Darin Adler <darin@apple.com>

634

635

Reviewed by Kevin Decker.

636

637

- fix <rdar://problem/5126394> REGRESSION: Crash after clicking back button in test application (13250)

638

http://bugs.webkit.org/show_bug.cgi?id=13250

639

640

* bindings/objc/objc_utility.mm: (KJS::Bindings::convertObjcValueToValue):

641

If the object returns 0 for _imp, convert that to "undefined", since callers

642

can't cope with a JSValue of 0.

643

644

2007-07-19 Geoffrey Garen <ggaren@apple.com>

645

646

Reviewed by Darin Adler.

647

648

Fixed http://bugs.webkit.org/show_bug.cgi?id=10880 | <rdar://problem/5335694>

649

REGRESSION: JavaScript menu doesn't appear on pricepoint.com (14595)

650

651

Though the ECMA spec says auto-semicolon insertion should not occur

652

without a newline or '}', Firefox treats do-while specially, and the

653

library used by pricepoint.com requires that special treatment.

654

655

* JavaScriptCore.xcodeproj/project.pbxproj:

656

* kjs/grammar.y:

657

658

2007-07-19 Darin Adler <darin@apple.com>

659

660

Reviewed by Geoff.

661

662

- fix <rdar://problem/5345440> PCRE computes wrong length for expressions with quantifiers

663

on named recursion or subexpressions

664

665

It's challenging to implement proper preflighting for compiling these advanced features.

666

But we don't want them in the JavaScript engine anyway.

667

668

Turned off the following features of PCRE (some of these are simply parsed and not implemented):

669

670

\C \E \G \L \N \P \Q \U \X \Z

671

\e \l \p \u \z

672

[::] [..] [==]

673

(?#) (?<=) (?<!) (?>)

674

(?C) (?P) (?R)

675

(?0) (and 1-9)

676

(?imsxUX)

677

678

Added the following:

679

680

\u \v

681

682

Because of \v, the js1_2/regexp/special_characters.js test now passes.

683

684

To be conservative, I left some features that JavaScript doesn't want, such as

685

\012 and \x{2013}, in place. We can revisit these later; they're not directly-enough

686

related to avoiding the incorrect preflighting.

687

688

I also didn't try to remove unused opcodes and remove code from the execution engine.

689

That could save code size and speed things up a bit, but it would require more changes.

690

691

* kjs/regexp.h:

692

* kjs/regexp.cpp: (KJS::RegExp::RegExp): Remove the sanitizePattern workaround for

693

lack of \u support, since the PCRE code now has \u support.

694

695

* pcre/pcre-config.h: Set JAVASCRIPT to 1.

696

* pcre/pcre_internal.h: Added ESC_v.

697

698

* pcre/pcre_compile.c: Added a different escape table for when JAVASCRIPT is set that

699

omits all the escapes we don't want interpreted and includes '\v'.

700

(check_escape): Put !JAVASCRIPT around the code for '\l', '\L', '\N', '\u', and '\U',

701

and added code to handle '\u2013' inside JAVASCRIPT.

702

(compile_branch): Put !JAVASCRIPT if around all the code implementing the features we

703

don't want.

704

(pcre_compile2): Ditto.

705

706

* tests/mozilla/expected.html: Updated since js1_2/regexp/special_characters.js now

707

passes.

708

709

2007-07-18 Darin Adler <darin@apple.com>

710

711

Reviewed by Oliver Hunt.

712

713

- fix <rdar://problem/5345432> PCRE computes length wrong for expressions such as "[**]"

714

715

Test: fast/js/regexp-charclass-crash.html

716

717

* pcre/pcre_compile.c: (pcre_compile2): Fix the preflight code that calls

718

check_posix_syntax to match the actual regular expression compilation code;

719

before it was missing the check of the first character.

720

721

2007-07-19 Holger Hans Peter Freyther <zecke@selfish.org>

722

723

Reviewed by Mark.

724

725

Define __BUILDING_GDK when building for Gdk to fix building testkjs on OSX.

726

727

* JavaScriptCore.pri:

728

729

2007-07-18 Simon Hausmann <hausmann@kde.org>

730

731

* Fix the Qt build, call dftables from the right directory.

732

733

Reviewed by Adam Treat.

734

735

* pcre/pcre.pri:

736

737

2007-07-18 Simon Hausmann <hausmann@kde.org>

738

739

Reviewed by Zack.

740

741

Don't call gcc directly when building the dftables tool but use a separate .pro file for the Qt build.

742

743

* pcre/dftables.pro: Added.

744

* pcre/pcre.pri:

745

746

2007-07-17 Cameron Zwarich <cwzwarich@uwaterloo.ca>

747

748

Reviewed by Darin, Maciej, and Adam.

749

750

Fixes <http://bugs.webkit.org/show_bug.cgi?id=9697>,

751

the failure of ecma/GlobalObject/15.1.2.2-2.js,

752

the failure of ecma/LexicalConventions/7.7.3-1.js,

753

and most of the failures of tests in ecma/TypeConversion/9.3.1-3.js.

754

755

Bug 9697: parseInt results may be inaccurate for numbers greater than 2^53

756

757

This patch also fixes similar issues in the lexer and UString::toDouble().

758

759

* kjs/function.cpp:

760

(KJS::parseIntOverflow):

761

(KJS::parseInt):

762

* kjs/function.h:

763

* kjs/lexer.cpp:

764

(KJS::Lexer::lex):

765

* kjs/ustring.cpp:

766

(KJS::UString::toDouble):

767

* tests/mozilla/expected.html:

768

769

2007-07-16 Sam Weinig <sam@webkit.org>

770

771

Reviewed by Oliver.

772

773

Turn off -Wshorten-64-to-32 warning for 64-bit builds.

774

775

* Configurations/Base.xcconfig:

776

777

2007-07-14 Brady Eidson <beidson@apple.com>

778

779

Reviewed by Sam Weinig

780

781

Initial check-in for <rdar://problem/3154486> - Supporting FTP directory listings in the browser

782

783

* wtf/Platform.h: Add ENABLE_FTPDIR feature to handle building on platforms that don't have the

784

proper network-layer support

785

786

2007-07-14 Cameron Zwarich <cwzwarich@uwaterloo.ca>

787

788

Reviewed by Darin.

789

790

Fixes http://bugs.webkit.org/show_bug.cgi?id=13517,

791

http://bugs.webkit.org/show_bug.cgi?id=14237, and

792

the failure of test js1_5/Scope/regress-185485.js

793

794

Bug 13517: DOM Exception 8 in finance.aol.com sub-page

795

Bug 14237: Javascript "var" statement interprets initialization in the topmost function scope

796

797

* kjs/nodes.cpp:

798

(VarDeclNode::evaluate):

799

* tests/mozilla/expected.html:

800

801

2007-07-12 Alexey Proskuryakov <ap@webkit.org>

802

803

Reviewed by Mitz.

804

805

http://bugs.webkit.org/show_bug.cgi?id=14596

806

Fix JSC compilation with KJS_VERBOSE.

807

808

* kjs/function.cpp:

809

(KJS::FunctionImp::passInParameters):

810

811

2007-07-11 George Staikos <staikos@kde.org>

812

813

Make it compile.

814

815

* ForwardingHeaders: Added.

816

* ForwardingHeaders/JavaScriptCore: Added.

817

* ForwardingHeaders/JavaScriptCore/APICast.h: Added.

818

* ForwardingHeaders/JavaScriptCore/JSBase.h: Added.

819

* ForwardingHeaders/JavaScriptCore/JSContextRef.h: Added.

820

* ForwardingHeaders/JavaScriptCore/JSLock.h: Added.

821

* ForwardingHeaders/JavaScriptCore/JSObjectRef.h: Added.

822

* ForwardingHeaders/JavaScriptCore/JSStringRef.h: Added.

823

* ForwardingHeaders/JavaScriptCore/JSStringRefCF.h: Added.

824

* ForwardingHeaders/JavaScriptCore/JSValueRef.h: Added.

825

* ForwardingHeaders/JavaScriptCore/JavaScriptCore.h: Added.

826

827

2007-07-11 Holger Hans Peter Freyther <zecke@selfish.org>

828

829

Reviewed by Darin.

830

831

As of http://bugs.webkit.org/show_bug.cgi?id=14527 move the

832

WebCore/ForwardingHeader/JavaScriptCore to JavaScriptCore

833

834

* ForwardingHeaders: Added.

835

* ForwardingHeaders/JavaScriptCore: Copied from WebCore/ForwardingHeaders/JavaScriptCore.

836

837

2007-07-11 Nikolas Zimmermann <zimmermann@kde.org>

838

839

Reviewed by Mark.

840

841

Forwardport the hash table fix from CodeGeneratorJS.pm to create_hash_table.

842

Reran run-jsc-tests, couldn't find any regressions. Suggested by Darin.

843

844

* kjs/create_hash_table:

845

846

2007-07-09 Maciej Stachowiak <mjs@apple.com>

847

848

Reviewed by Oliver.

849

850

- JavaScriptCore part of fix for: <rdar://problem/5295734> Repro crash closing tab/window @ maps.google.com in WTF::HashSet<KJS::RuntimeObjectImp*, WTF::PtrHash<KJS::RuntimeObjectImp*>, WTF::HashTraits<KJS::RuntimeObjectImp*> >::add + 11

851

852

* JavaScriptCore.exp: Added needed export.

853

854

2007-07-06 Maciej Stachowiak <mjs@apple.com>

855

856

Reviewed by Antti.

857

858

- <rdar://problem/5311093> JavaScriptCore fails to build with strict-aliasing warnings

859

860

* Configurations/Base.xcconfig: Re-enable -Wstrict-aliasing

861

* bindings/jni/jni_utility.cpp:

862

(KJS::Bindings::getJNIEnv): Type-pun via a union instead of a pointer cast.

863

* wtf/HashMap.h:

864

(WTF::): Instead of doing type-punned assignments via pointer cast, do one of three things:

865

(1) assign directly w/o cast if storage type matches real type; (2) assign using cast

866

via union if type does not need reffing; (3) copy with memcpy and ref/deref manually if type

867

needs reffing. This is ok peref-wise because memcpy of a constant length gets optomized.

868

HashTraits are now expected to make ref()/deref() take the storage type, not the true type.

869

* wtf/HashSet.h:

870

(WTF::): Same basic idea.

871

* wtf/HashTable.h:

872

(WTF::): Added Assigner template for use by HashMap/HashSet. Change RefCounter to call ref()

873

and deref() via storage type, avoiding the need to

874

type-pun.

875

(WTF::RefCounter::ref): ditto

876

(WTF::RefCounter::deref): ditto

877

* wtf/HashTraits.h:

878

(WTF::): Change ref() and deref() for RefPtr HashTraits to take the storage type; cast

879

via union to pointer type.

880

* wtf/FastMalloc.cpp:

881

(WTF::TCMalloc_PageHeap::init): Changed from constructor to init function so this can go in a union.

882

(WTF::): redefine pageheap macro in terms of getPageHeap().

883

(WTF::getPageHeap): new inline function, helper for pageheap macro. This hides the cast in a union.

884

(WTF::TCMalloc_ThreadCache::InitModule): Call init() instead of using placement new to initialize page

885

heap.

886

* wtf/TCPageMap.h:

887

(TCMalloc_PageMap1::init): Changed from constructor to init function.

888

(TCMalloc_PageMap2::init): ditto

889

(TCMalloc_PageMap3::init): ditto

890

891

892

2007-07-06 George Staikos <staikos@kde.org>

893

894

Reviewed by Maciej.

895

896

Switch USE(ICONDATABASE) to ENABLE(ICONDATABASE)

897

898

* wtf/Platform.h:

899

900

2007-07-03 Sam Weinig <sam@webkit.org>

901

902

Reviewed by Darin.

903

904

Eleventh round of fixes for implicit 64-32 bit conversion errors.

905

<rdar://problem/5292262>

906

907

- Fixes a real bug where where we were setting long long and unsigned long long

908

values to a long field.

909

910

* bindings/objc/objc_utility.mm:

911

(KJS::Bindings::convertValueToObjcValue):

912

913

2007-07-03 Sam Weinig <sam@webkit.org>

914

915

Reviewed by Brady Eidson.

916

917

Tenth round of fixes for implicit 64-32 bit conversion errors.

918

<rdar://problem/5292262>

919

920

- Add explicit casts.

921

922

* kjs/dtoa.cpp:

923

(Bigint::):

924

925

2007-07-02 Sam Weinig <sam@webkit.org>

926

927

Reviewed by Kevin McCullough.

928

929

Fourth round of fixes for implicit 64-32 bit conversion errors.

930

<rdar://problem/5292262>

931

932

Add custom piDouble and piFloat constants to use instead of M_PI.

933

934

* kjs/math_object.cpp:

935

(MathObjectImp::getValueProperty):

936

* wtf/MathExtras.h:

937

(wtf_atan2):

938

939

2007-06-29 Sam Weinig <sam@webkit.org>

940

941

Reviewed by Darin.

942

943

Second pass at fixing implicit 64-32 bit conversion errors.

944

<rdar://problem/5292262>

945

946

- Add a toFloat() method to JSValue for float conversion.

947

948

* JavaScriptCore.exp:

949

* kjs/value.cpp:

950

(KJS::JSValue::toFloat):

951

* kjs/value.h:

952

953

2007-06-27 Kevin McCullough <kmccullough@apple.com>

954

955

Reviewed by Darin.

956

957

- <rdar://problem/5271937> REGRESSION: Apparent WebKit JavaScript memory smasher when submitting comment to iWeb site (crashes in kjs_pcre_compile2)

958

- Correctly evaluate the return value of _pcre_ucp_findchar.

959

960

* pcre/pcre_compile.c:

961

(compile_branch):

962

* pcre/pcre_exec.c:

963

(match):

964

965

2007-06-27 Sam Weinig <sam@webkit.org>

966

967

Reviewed by Darin.

968

969

First pass at fixing implicit 64-32 bit conversion errors.

970

<rdar://problem/5292262>

971

972

- Add 'f' suffix where necessary.

973

974

* kjs/testkjs.cpp:

975

(StopWatch::getElapsedMS):

976

977

2007-06-26 Geoffrey Garen <ggaren@apple.com>

978

979

Reviewed by Maciej Stachowiak.

980

981

Fixed <rdar://problem/5296627> JSGarbageCollect headerdoc suggests that

982

using JavaScriptCore requires leaking memory

983

984

* API/JSBase.h: Changed documentation to explain that you can pass NULL

985

to JSGarbageCollect.

986

987

2007-06-26 Adam Treat <adam@staikos.net>

988

989

Reviewed by Adam Roben.

990

991

Make the SQLite icon database optional.

992

993

* wtf/Platform.h:

994

995

2007-06-15 George Staikos <staikos@kde.org>

996

997

More missing files for Qt.

998

999

* JavaScriptCore.pri:

1000

* kjs/testkjs.pro:

1001

1002

2007-06-15 George Staikos <staikos@kde.org>

1003

1004

Another Qt build fix.

1005

1006

* JavaScriptCore.pri:

1007

* kjs/testkjs.pro:

1008

1009

2007-06-15 George Staikos <staikos@kde.org>

1010

1011

Fixing Qt build.

1012

1013

* JavaScriptCore.pri:

1014

1015

2007-06-20 Mark Rowe <mrowe@apple.com>

1016

1017

Reviewed by Mitz.

1018

1019

Fix http://bugs.webkit.org/show_bug.cgi?id=14244

1020

Bug 14244: Data corruption when using a replace() callback function with data containing "$"

1021

1022

* kjs/string_object.cpp:

1023

(KJS::replace): When 'replacement' is a function, do not replace $n placeholders in its return value.

1024

This matches the behaviour described in ECMA 262 3rd Ed section 15.5.4.1, and as implemented in Firefox.

1025

1026

2007-06-14 Anders Carlsson <andersca@apple.com>

1027

1028

Fix Windows build.

1029

1030

* bindings/runtime_object.cpp:

1031

(RuntimeObjectImp::canPut):

1032

1033

2007-06-14 Anders Carlsson <andersca@apple.com>

1034

1035

Reviewed by Darin.

1036

1037

<rdar://problem/5103077>

1038

Crash at _NPN_ReleaseObject when quitting page at http://eshop.macsales.com/shop/ModBook

1039

1040

<rdar://problem/5183692>

1041

http://bugs.webkit.org/show_bug.cgi?id=13547

1042

REGRESSION: Crash in _NPN_ReleaseObject when closing Safari on nba.com (13547)

1043

1044

<rdar://problem/5261499>

1045

CrashTracer: [USER] 75 crashes in Safari at com.apple.JavaScriptCore: KJS::Bindings::CInstance::~CInstance + 40

1046

1047

Have the root object track all live instances of RuntimeObjectImp. When invalidating

1048

the root object, also invalidate all live runtime objects by zeroing out their instance ivar.

1049

This prevents instances from outliving their plug-ins which lead to crashes.

1050

1051

* bindings/c/c_utility.cpp:

1052

(KJS::Bindings::convertValueToNPVariant):

1053

* bindings/jni/jni_jsobject.cpp:

1054

(JavaJSObject::convertValueToJObject):

1055

* bindings/jni/jni_utility.cpp:

1056

(KJS::Bindings::convertValueToJValue):

1057

* bindings/objc/objc_runtime.mm:

1058

(ObjcFallbackObjectImp::callAsFunction):

1059

* bindings/runtime_array.cpp:

1060

(RuntimeArray::RuntimeArray):

1061

* bindings/runtime_array.h:

1062

(KJS::RuntimeArray::getConcreteArray):

1063

* bindings/runtime_method.cpp:

1064

(RuntimeMethod::callAsFunction):

1065

* bindings/runtime_method.h:

1066

* bindings/runtime_object.cpp:

1067

(RuntimeObjectImp::RuntimeObjectImp):

1068

(RuntimeObjectImp::~RuntimeObjectImp):

1069

(RuntimeObjectImp::invalidate):

1070

(RuntimeObjectImp::fallbackObjectGetter):

1071

(RuntimeObjectImp::fieldGetter):

1072

(RuntimeObjectImp::methodGetter):

1073

(RuntimeObjectImp::getOwnPropertySlot):

1074

(RuntimeObjectImp::put):

1075

(RuntimeObjectImp::canPut):

1076

(RuntimeObjectImp::defaultValue):

1077

(RuntimeObjectImp::implementsCall):

1078

(RuntimeObjectImp::callAsFunction):

1079

(RuntimeObjectImp::getPropertyNames):

1080

(RuntimeObjectImp::throwInvalidAccessError):

1081

* bindings/runtime_object.h:

1082

* bindings/runtime_root.cpp:

1083

(KJS::Bindings::RootObject::invalidate):

1084

(KJS::Bindings::RootObject::addRuntimeObject):

1085

(KJS::Bindings::RootObject::removeRuntimeObject):

1086

* bindings/runtime_root.h:

1087

1088

2007-06-14 Anders Carlsson <andersca@apple.com>

1089

1090

Reviewed by Mitz.

1091

1092

<rdar://problem/5244948>

1093

Safari keeps on complaining about slow script playing NBC TV video (14133)

1094

1095

http://bugs.webkit.org/show_bug.cgi?id=14133

1096

Runaway JavaScript timer fires when spinning around in Google Maps street view

1097

1098

Make sure to start and stop the timeout checker around calls to JS.

1099

1100

* bindings/NP_jsobject.cpp:

1101

(_NPN_InvokeDefault):

1102

(_NPN_Invoke):

1103

(_NPN_Evaluate):

1104

* bindings/jni/jni_jsobject.cpp:

1105

(JavaJSObject::call):

1106

(JavaJSObject::eval):

1107

1108

2007-06-13 Darin Adler <darin@apple.com>

1109

1110

Reviewed by Mark Rowe.

1111

1112

- fix http://bugs.webkit.org/show_bug.cgi?id=14132

1113

array sort with > 10000 elements sets elements > 10000 undefined

1114

1115

Test: fast/js/sort-large-array.html

1116

1117

* kjs/array_instance.h: Replaced pushUndefinedObjectsToEnd with

1118

compactForSorting, and removed ExecState parameters.

1119

1120

* kjs/array_object.cpp:

1121

(ArrayInstance::sort): Changed to call compactForSorting.

1122

(ArrayInstance::compactForSorting): Do the get and delete of the

1123

properties directly on the property map instead of using public

1124

calls from JSObject. The public calls would just read the undefined

1125

values from the compacted sort results array!

1126

1127

2007-06-13 George Staikos <staikos@kde.org>

1128

1129

Reviewed by Lars.

1130

1131

Fix Mac OS X build after last checkin.

1132

1133

* wtf/FastMalloc.h:

1134

1135

2007-06-14 Lars Knoll <lars@trolltech.com>

1136

1137

Reviewed by Maciej.

1138

1139

Disable FastMalloc for the Qt build and make sure we

1140

don't reimplement the global new/delete operators

1141

when using the system malloc.

1142

1143

* wtf/FastMalloc.cpp:

1144

* wtf/FastMalloc.h:

1145

* wtf/Platform.h:

1146

1147

2007-06-13 Anders Carlsson <andersca@apple.com>

1148

1149

Reviewed by Geoff.

1150

1151

Make sure that bindings instances get correct root objects.

1152

1153

* JavaScriptCore.exp:

1154

* bindings/NP_jsobject.cpp:

1155

(listFromVariantArgs):

1156

(_NPN_InvokeDefault):

1157

(_NPN_Invoke):

1158

(_NPN_SetProperty):

1159

* bindings/c/c_instance.cpp:

1160

(KJS::Bindings::CInstance::invokeMethod):

1161

(KJS::Bindings::CInstance::invokeDefaultMethod):

1162

* bindings/c/c_runtime.cpp:

1163

(KJS::Bindings::CField::valueFromInstance):

1164

* bindings/c/c_utility.cpp:

1165

(KJS::Bindings::convertNPVariantToValue):

1166

* bindings/c/c_utility.h:

1167

* bindings/objc/objc_instance.mm:

1168

(ObjcInstance::invokeMethod):

1169

(ObjcInstance::invokeDefaultMethod):

1170

(ObjcInstance::getValueOfUndefinedField):

1171

* bindings/objc/objc_runtime.mm:

1172

(ObjcField::valueFromInstance):

1173

(ObjcArray::valueAt):

1174

* bindings/objc/objc_utility.h:

1175

* bindings/objc/objc_utility.mm:

1176

(KJS::Bindings::convertObjcValueToValue):

1177

* bindings/runtime.h:

1178

1179

2007-06-13 Simon Hausmann <hausmann@kde.org>

1180

1181

Reviewed by Lars.

1182

1183

* kjs/testkjs.pro: WebKitQt is now called QtWebKit.

1184

1185

2007-06-12 Anders Carlsson <andersca@apple.com>

1186

1187

Another build fix.

1188

1189

* bindings/qt/qt_instance.cpp:

1190

(KJS::Bindings::QtInstance::invokeMethod):

1191

1192

2007-06-12 Anders Carlsson <andersca@apple.com>

1193

1194

Reviewed by Geoff.

1195

1196

Move the notion of field type to the JNI runtime since that's the only

1197

one that was actually using it.

1198

1199

* bindings/c/c_runtime.h:

1200

(KJS::Bindings::CField::CField):

1201

* bindings/jni/jni_runtime.h:

1202

* bindings/objc/objc_runtime.h:

1203

* bindings/objc/objc_runtime.mm:

1204

* bindings/qt/qt_runtime.h:

1205

* bindings/runtime.h:

1206

* bindings/runtime_method.cpp:

1207

1208

2007-06-12 Anders Carlsson <andersca@apple.com>

1209

1210

Build fix.

1211

1212

* bindings/qt/qt_class.cpp:

1213

(KJS::Bindings::QtClass::methodsNamed):

1214

* bindings/qt/qt_instance.cpp:

1215

(KJS::Bindings::QtInstance::invokeMethod):

1216

1217

2007-06-12 Anders Carlsson <andersca@apple.com>

1218

1219

Reviewed by Oliver.

1220

1221

Get rid of the MethodList class and use a good ol' Vector instead.

1222

1223

* bindings/c/c_class.cpp:

1224

(KJS::Bindings::CClass::methodsNamed):

1225

* bindings/c/c_instance.cpp:

1226

(KJS::Bindings::CInstance::invokeMethod):

1227

* bindings/jni/jni_class.cpp:

1228

(JavaClass::JavaClass):

1229

(JavaClass::~JavaClass):

1230

* bindings/jni/jni_instance.cpp:

1231

(JavaInstance::invokeMethod):

1232

* bindings/objc/objc_class.mm:

1233

(KJS::Bindings::ObjcClass::methodsNamed):

1234

* bindings/objc/objc_instance.mm:

1235

(ObjcInstance::invokeMethod):

1236

* bindings/objc/objc_runtime.mm:

1237

(ObjcFallbackObjectImp::callAsFunction):

1238

* bindings/runtime.cpp:

1239

* bindings/runtime.h:

1240

* bindings/runtime_method.cpp:

1241

(RuntimeMethod::lengthGetter):

1242

(RuntimeMethod::callAsFunction):

1243

* bindings/runtime_object.cpp:

1244

(RuntimeObjectImp::getOwnPropertySlot):

1245

1246

2007-06-12 Anders Carlsson <andersca@apple.com>

1247

1248

Reviewed by Geoff.

1249

1250

Make RuntimeMethod's method list a pointer so that the object size doesn't

1251

grow beyond 32 bytes when we later will replace MethodList with a Vector.

1252

1253

* bindings/runtime_method.cpp:

1254

(RuntimeMethod::RuntimeMethod):

1255

(RuntimeMethod::lengthGetter):

1256

(RuntimeMethod::callAsFunction):

1257

* bindings/runtime_method.h:

1258

1259

2007-06-12 Anders Carlsson <andersca@apple.com>

1260

1261

Reviewed by Geoff.

1262

1263

Get rid of the Parameter class.

1264

1265

* bindings/jni/jni_instance.cpp:

1266

(JavaInstance::invokeMethod):

1267

* bindings/jni/jni_runtime.cpp:

1268

(JavaMethod::signature):

1269

* bindings/jni/jni_runtime.h:

1270

(KJS::Bindings::JavaParameter::JavaParameter):

1271

(KJS::Bindings::JavaParameter::~JavaParameter):

1272

(KJS::Bindings::JavaParameter::type):

1273

(KJS::Bindings::JavaMethod::parameterAt):

1274

(KJS::Bindings::JavaMethod::numParameters):

1275

* bindings/runtime.h:

1276

1277

2007-06-12 Anders Carlsson <andersca@apple.com>

1278

1279

Build fix.

1280

1281

* bindings/qt/qt_class.h:

1282

1283

2007-06-12 Mark Rowe <mrowe@apple.com>

1284

1285

Build fix.

1286

1287

* bindings/objc/objc_runtime.h:

1288

1289

2007-06-12 Anders Carlsson <andersca@apple.com>

1290

1291

Reviewed by Geoff.

1292

1293

Get rid of Constructor and its only subclass JavaConstructor.

1294

1295

* bindings/c/c_class.h:

1296

* bindings/jni/jni_class.cpp:

1297

(JavaClass::JavaClass):

1298

(JavaClass::~JavaClass):

1299

* bindings/jni/jni_class.h:

1300

* bindings/jni/jni_runtime.cpp:

1301

* bindings/jni/jni_runtime.h:

1302

* bindings/objc/objc_class.h:

1303

* bindings/runtime.h:

1304

1305

2007-06-12 Anders Carlsson <andersca@apple.com>

1306

1307

Reviewed by Geoff.

1308

1309

Use RetainPtr throughout the bindings code.

1310

1311

* bindings/objc/objc_class.h:

1312

* bindings/objc/objc_class.mm:

1313

(KJS::Bindings::ObjcClass::ObjcClass):

1314

(KJS::Bindings::ObjcClass::methodsNamed):

1315

(KJS::Bindings::ObjcClass::fieldNamed):

1316

* bindings/objc/objc_instance.h:

1317

(KJS::Bindings::ObjcInstance::getObject):

1318

* bindings/objc/objc_instance.mm:

1319

(ObjcInstance::ObjcInstance):

1320

(ObjcInstance::~ObjcInstance):

1321

(ObjcInstance::implementsCall):

1322

(ObjcInstance::invokeMethod):

1323

(ObjcInstance::invokeDefaultMethod):

1324

(ObjcInstance::defaultValue):

1325

* bindings/objc/objc_runtime.h:

1326

(KJS::Bindings::ObjcMethod::setJavaScriptName):

1327

(KJS::Bindings::ObjcMethod::javaScriptName):

1328

(KJS::Bindings::ObjcArray::getObjcArray):

1329

* bindings/objc/objc_runtime.mm:

1330

(ObjcField::name):

1331

(ObjcArray::ObjcArray):

1332

(ObjcArray::setValueAt):

1333

(ObjcArray::valueAt):

1334

(ObjcArray::getLength):

1335

* wtf/RetainPtr.h:

1336

1337

2007-06-12 Anders Carlsson <andersca@apple.com>

1338

1339

Reviewed by Maciej.

1340

1341

Have JSCell inherit from Noncopyable.

1342

1343

* bindings/objc/objc_runtime.h:

1344

* bindings/runtime_object.h:

1345

* kjs/value.h:

1346

1347

2007-06-12 Anders Carlsson <andersca@apple.com>

1348

1349

Reviewed by Darin and Maciej.

1350

1351

More cleanup. Use our Noncopyable WTF class, add a root object member

1352

to the Array class.

1353

1354

* bindings/c/c_class.h:

1355

* bindings/jni/jni_class.h:

1356

* bindings/jni/jni_instance.h:

1357

* bindings/jni/jni_runtime.cpp:

1358

(JavaArray::JavaArray):

1359

* bindings/jni/jni_runtime.h:

1360

* bindings/objc/objc_class.h:

1361

* bindings/objc/objc_runtime.h:

1362

* bindings/objc/objc_runtime.mm:

1363

(ObjcArray::ObjcArray):

1364

* bindings/objc/objc_utility.mm:

1365

(KJS::Bindings::convertObjcValueToValue):

1366

* bindings/runtime.cpp:

1367

(KJS::Bindings::Array::Array):

1368

(KJS::Bindings::Array::~Array):

1369

* bindings/runtime.h:

1370

* bindings/runtime_object.h:

1371

* bindings/runtime_root.h:

1372

1373

2007-06-08 Zack Rusin <zrusin@trolltech.com>

1374

1375

Fix the Qt build

1376

1377

* bindings/qt/qt_instance.cpp:

1378

(KJS::Bindings::QtInstance::QtInstance):

1379

* bindings/qt/qt_instance.h:

1380

1381

2007-06-07 Anders Carlsson <andersca@apple.com>

1382

1383

Reviewed by Geoff.

1384

1385

Get rid of Instance::setRootObject and pass the root object to the instance constructor instead.

1386

1387

* bindings/c/c_instance.cpp:

1388

(KJS::Bindings::CInstance::CInstance):

1389

* bindings/c/c_instance.h:

1390

* bindings/jni/jni_instance.cpp:

1391

(JavaInstance::JavaInstance):

1392

* bindings/jni/jni_instance.h:

1393

* bindings/jni/jni_jsobject.cpp:

1394

(JavaJSObject::convertJObjectToValue):

1395

* bindings/objc/objc_instance.h:

1396

* bindings/objc/objc_instance.mm:

1397

(ObjcInstance::ObjcInstance):

1398

* bindings/runtime.cpp:

1399

(KJS::Bindings::Instance::Instance):

1400

(KJS::Bindings::Instance::createBindingForLanguageInstance):

1401

* bindings/runtime.h:

1402

1403

2007-06-07 Anders Carlsson <andersca@apple.com>

1404

1405

Reviewed by Adam.

1406

1407

Don't use a JavaInstance to store the field when all we want to do is to keep the field

1408

from being garbage collected. Instead, use a JObjectWrapper.

1409

1410

* bindings/jni/jni_instance.h:

1411

* bindings/jni/jni_runtime.cpp:

1412

(JavaField::JavaField):

1413

(JavaField::dispatchValueFromInstance):

1414

(JavaField::dispatchSetValueToInstance):

1415

* bindings/jni/jni_runtime.h:

1416

(KJS::Bindings::JavaField::JavaField):

1417

(KJS::Bindings::JavaField::operator=):

1418

1419

2007-05-30 Alp Toker <alp.toker@collabora.co.uk>

1420

1421

Reviewed by Brady.

1422

1423

Enable logging in the Gdk port.

1424

http://bugs.webkit.org/show_bug.cgi?id=13936

1425

1426

* wtf/Assertions.cpp:

1427

* wtf/Assertions.h: Add WTFLogVerbose which also logs

1428

the file, line number and function.

1429

1430

2007-05-30 Mark Rowe <mrowe@apple.com>

1431

1432

Mac build fix. Update #include.

1433

1434

* API/JSCallbackFunction.h:

1435

1436

2007-05-30 Luciano Montanaro <mikelima@cirulla.net>

1437

1438

Reviewed by Maciej.

1439

1440

- cross-port Harri Porten's commits 636099 and 636108 from KJS:

1441

"publish a class anyway public already" and "class is being used from

1442

outside for quite some time" in preparation for further syncronizations

1443

1444

* kjs/context.h:

1445

* kjs/date_object.cpp:

1446

* kjs/date_object.h:

1447

* kjs/function.h:

1448

(KJS::):

1449

(KJS::InternalFunctionImp::classInfo):

1450

(KJS::InternalFunctionImp::functionName):

1451

* kjs/function_object.h:

1452

* kjs/internal.h:

1453

* kjs/lookup.h:

1454

(KJS::getStaticPropertySlot):

1455

(KJS::getStaticFunctionSlot):

1456

(KJS::getStaticValueSlot):

1457

* kjs/object_object.h:

1458

1459

2007-05-29 Sam Weinig <sam@webkit.org>

1460

1461

Reviewed by Adam Roben.

1462

1463

Cleanup function and fix to match comparison API.

1464

1465

* kjs/string_object.cpp:

1466

(KJS::substituteBackreferences):

1467

(KJS::localeCompare):

1468

1469

2007-05-28 Geoffrey Garen <ggaren@apple.com>

1470

1471

Slight clarification to an exception message.

1472

1473

* API/JSCallbackObject.cpp:

1474

(KJS::JSCallbackObject::put):

1475

1476

2007-05-27 Holger Freyther <zecke@selfish.org>

1477

1478

Reviewed by Mark Rowe.

1479

1480

* wtf/Platform.h: Move Gdk up to allow building WebKit/Gdk on Darwin

1481

1482

2007-05-27 Darin Adler <darin@apple.com>

1483

1484

- fix a couple ifdefs that said WIN instead of WIN_OS

1485

1486

* kjs/collector.cpp:

1487

(KJS::allocateBlock): WIN -> WIN_OS

1488

(KJS::freeBlock): Ditto.

1489

1490

2007-05-26 Sam Weinig <sam@webkit.org>

1491

1492

Reviewed by Darin.

1493

1494

Patch for http://bugs.webkit.org/show_bug.cgi?id=13854

1495

Port of commit 667785 from kjs

1496

1497

- special case calling String.localeCompare() with no parameters to return 0.

1498

1499

* kjs/string_object.cpp:

1500

(KJS::StringProtoFunc::callAsFunction):

1501

1502

2007-05-25 Kimmo Kinnunen <kimmok@iki.fi>

1503

1504

Reviewed by Darin.

1505

1506

- Fix for http://bugs.webkit.org/show_bug.cgi?id=13456

1507

REGRESSION: setTimeout "arguments" object gets shadowed by a local variable

1508

1509

- Add a explicit check for arguments. Previously check was done with getDirect,

1510

but since the arguments is created on-demand in ActivationImp, it doesn't

1511

show up in the test. 'arguments' should always be in the VarDeclNode's

1512

evaluation scope.

1513

1514

* kjs/nodes.cpp:

1515

(VarDeclNode::evaluate): Additional check if the var decl identifier is 'arguments'

1516

1517

2007-05-25 George Staikos <staikos@kde.org>

1518

1519

Reviewed by Maciej.

1520

1521

- Use COMPILER(GCC), not PLATFORM(GCC) - as Platform.h defines

1522

1523

* wtf/FastMalloc.h:

1524

1525

2007-05-25 Kimmo Kinnunen <kimmok@iki.fi>

1526

1527

Reviewed by Darin.

1528

1529

- http://bugs.webkit.org/show_bug.cgi?id=13623 (Decompilation of function

1530

doesn't compile with "++(x,y)")

1531

- Create the error node based on the actual node, not the node inside

1532

parenthesis

1533

- Fix applies to postfix, prefix and typeof operators

1534

- Produces run-time ReferenceError like other non-lvalue assignments etc.

1535

1536

* kjs/grammar.y: Create {Prefix,Postfix}ErrorNode based on the actual node,

1537

not the based on the node returned by "nodeInsideAllParens()". Same for

1538

TypeOfValueNode.

1539

1540

2007-05-25 Simon Hausmann <hausmann@kde.org>

1541

1542

Reviewed by Zack.

1543

1544

Fix crash in Qt JavaScript bindings when the arguments used on the Qt side are not

1545

registered with QMetaType.

1546

1547

* bindings/qt/qt_instance.cpp:

1548

(KJS::Bindings::QtInstance::invokeMethod):

1549

* bindings/qt/qt_runtime.cpp:

1550

1551

2007-05-24 Luciano Montanaro <mikelima@cirulla.net>

1552

1553

Reviewed by Darin

1554

1555

Patch for http://bugs.webkit.org/show_bug.cgi?id=13855

1556

Port patch 666176 to JavaScriptCore

1557

1558

- Renamed JSValue::downcast() to JSValue::asCell() which makes the

1559

function meaning cleaner. It's modeled after Harri Porten change in

1560

KDE trunk.

1561

1562

* kjs/collector.cpp:

1563

(KJS::Collector::protect):

1564

(KJS::Collector::unprotect):

1565

(KJS::Collector::collectOnMainThreadOnly):

1566

* kjs/object.h:

1567

(KJS::JSValue::isObject):

1568

* kjs/string_object.cpp:

1569

(KJS::StringProtoFunc::callAsFunction):

1570

* kjs/value.h:

1571

(KJS::JSValue::asCell):

1572

(KJS::JSValue::isNumber):

1573

(KJS::JSValue::isString):

1574

(KJS::JSValue::isObject):

1575

(KJS::JSValue::getNumber):

1576

(KJS::JSValue::getString):

1577

(KJS::JSValue::getObject):

1578

(KJS::JSValue::getUInt32):

1579

(KJS::JSValue::mark):

1580

(KJS::JSValue::marked):

1581

(KJS::JSValue::type):

1582

(KJS::JSValue::toPrimitive):

1583

(KJS::JSValue::toBoolean):

1584

(KJS::JSValue::toNumber):

1585

(KJS::JSValue::toString):

1586

(KJS::JSValue::toObject):

1587

1588

2007-05-18 Holger Hans Peter Freyther <zecke@selfish.org>

1589

1590

Reviewed by Mark Rowe.

1591

1592

* kjs/testkjs.pro: Make the Gdk port link to icu

1593

1594

2007-05-15 Geoffrey Garen <ggaren@apple.com>

1595

1596

Reviewed by Adele Peterson.

1597

1598

It helps if you swap the right variable.

1599

1600

* wtf/HashSet.h:

1601

(WTF::::operator):

1602

1603

2007-05-15 Lars Knoll <lars@trolltech.com>

1604

1605

Reviewed by Zack

1606

1607

Extend the QObject JavaScript bindings to work for slots with

1608

arguments.

1609

1610

* bindings/qt/qt_instance.cpp:

1611

(KJS::Bindings::QtInstance::invokeMethod):

1612

1613

2007-05-14 Kimmo Kinnunen <kimmok@iki.fi>

1614

1615

Reviewed by Darin.

1616

1617

- Fixes http://bugs.webkit.org/show_bug.cgi?id=13622 (Decompiler

1618

omits trailing comma in array literal)

1619

1620

* kjs/nodes2string.cpp:

1621

(ArrayNode::streamTo): print extra ',' in case there was elision

1622

commas (check opt member var) and array elements present

1623

in the array expression

1624

1625

2007-05-14 Geoffrey Garen <ggaren@apple.com>

1626

1627

Reviewed by Oliver Hunt.

1628

1629

Added HashMap::swap and HashSet::swap. WebCore now uses HashSet::swap.

1630

I figured while I was in the neighborhood I might as well add HashMap::swap,

1631

too.

1632

1633

* wtf/HashMap.h:

1634

(WTF::::operator):

1635

(WTF::::swap):

1636

* wtf/HashSet.h:

1637

(WTF::::operator):

1638

(WTF::::swap):

1639

1640

2007-05-11 Kimmo Kinnunen <kimmok@iki.fi>

1641

1642

Reviewed by Darin.

1643

1644

- Fix for bug http://bugs.webkit.org/show_bug.cgi?id=13620

1645

Bogus decompilation of "for (var j = 1 in [])"

1646

- ForInNode toString()'ed to syntax error if there was var decl

1647

and initializer

1648

- ForNode toStringed()'ed lost 'var ' if it was present

1649

1650

* kjs/nodes2string.cpp:

1651

(VarDeclListNode::streamTo): Print "var " here

1652

(VarStatementNode::streamTo): Don't print "var " here

1653

(ForNode::streamTo): Remove TODO comment, VarDeclListNode will

1654

stream the "var "

1655

(ForInNode::streamTo): ForIn initializer is printed by VarDeclNode

1656

1657

2007-05-11 Kimmo Kinnunen <kimmok@iki.fi>

1658

1659

Reviewed by Darin.

1660

1661

- Fixes http://bugs.webkit.org/show_bug.cgi?id=10878

1662

(Incorrect decompilation for "4..x")

1663

- Group numbers in dotted expressions in toString() output, so we

1664

avoid the 4.x constructs when the original input is 4..x.

1665

4..x means the same as 4. .x or (4).x or Number(4).x

1666

1667

* kjs/nodes2string.cpp:

1668

(KJS::SourceStream::):

1669

Add boolean flag to indicate that if next item is a number, it should be grouped.

1670

Add new formatting enum which turns on the boolean flag.

1671

(KJS::SourceStream::SourceStream): Added. Initialize the flag.

1672

(SourceStream::operator<<): Added. New overloaded operator with double value as parameter.

1673

(NumberNode::streamTo): Use the double operator

1674

(ArrayNode::streamTo):

1675

(DotAccessorNode::streamTo):

1676

(FunctionCallDotNode::streamTo):

1677

(FunctionCallParenDotNode::streamTo):

1678

(PostfixDotNode::streamTo):

1679

(DeleteDotNode::streamTo):

1680

(PrefixDotNode::streamTo):

1681

(AssignDotNode::streamTo): Use the new formatting enum to turn on the grouping flag.

1682

1683

2007-05-10 Lars Knoll <lars@trolltech.com>

1684

1685

Reviewed by Zack

1686

1687

Fix our last three test failures in the JavaScript

1688

tests.

1689

1690

* wtf/unicode/qt4/UnicodeQt4.h:

1691

(WTF::Unicode::toLower):

1692

(WTF::Unicode::toUpper):

1693

1694

2007-05-08 Geoffrey Garen <ggaren@apple.com>

1695

1696

Reviewed by Darin Adler.

1697

1698

Fixed #includes of JSStringRefCF.h and use of CF datatypes. I think I

1699

misunderstood this issue before.

1700

1701

* API/JavaScriptCore.h: #include JSStringRefCF.h. Platforms that don't

1702

want this behavior can just #include individual headers, instead of the

1703

umbrella framework header. But we definitely want Mac OS X clients to

1704

get the #include of JSStringRefCF.h "for free."

1705

* API/minidom.c: Don't #include JSStringRefCF.h. (Don't need to #include

1706

JavaScriptCore.h, either.)

1707

* API/testapi.c: Don't #include JSStringRefCF.h. Do use CF datatypes

1708

regardless of whether __APPLE__ is defined. Platforms that don't support

1709

CF just shouldn't compile this file.

1710

(main):

1711

1712

2007-05-09 Eric Seidel <eric@webkit.org>

1713

1714

Reviewed by mjs.

1715

1716

http://bugs.webkit.org/show_bug.cgi?id=6985

1717

Cyclic __proto__ values cause WebKit to hang

1718

1719

* kjs/object.cpp:

1720

(KJS::JSObject::put): do a cycle check before setting __proto__

1721

1722

2007-05-08 Kimmo Kinnunen <kimmok@iki.fi>

1723

1724

Reviewed by darin. Landed by eseidel.

1725

1726

- http://bugs.webkit.org/show_bug.cgi?id=10880 (Do..while loop gains

1727

a semicolon each time it is toStringed)

1728

Grammar in Ecma-66262, 12.6: "do Statement while ( Expression );"

1729

EmptyStatement was created after every do..while(expr) which

1730

had semicolon at the end.

1731

1732

* kjs/grammar.y: Require semicolon at the end of do..while

1733

1734

2007-05-08 Geoffrey Garen <ggaren@apple.com>

1735

1736

Build fix -- this time for sure.

1737

1738

APICast.h, being private, ends up in a different folder than JSValueRef.h,

1739

so we can't include one from the other using "". Instead, just forward

1740

declare the relevant data types.

1741

1742

* API/APICast.h:

1743

1744

2007-05-08 Geoffrey Garen <ggaren@apple.com>

1745

1746

Build fix: export APICast.h for WebCore and WebKit.

1747

1748

* JavaScriptCore.xcodeproj/project.pbxproj:

1749

1750

2007-05-04 Darin Adler <darin@apple.com>

1751

1752

Reviewed by Adele.

1753

1754

- fix http://bugs.webkit.org/show_bug.cgi?id=12821

1755

<rdar://problem/5007921> Number.toExponential doesn't work for negative numbers

1756

1757

* kjs/number_object.cpp: (NumberProtoFunc::callAsFunction):

1758

Added a call to fabs before calling log10.

1759

1760

2007-05-03 Holger Freyther <freyther@kde.org>

1761

1762

Reviewed by Zack, landed by Simon.

1763

This is bugzilla bug 13499.

1764

1765

* JavaScriptCore.pri: Place Qt into the qt-port scope

1766

* bindings/testbindings.pro: Place Qt into the qt-port scope

1767

* kjs/testkjs.pro: Place Qt into the qt-port scope

1768

* pcre/pcre.pri: Place Qt into the qt-port scope

1769

1770

2007-05-02 David Harrison <harrison@apple.com>

1771

1772

Reviewed by Antti.

1773

1774

<rdar://problem/5174862> Crash resulting from DeprecatedString::insert()

1775

1776

Added insertion support for more than one value.

1777

1778

* wtf/Vector.h:

1779

(WTF::::insert):

1780

Added support for inserting multiple values.

1781

1782

(WTF::::prepend):

1783

New. Insert at the start of vectors. Convenient for vectors used as strings.

1784

1785

2007-05-01 Jungshik Shin <jungshik.shin@gmail.com>

1786

1787

Reviewed by Alexey.

1788

1789

- get rid of non-ASCII lteral characters : suppress compiler warnings

1790

http://bugs.webkit.org/show_bug.cgi?id=13551

1791

1792

* kjs/testkjs.cpp:

1793

* pcre/pcre_compile.c:

1794

1795

2007-04-28 Jungshik Shin <jungshik.shin@gmail.com>

1796

1797

Reviewed by Sam Weinig.

1798

1799

- Replace copyright sign in Latin-1 (0xA9) with '(C)'

1800

http://bugs.webkit.org/show_bug.cgi?id=13531

1801

1802

* bindings/npruntime.h:

1803

1804

2007-04-28 Darin Adler <darin@apple.com>

1805

1806

Reviewed by Maciej.

1807

1808

- fix <rdar://problem/5154144> Hamachi test fails: assertion failure in ListHashSet

1809

1810

Test: fast/forms/add-remove-form-elements-stress-test.html

1811

1812

* wtf/ListHashSet.h:

1813

(WTF::ListHashSetNodeAllocator::ListHashSetNodeAllocator): Initialize

1814

m_isDoneWithInitialFreeList to false.

1815

(WTF::ListHashSetNodeAllocator::allocate): Added assertions based on a debug-only

1816

m_isAllocated flag that make sure we don't allocate a block that's already allocated.

1817

These assertions helped pinpoint the bug. Set m_isDoneWithInitialFreeList when we

1818

allocate the last block of the initial free list. Once we're done with the initial

1819

free list, turn off the rule that says that the next node in the pool after the last

1820

node in the free list is also free. This rule works because any free nodes are added

1821

to the head of the free list, so a node that hasn't been allocated even once is always

1822

at the tail of the free list and all the nodes after it also haven't been allocated

1823

even once. But it doesn't work any longer once the entire pool has been used at least

1824

once, because there's nothing special about the last node on the free list any more.

1825

(WTF::ListHashSetNodeAllocator::deallocate): Set the node's m_isAllocated to false.

1826

(WTF::ListHashSetNodeAllocator::pastPool): Added. Used above.

1827

(WTF::ListHashSetNodeAllocator::inPool): Changed to use the pastPool function.

1828

(WTF::ListHashSetNode::ListHashSetNode): Initialize m_isAllocated to true.

1829

(WTF::ListHashSetNode::operator new): Removed variable name for unused size

1830

parameter.

1831

(WTF::ListHashSetNode::destroy): Changed to call the destructor rather than

1832

delete -- this gets rid of the need to define an operator delete.

1833

1834

2007-04-27 Christopher Brichford <chrisb@adobe.com>

1835

1836

Reviewed by Timothy Hatcher.

1837

1838

Fix for: Bug 13211: Move JavaScriptCore mac project files for apollo port

1839

http://bugs.webkit.org/show_bug.cgi?id=13211

1840

1841

* JavaScriptCore.apolloproj/mac/JavaScriptCore.Debug.xcconfig: Added.

1842

* JavaScriptCore.apolloproj/mac/JavaScriptCore.Release.xcconfig: Added.

1843

* JavaScriptCore.apolloproj/mac/JavaScriptCore.xcconfig: Added.

1844

* JavaScriptCore.apolloproj/mac/JavaScriptCore.xcodeproj/project.pbxproj: Added.

1845

* JavaScriptCore.apolloproj/mac/JavaScriptCore/JavaScriptCore.Debug.xcconfig: Removed.

1846

* JavaScriptCore.apolloproj/mac/JavaScriptCore/JavaScriptCore.Release.xcconfig: Removed.

1847

* JavaScriptCore.apolloproj/mac/JavaScriptCore/JavaScriptCore.xcconfig: Removed.

1848

* JavaScriptCore.apolloproj/mac/JavaScriptCore/JavaScriptCore.xcodeproj/project.pbxproj: Removed.

1849

1850

2007-04-27 Holger Freyther <freyther@kde.org>

1851

1852

Reviewed by Maciej.

1853

1854

Remove unmaintained CMake build system.

1855

1856

* CMakeLists.txt: Removed.

1857

* pcre/CMakeLists.txt: Removed.

1858

1859

2007-04-27 Mark Rowe <mrowe@apple.com>

1860

1861

Reviewed by Oliver.

1862

1863

* JavaScriptCore.xcodeproj/project.pbxproj: Improve dependencies in Xcode project

1864

by marking dftables as a dependency of Generate Derived Sources rather than of

1865

JavaScriptCore itself.

1866

1867

2007-04-26 Geoffrey Garen <ggaren@apple.com>

1868

1869

Build fix -- added #includes that we used to get implicitly through

1870

JSStringRef.h.

1871

1872

* API/JSNode.c:

1873

* API/JSNodeList.c:

1874

* API/minidom.c:

1875

* API/testapi.c:

1876

1877

2007-04-26 Geoffrey Garen <ggaren@apple.com>

1878

1879

Reviewed by Maciej Stachowiak, Adam Roben.

1880

1881

Fixed

1882

<rdar://problem/4885130> Remove #include of JSStringRefCF.h from JSStringRef.h

1883

<rdar://problem/4885123> JavaScriptCore is not cross-platform -- JSStringRef.h references CF datatypes

1884

1885

* API/JSStringRef.h: Removed #include -- no clients need it anymore.

1886

1887

2007-04-25 David Kilzer <ddkilzer@apple.com>

1888

1889

Reviewed by Maciej.

1890

1891

Add assertions for debug builds.

1892

1893

* kjs/JSLock.cpp:

1894

(KJS::JSLock::lock): Assert the return value of pthread_mutex_lock() in debug builds.

1895

(KJS::JSLock::unlock): Assert the return value of pthread_mutex_unlock() in debug builds.

1896

1897

2007-04-25 Maciej Stachowiak <mjs@apple.com>

1898

1899

Reviewed by Anders.

1900

1901

- fix build problems

1902

1903

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: Disable warning that

1904

gives often downright incorrect results based on guessing what will happen in 64-bit.

1905

1906

2007-04-25 Darin Adler <darin@apple.com>

1907

1908

Reviewed by Geoff.

1909

1910

- tweak the allocator for a small speedup -- Shark showed this was a win, but I can't

1911

measure an improvement right now, but it's also clear these changes do no harm

1912

1913

* wtf/FastMalloc.cpp:

1914

(WTF::LgFloor): Use ALWAYS_INLINE here; in testing I did a while back this was necessary

1915

to get this single-instruction function to be inlined.

1916

(WTF::SizeClass): Use ALWAYS_INLINE here too for the same reason. Also change the special

1917

case for a size of 0 to work without a branch for a bit of extra speed.

1918

(WTF::ByteSizeForClass): Use ALWAYS_INLINE here too for the same reason.

1919

1920

2007-04-24 Maciej Stachowiak <mjs@apple.com>

1921

1922

Reviewed by Oliver.

1923

1924

- use custom calling convention for everything in nodes.cpp on intel gcc for 1.5% speed boost

1925

1926

Nearly all functions in nodes.cpp were marked up to use the

1927

regparm(3) calling convention under GCC for x86, since this is

1928

faster and they are all guaranteed to be called only internally to

1929

kjs.

1930

1931

The only exception is destructors, since delete doesn't know how to use a custom calling convention.

1932

1933

* kjs/nodes.cpp:

1934

(dotExprDoesNotAllowCallsString):

1935

* kjs/nodes.h:

1936

(KJS::Node::):

1937

(KJS::StatementNode::):

1938

(KJS::NullNode::):

1939

(KJS::BooleanNode::):

1940

(KJS::NumberNode::):

1941

(KJS::StringNode::):

1942

(KJS::RegExpNode::):

1943

(KJS::ThisNode::):

1944

(KJS::ResolveNode::):

1945

(KJS::GroupNode::):

1946

(KJS::ElementNode::):

1947

(KJS::ArrayNode::):

1948

(KJS::PropertyNameNode::):

1949

(KJS::PropertyNode::):

1950

(KJS::PropertyListNode::):

1951

(KJS::ObjectLiteralNode::):

1952

(KJS::BracketAccessorNode::):

1953

(KJS::DotAccessorNode::):

1954

(KJS::ArgumentListNode::):

1955

(KJS::ArgumentsNode::):

1956

(KJS::NewExprNode::):

1957

(KJS::FunctionCallValueNode::):

1958

(KJS::FunctionCallResolveNode::):

1959

(KJS::FunctionCallBracketNode::):

1960

(KJS::FunctionCallParenBracketNode::):

1961

(KJS::FunctionCallDotNode::):

1962

(KJS::FunctionCallParenDotNode::):

1963

(KJS::PostfixResolveNode::):

1964

(KJS::PostfixBracketNode::):

1965

(KJS::PostfixDotNode::):

1966

(KJS::PostfixErrorNode::):

1967

(KJS::DeleteResolveNode::):

1968

(KJS::DeleteBracketNode::):

1969

(KJS::DeleteDotNode::):

1970

(KJS::DeleteValueNode::):

1971

(KJS::VoidNode::):

1972

(KJS::TypeOfResolveNode::):

1973

(KJS::TypeOfValueNode::):

1974

(KJS::PrefixResolveNode::):

1975

(KJS::PrefixBracketNode::):

1976

(KJS::PrefixDotNode::):

1977

(KJS::PrefixErrorNode::):

1978

(KJS::UnaryPlusNode::):

1979

(KJS::NegateNode::):

1980

(KJS::BitwiseNotNode::):

1981

(KJS::LogicalNotNode::):

1982

(KJS::MultNode::):

1983

(KJS::AddNode::):

1984

(KJS::ShiftNode::):

1985

(KJS::RelationalNode::):

1986

(KJS::EqualNode::):

1987

(KJS::BitOperNode::):

1988

(KJS::BinaryLogicalNode::):

1989

(KJS::ConditionalNode::):

1990

(KJS::AssignResolveNode::):

1991

(KJS::AssignBracketNode::):

1992

(KJS::AssignDotNode::):

1993

(KJS::AssignErrorNode::):

1994

(KJS::CommaNode::):

1995

(KJS::AssignExprNode::):

1996

(KJS::VarDeclListNode::):

1997

(KJS::VarStatementNode::):

1998

(KJS::EmptyStatementNode::):

1999

(KJS::ExprStatementNode::):

2000

(KJS::IfNode::):

2001

(KJS::DoWhileNode::):

2002

(KJS::WhileNode::):

2003

(KJS::ForNode::):

2004

(KJS::ContinueNode::):

2005

(KJS::BreakNode::):

2006

(KJS::ReturnNode::):

2007

(KJS::WithNode::):

2008

(KJS::LabelNode::):

2009

(KJS::ThrowNode::):

2010

(KJS::TryNode::):

2011

(KJS::ParameterNode::):

2012

(KJS::Parameter::):

2013

(KJS::FunctionBodyNode::):

2014

(KJS::FuncExprNode::):

2015

(KJS::FuncDeclNode::):

2016

(KJS::SourceElementsNode::):

2017

(KJS::CaseClauseNode::):

2018

(KJS::ClauseListNode::):

2019

(KJS::SwitchNode::):

2020

2021

2007-04-24 Oliver Hunt <oliver@apple.com>

2022

2023

GTK Build fix, ::findEntry->KJS::findEntry

2024

2025

* kjs/lookup.cpp:

2026

(KJS::Lookup::findEntry):

2027

(KJS::Lookup::find):

2028

2029

2007-04-23 Maciej Stachowiak <mjs@apple.com>

2030

2031

Reviewed by Geoff.

2032

2033

- compile most of JavaScriptCore as one file for 4% JS iBench speed improvement

2034

2035

* JavaScriptCore.xcodeproj/project.pbxproj: Add AllInOneFile.cpp, and remove files it includes

2036

from the build.

2037

* kjs/AllInOneFile.cpp: Added.

2038

* kjs/dtoa.cpp: Renamed CONST to CONST_ to avoid conflict.

2039

(Bigint::):

2040

(Bigint::nrv_alloc):

2041

* kjs/lookup.cpp: Use "namspace KJS { ... }" instead of "using namespace KJS;"

2042

2043

2007-04-23 Maciej Stachowiak <mjs@apple.com>

2044

2045

Build fix, not reviewed.

2046

2047

* kjs/collector.h: Fix struct/class mismatch.

2048

2049

2007-04-23 Maciej Stachowiak <mjs@apple.com>

2050

2051

Reviewed by Darin.

2052

2053

- raise ALLOCATIONS_PER_COLLECTION to 4000, for 3.7% iBench speed improvement

2054

2055

Now that the cell size is smaller and the block size is bigger, we can fit 4000 objects in

2056

the two spare cells the collector is willing to keep around, so collect a bit less often.

2057

2058

* kjs/collector.cpp:

2059

2060

2007-04-23 Maciej Stachowiak <mjs@apple.com>

2061

2062

Reviewed by Darin and Geoff.

2063

2064

- move mark and collectOnMainThreadOnly bits into separate bitmaps

2065

2066

This saves 4 bytes per cell, allowing shrink of cell size to 32,

2067

which leads to a .8% speed improvement on iBench.

2068

2069

This is only feasible because of all the previous changes on the branch.

2070

2071

* kjs/collector.cpp:

2072

(KJS::allocateBlock): Adjust for some renames of constants.

2073

(KJS::Collector::markStackObjectsConservatively): Now that cells are 32 bytes (64

2074

bytes on 64-bit) the cell alignment check can be made much more strict, and also

2075

obsoletes the need for a % sizeof(CollectorCell) check. Also, we can mask off the low

2076

bits of the pointer to have a potential block pointer to look for.

2077

(KJS::Collector::collectOnMainThreadOnly): Use bitmap.

2078

(KJS::Collector::markMainThreadOnlyObjects): Use bitmap.

2079

(KJS::Collector::collect): When sweeping, use bitmaps directly to find mark bits.

2080

* kjs/collector.h:

2081

(KJS::): Move needed constants and type declarations here.

2082

(KJS::CollectorBitmap::get): Bit twiddling to get a bitmap value.

2083

(KJS::CollectorBitmap::set): Bit twiddling to set a bitmap bit to true.

2084

(KJS::CollectorBitmap::clear): Bit twiddling to set a bitmap bit to false.

2085

(KJS::CollectorBitmap::clearAll): Clear whole bitmap at one go.

2086

(KJS::Collector::cellBlock): New operation, compute the block pointer for

2087

a cell by masking off low bits.

2088

(KJS::Collector::cellOffset): New operation, compute the cell offset for a

2089

cell by masking off high bits and dividing (actually a shift).

2090

(KJS::Collector::isCellMarked): Check mark bit in bitmap

2091

(KJS::Collector::markCell): Set mark bit in bitmap.

2092

* kjs/value.h:

2093

(KJS::JSCell::JSCell): No more bits.

2094

(KJS::JSCell::marked): Let collector handle it.

2095

(KJS::JSCell::mark): Let collector handle it.

2096

2097

2007-04-23 Anders Carlsson <andersca@apple.com>

2098

2099

Build fix.

2100

2101

* kjs/regexp_object.h:

2102

RegExpObjectImpPrivate is a struct, not a class.

2103

2104

2007-04-23 Maciej Stachowiak <mjs@apple.com>

2105

2106

Reviewed by Darin.

2107

2108

- shrink FunctionImp / DeclaredFunctionImp by 4 bytes, by moving parameter list to function body

2109

2110

I reconciled this with a similar change in KDE kjs by Maks Orlovich <maksim@kde.org>.

2111

2112

* kjs/function.cpp:

2113

(KJS::FunctionImp::callAsFunction):

2114

(KJS::FunctionImp::passInParameters):

2115

(KJS::FunctionImp::lengthGetter):

2116

(KJS::FunctionImp::getParameterName):

2117

* kjs/function.h:

2118

* kjs/function_object.cpp:

2119

(FunctionProtoFunc::callAsFunction):

2120

(FunctionObjectImp::construct):

2121

* kjs/nodes.cpp:

2122

(FunctionBodyNode::addParam):

2123

(FunctionBodyNode::paramString):

2124

(FuncDeclNode::addParams):

2125

(FuncDeclNode::processFuncDecl):

2126

(FuncExprNode::addParams):

2127

(FuncExprNode::evaluate):

2128

* kjs/nodes.h:

2129

(KJS::Parameter::Parameter):

2130

(KJS::FunctionBodyNode::numParams):

2131

(KJS::FunctionBodyNode::paramName):

2132

(KJS::FunctionBodyNode::parameters):

2133

(KJS::FuncExprNode::FuncExprNode):

2134

(KJS::FuncDeclNode::FuncDeclNode):

2135

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: Disable 64-bit warnings because

2136

they handle size_t badly.

2137

2138

2007-04-23 Maciej Stachowiak <mjs@apple.com>

2139

2140

Reviewed by Darin.

2141

2142

- shrink RegexpObjectImp by 4 bytes

2143

2144

Somewhat inexplicably, this seems to be a .33% speedup on JS iBench.

2145

2146

* kjs/regexp_object.cpp:

2147

(KJS::RegExpObjectImpPrivate::RegExpObjectImpPrivate):

2148

(RegExpObjectImp::RegExpObjectImp):

2149

(RegExpObjectImp::performMatch):

2150

(RegExpObjectImp::arrayOfMatches):

2151

(RegExpObjectImp::getBackref):

2152

(RegExpObjectImp::getLastMatch):

2153

(RegExpObjectImp::getLastParen):

2154

(RegExpObjectImp::getLeftContext):

2155

(RegExpObjectImp::getRightContext):

2156

(RegExpObjectImp::getValueProperty):

2157

(RegExpObjectImp::putValueProperty):

2158

* kjs/regexp_object.h:

2159

2160

2007-04-23 Maciej Stachowiak <mjs@apple.com>

2161

2162

Reviewed by Oliver.

2163

2164

- change to 1-bit bitfields instead of 8-bit, this turns out to lead to a .51% speedup on JS iBench

2165

2166

The 1-bit bitfields are actually faster than just plain bools, at least on Intel (go figure).

2167

2168

* kjs/property_map.h:

2169

2170

2007-04-23 Maciej Stachowiak <mjs@apple.com>

2171

2172

Reviewed by Darin.

2173

2174

- shrink ArrayInstance objects by 4 bytes

2175

http://bugs.webkit.org/show_bug.cgi?id=13386

2176

2177

I did this by storing the capacity before the beginning of the storage array. It turns out

2178

it is rarely needed and is by definition 0 when the storage array is null.

2179

2180

* kjs/array_instance.h:

2181

(KJS::ArrayInstance::capacity): Get it from the secret stash

2182

* kjs/array_object.cpp:

2183

(allocateStorage): New function to encapsulate allocating the storage with extra space ahead

2184

for the capacity.

2185

(reallocateStorage): ditto for realloc

2186

(ArrayInstance::ArrayInstance):

2187

(ArrayInstance::~ArrayInstance):

2188

(ArrayInstance::resizeStorage):

2189

2190

2007-04-23 Darin Adler <darin@apple.com>

2191

2192

Reviewed by Maciej.

2193

2194

- fix <rdar://problem/4840688> REGRESSION (r10588, r10621): JavaScript won't parse

2195

modifications of non-references (breaks 300themovie.warnerbros.com, fedex.com)

2196

2197

Despite the ECMAScript specification's claim that you can treat these as syntax

2198

errors, doing so creates some website incompatibilities. So this patch turns them back

2199

into evaluation errors instead.

2200

2201

Test: fast/js/modify-non-references.html

2202

2203

* kjs/grammar.y: Change makeAssignNode, makePrefixNode, and makePostfixNode so that they

2204

never fail to parse. Update rules that use them. Fix a little bit of indenting. Use

2205

new PostfixErrorNode, PrefixErrorNode, and AssignErrorNode classes.

2206

2207

* kjs/nodes.h: Added an overload of throwError that takes a char* argument.

2208

Replaced setExceptionDetailsIfNeeded and debugExceptionIfNeeded with handleException,

2209

which does both. Added PostfixErrorNode, PrefixErrorNode, and AssignErrorNode classes.

2210

2211

* kjs/nodes.cpp: Changed exception macros to use handleException; simpler and smaller

2212

code size than the two functions that we used before.

2213

(Node::throwError): Added the overload mentioned above.

2214

(Node::handleException): Added. Contains the code from both setExceptionDetailsIfNeeded

2215

and debugExceptionIfNeeded.

2216

(PostfixErrorNode::evaluate): Added. Throws an exception.

2217

(PrefixErrorNode::evaluate): Ditto.

2218

(AssignErrorNode::evaluate): Ditto.

2219

(ThrowNode::execute): Call handleException instead of debugExceptionIfNeeded; this

2220

effectively adds a call to setExceptionDetailsIfNeeded, which may help with getting

2221

the correct file and line number for these exceptions.

2222

2223

* kjs/nodes2string.cpp:

2224

(PostfixErrorNode::streamTo): Added.

2225

(PrefixErrorNode::streamTo): Added.

2226

(AssignErrorNode::streamTo): Added.

2227

2228

2007-04-23 Maciej Stachowiak <mjs@apple.com>

2229

2230

Reviewed by Darin.

2231

2232

- fix test failures / crashes on PPC

2233

2234

* kjs/property_map.h: Make the bool fields explicitly 8-bit bitfields, since bool is a full

2235

word there otherwise :-(

2236

2237

2007-04-23 Maciej Stachowiak <mjs@apple.com>

2238

2239

Reviewed by Darin.

2240

2241

- fix more test case failures

2242

2243

* bindings/runtime_array.cpp:

2244

(RuntimeArray::RuntimeArray): inherit from JSObject instead of ArrayInstance; it turns

2245

out that this class only needs the prototype and classInfo from ArrayInstance, not the

2246

actual class itself, and it was too big otherwise.

2247

(RuntimeArray::getOwnPropertySlot):

2248

* bindings/runtime_array.h:

2249

2250

2007-04-23 Maciej Stachowiak <mjs@apple.com>

2251

2252

Reviewed by Darin.

2253

2254

- fix some test failures

2255

2256

* bindings/runtime_method.cpp:

2257

(RuntimeMethod::RuntimeMethod): inherit from InternalFunctionImp instead of FunctionImpl,

2258

otherwise this is too big

2259

(RuntimeMethod::getOwnPropertySlot):

2260

* bindings/runtime_method.h:

2261

2262

2007-04-22 Maciej Stachowiak <mjs@apple.com>

2263

2264

Reviewed by Darin.

2265

2266

- discard the arguments List for an ActivationImp when the corresponding Context is destroyed (1.7% speedup)

2267

http://bugs.webkit.org/show_bug.cgi?id=13385

2268

2269

Based an idea by Christopher E. Hyde <C.Hyde@parableuk.force9.co.uk>. His patch to do

2270

this also had many other List changes and I found this much simpler subset of the changes

2271

was actually a hair faster.

2272

2273

This optimization is valid because the arguments list is only kept around to

2274

lazily make the arguments object. If it's not made by the time the function

2275

exits, it never will be, since any function that captures the continuation will

2276

have its own local arguments variable in scope.

2277

2278

Besides the 1.7% speed improvement, it shrinks List by 4 bytes

2279

(which in turn shrinks ActivationImp by 4 bytes).

2280

2281

* kjs/Context.cpp:

2282

(KJS::Context::~Context): Clear the activation's arguments list.

2283

* kjs/function.cpp:

2284

(KJS::ActivationImp::ActivationImp): Adjusted for list changes.

2285

(KJS::ActivationImp::mark): No need to mark, lists are always protected (this doesn't

2286

cause a ref-cycle for reasons stated above).

2287

(KJS::ActivationImp::createArgumentsObject): Clear arguments list.

2288

* kjs/function.h:

2289

* kjs/list.cpp:

2290

(KJS::List::List): No more needsMarking boolean

2291

(KJS::List::operator=): ditto

2292

* kjs/list.h:

2293

(KJS::List::List): ditto

2294

(KJS::List::reset): ditto

2295

(KJS::List::deref): ditto

2296

2297

2007-04-22 Maciej Stachowiak <mjs@apple.com>

2298

2299

Reviewed by Darin.

2300

2301

- shrink PropertyMap by 8 bytes and therefore shrink CELL_SIZE to 40 (for 32-bit;

2302

similar shrinkage for 64-bit)

2303

http://bugs.webkit.org/show_bug.cgi?id=13384

2304

2305

Inspired by similar changes by Christopher E. Hyde <C.Hyde@parableuk.force9.co.uk>

2306

done in the kjs-tweaks branch of KDE's kjs. However, this version is somewhat

2307

cleaner style-wise and avoids some of the negative speed impact (at least on gcc/x86)

2308

of his version.

2309

2310

This is nearly a wash performance-wise, maybe a slight slowdown, but worth doing

2311

to eventually reach cell size 32.

2312

2313

* kjs/collector.cpp:

2314

(KJS::):

2315

* kjs/property_map.cpp:

2316

(KJS::PropertyMap::~PropertyMap):

2317

(KJS::PropertyMap::clear):

2318

(KJS::PropertyMap::get):

2319

(KJS::PropertyMap::getLocation):

2320

(KJS::PropertyMap::put):

2321

(KJS::PropertyMap::insert):

2322

(KJS::PropertyMap::expand):

2323

(KJS::PropertyMap::rehash):

2324

(KJS::PropertyMap::remove):

2325

(KJS::PropertyMap::mark):

2326

(KJS::PropertyMap::containsGettersOrSetters):

2327

(KJS::PropertyMap::getEnumerablePropertyNames):

2328

(KJS::PropertyMap::getSparseArrayPropertyNames):

2329

(KJS::PropertyMap::save):

2330

(KJS::PropertyMap::checkConsistency):

2331

* kjs/property_map.h:

2332

(KJS::PropertyMap::hasGetterSetterProperties):

2333

(KJS::PropertyMap::setHasGetterSetterProperties):

2334

(KJS::PropertyMap::):

2335

(KJS::PropertyMap::PropertyMap):

2336

2337

2007-04-22 Maciej Stachowiak <mjs@apple.com>

2338

2339

Reviewed by Darin.

2340

2341

- change blocks to 64k in size, and use various platform-specific calls to allocate at 64k-aligned addresses

2342

http://bugs.webkit.org/show_bug.cgi?id=13383

2343

2344

* kjs/collector.cpp:

2345

(KJS::allocateBlock): New function to allocate 64k of 64k-aligned memory

2346

(KJS::freeBlock): Corresponding free

2347

(KJS::Collector::allocate):

2348

(KJS::Collector::collect):

2349

2350

2007-04-22 Maciej Stachowiak <mjs@apple.com>

2351

2352

Reviewed by Darin and Geoff.

2353

2354

- remove the concept of oversize objects, now that there aren't any (for now

2355

only enforced with an assert).

2356

http://bugs.webkit.org/show_bug.cgi?id=13382

2357

2358

This change is a .66% speedup on JS iBench for 32-bit platforms, probably much more

2359

for 64-bit since it finally gives a reasonable cell size, but I did not test that.

2360

2361

* kjs/collector.cpp:

2362

(KJS::): Use different cell size for 32-bit and 64-bit, now that there is no

2363

oversize allocation.

2364

(KJS::Collector::allocate): Remove oversize allocator.

2365

(KJS::Collector::markStackObjectsConservatively): Don't check oversize objects.

2366

(KJS::Collector::markMainThreadOnlyObjects): Ditto.

2367

(KJS::Collector::collect): Ditto.

2368

2369

2007-04-21 Mitz Pettel <mitz@webkit.org>

2370

2371

Reviewed by Adam.

2372

2373

- fix http://bugs.webkit.org/show_bug.cgi?id=13428

2374

REGRESSION (r20973-r20976): Failing ecma/Array/15.4.4.5-3.js

2375

2376

- fix http://bugs.webkit.org/show_bug.cgi?id=13429

2377

REGRESSION (r20973-r20976): Crashing in fast/dom/plugin-attributes-enumeration.html

2378

2379

* kjs/array_object.cpp:

2380

(ArrayInstance::sort): Free the old storage, not the new one.

2381

2382

2007-04-20 Maciej Stachowiak <mjs@apple.com>

2383

2384

Not reviewed, build fix.

2385

2386

- fix build problem with last change - -O3 complains more about uninitialized variables

2387

2388

* pcre/pcre_compile.c:

2389

(compile_branch):

2390

(pcre_compile2):

2391

2392

2007-04-20 Maciej Stachowiak <mjs@apple.com>

2393

2394

Reviewed by Darin.

2395

2396

- <rdar://problem/5149915> use mergesort when possible, since it leads to fewer compares (2% JS iBench speedup)

2397

2398

* kjs/array_object.cpp:

2399

(ArrayInstance::sort): Use mergesort(3) on platforms that have it, since it tends

2400

to do fewer compares than qsort; but avoid it very on large arrays since it uses extra

2401

memory. Also added comments identifying possibly even better sorting algorithms

2402

for sort by string value and sort by compare function.

2403

* kjs/config.h:

2404

2405

2007-04-20 Maciej Stachowiak <mjs@apple.com>

2406

2407

Reviewed by Darin.

2408

2409

- bump optimization flags up to -O3 for 1% JS iBench speed improvement

2410

2411

* Configurations/Base.xcconfig:

2412

2413

2007-04-20 Mark Rowe <mrowe@apple.com>

2414

2415

Reviewed by Maciej.

2416

2417

Fix bogus optimisation in the generic pthread code path.

2418

2419

* kjs/collector.cpp:

2420

(KJS::currentThreadStackBase):

2421

2422

2007-04-20 Mark Rowe <mrowe@apple.com>

2423

2424

Reviewed by Anders.

2425

2426

Improve FreeBSD compatibility, as suggested by Alexander Botero-Lowry.

2427

2428

* kjs/collector.cpp:

2429

(KJS::currentThreadStackBase): FreeBSD requires that pthread_attr_t's are

2430

initialized via pthread_attr_init before being used in any context.

2431

2432

2007-04-19 Mark Rowe <mrowe@apple.com>

2433

2434

Reviewed by Darin.

2435

2436

Fix http://bugs.webkit.org/show_bug.cgi?id=13401

2437

Bug 13401: Reproducible crash calling myArray.sort(compareFn) from within

2438

a sort comparison function

2439

2440

* kjs/array_object.cpp:

2441

(ArrayInstance::sort): Save/restore the static variables around calls to qsort

2442

to ensure nested calls to ArrayInstance::sort behave correctly.

2443

2444

2007-04-12 Deneb Meketa <dmeketa@adobe.com>

2445

2446

Reviewed by Darin Adler.

2447

2448

http://bugs.webkit.org/show_bug.cgi?id=13029

2449

rdar://problem/4994849

2450

Bug 13029: Permit NPAPI plug-ins to see HTTP response headers.

2451

This doesn't actually change JavaScriptCore, but that's where npapi.h is.

2452

2453

* bindings/npapi.h:

2454

Add headers member to NPStream struct. Also increase NP_VERSION_MINOR to 18.

2455

Increasing to >= 17 allows plug-ins to safely detect whether to look for

2456

NPStream::headers. Increasing from 17 to 18 reflects presence of NPObject

2457

enumeration, which was added in a prior patch, and which has been agreed to

2458

constitute version 18 by the plugin-futures list. Also add other missing

2459

bits of npapi.h to catch up from 14 to 18. This includes features that are

2460

not implemented in WebKit, but those are safely stubbed.

2461

2462

2007-04-10 Geoffrey Garen <ggaren@apple.com>

2463

2464

Reviewed by Mark Rowe.

2465

2466

Fixed last check-in to print in release builds, too.

2467

2468

* kjs/collector.cpp:

2469

(KJS::getPlatformThreadRegisters):

2470

2471

2007-04-10 Geoffrey Garen <ggaren@apple.com>

2472

2473

Reviewed by John Sullivan, Darin Adler.

2474

2475

Fixed <rdar://problem/5121899> JavaScript garbage collection leads to

2476

later crash under Rosetta (should abort or leak instead?)

2477

2478

Log an error message and crash if the kernel reports failure during GC.

2479

We decided to do this instead of just leaking because we don't want people

2480

to get the mistaken impression that running in Rosetta is a supported

2481

configurtion.

2482

2483

The CRASH macro will also hook into CrashReporter, which will tell us if

2484

many (any?) users run into this issue.

2485

2486

* kjs/collector.cpp:

2487

(KJS::getPlatformThreadRegisters):

2488

2489

2007-04-06 Krzysztof Kowalczyk <kkowalczyk@gmail.com>

2490

2491

Reviewed by darin.

2492

2493

Coverity fix. Coverity says:

2494

"Event var_deref_model: Variable "sourceRanges" tracked as NULL was passed to a

2495

function that dereferences it"

2496

2497

* kjs/string_object.cpp:

2498

(KJS::replace):

2499

2500

2007-04-06 Geoffrey Garen <ggaren@apple.com>

2501

2502

Rubber stamped by Adele Peterson.

2503

2504

* kjs/ExecState.h: Removed obsolete forward/friend declaration of

2505

RuntimeMethodImp.

2506

2507

2007-04-05 Krzysztof Kowalczyk <kkowalczyk@gmail.com>

2508

2509

Reviewed by darin.

2510

2511

Coverity fix. Coverity says:

2512

"Event check_after_deref: Pointer "dateString" dereferenced before NULL check"

2513

2514

* kjs/date_object.cpp:

2515

(KJS::parseDate):

2516

2517

2007-04-05 Krzysztof Kowalczyk <kkowalczyk@gmail.com>

2518

2519

Reviewed by darin.

2520

2521

Coverity fix. Coverity says:

2522

"Event check_after_deref: Pointer "re" dereferenced before NULL check"

2523

2524

* pcre/pcre_study.c:

2525

(pcre_study):

2526

2527

2007-04-05 Krzysztof Kowalczyk <kkowalczyk@gmail.com>

2528

2529

Reviewed by darin.

2530

2531

Coverity fixes. Coverity says:

2532

"Event leaked_storage: Returned without freeing storage "buffer""

2533

and:

2534

"Event leaked_storage: Returned without freeing storage "script""

2535

2536

* kjs/testkjs.cpp:

2537

(doIt):

2538

(createStringWithContentsOfFile):

2539

2540

2007-04-05 Krzysztof Kowalczyk <kkowalczyk@gmail.com>

2541

2542

Reviewed by darin.

2543

2544

Coverity fix: in single-threaded case currentThreadIsMainThread is always true

2545

so the code in if (!currentThreadIsMainThread) cannot possibly be reached

2546

and Coverity complains about dead code.

2547

2548

* kjs/collector.cpp:

2549

(KJS::Collector::collect):

2550

2551

=== Safari-5522.6 ===

2552

2553

2007-04-03 Kevin McCullough <kmccullough@apple.com>

2554

2555

Reviewed by Adam.

2556

2557

- Testing a post-commit hook.

2558

2559

* JavaScriptCore.vcproj/testkjs/testkjs.vcproj:

2560

2561

2007-04-03 Anders Carlsson <andersca@apple.com>

2562

2563

Reviewed by Adam.

2564

2565

<rdar://problem/5107534>

2566

http://bugs.webkit.org/show_bug.cgi?id=13265

2567

REGRESSION: Crash in KJS::Bindings::convertValueToNPVariant

2568

2569

* bindings/NP_jsobject.cpp:

2570

(_NPN_InvokeDefault):

2571

Return false if the object isn't a function. Set the return value to undefined by default

2572

(to match Firefox).

2573

2574

2007-03-30 Anders Carlsson <andersca@apple.com>

2575

2576

Build fix.

2577

2578

* bindings/NP_jsobject.cpp:

2579

(_NPN_Enumerate):

2580

2581

2007-03-30 Anders Carlsson <andersca@apple.com>

2582

2583

Reviewed by Geoff.

2584

2585

Implement _NPN_Enumerate support.

2586

2587

* JavaScriptCore.exp:

2588

* bindings/NP_jsobject.cpp:

2589

(_NPN_Enumerate):

2590

* bindings/c/c_instance.cpp:

2591

(KJS::Bindings::CInstance::getPropertyNames):

2592

* bindings/c/c_instance.h:

2593

* bindings/npapi.h:

2594

* bindings/npruntime.h:

2595

* bindings/npruntime_impl.h:

2596

* bindings/runtime.h:

2597

(KJS::Bindings::Instance::getPropertyNames):

2598

* bindings/runtime_object.cpp:

2599

(RuntimeObjectImp::getPropertyNames):

2600

* bindings/runtime_object.h:

2601

(KJS::RuntimeObjectImp::getInternalInstance):

2602

2603

2007-03-28 Jeff Walden <jwalden+code@mit.edu>

2604

2605

Reviewed by Darin.

2606

2607

http://bugs.webkit.org/show_bug.cgi?id=12963

2608

Fix some inconsistencies in the Mozilla JS Array extras implementations

2609

with respect to the Mozilla implementation:

2610

2611

- holes in arrays should be skipped, not treated as undefined,

2612

by all such methods

2613

- an element with value undefined is not a hole

2614

- Array.prototype.forEach should return undefined

2615

2616

* kjs/array_object.cpp:

2617

(ArrayInstance::getOwnPropertySlot):

2618

(ArrayProtoFunc::callAsFunction):

2619

2620

2007-03-27 Anders Carlsson <acarlsson@apple.com>

2621

2622

Reviewed by Geoff.

2623

2624

* bindings/NP_jsobject.cpp:

2625

(_NPN_InvokeDefault):

2626

Call JSObject:call for native JavaScript objects.

2627

2628

2007-03-26 David Carson <dacarson@gmail.com>

2629

2630

Reviewed by Darin, landed by Anders.

2631

2632

Fix for: REGRESSION (r19559): Java applet crash

2633

http://bugs.webkit.org/show_bug.cgi?id=13142

2634

<rdar://problem/5080340>

2635

2636

The previous fix http://bugs.webkit.org/show_bug.cgi?id=12636

2637

introduced new JNIType to enum in jni_utility.h This is a

2638

problem on the Mac as it seems that the JNIType enum is also

2639

used in the JVM, it is used to specify the return type in

2640

jni_objc.mm

2641

Corrected the fix by moving type to the end, and changing

2642

jni_objc.mm to convert the new type to an old compatible

2643

type.

2644

2645

* bindings/jni/jni_objc.mm:

2646

(KJS::Bindings::dispatchJNICall):

2647

* bindings/jni/jni_utility.h:

2648

2649

2007-03-26 Christopher Brichford <chrisb@adobe.com>

2650

2651

Reviewed/landed by Adam.

2652

2653

Bug 13198: Move build settings from project file to xcconfig file for apollo

2654

port JSCore

2655

http://bugs.webkit.org/show_bug.cgi?id=13198

2656

2657

- Moving build settings from xcode project file to xcconfig files.

2658

2659

* JavaScriptCore.apolloproj/mac/JavaScriptCore/JavaScriptCore.Debug.xcconfig:

2660

* JavaScriptCore.apolloproj/mac/JavaScriptCore/JavaScriptCore.Release.xcconfig:

2661

* JavaScriptCore.apolloproj/mac/JavaScriptCore/JavaScriptCore.xcconfig:

2662

* JavaScriptCore.apolloproj/mac/JavaScriptCore/JavaScriptCore.xcodeproj/project.pbxproj:

2663

2664

2007-03-26 Brady Eidson <beidson@apple.com>

2665

2666

Rubberstamped by Anders and Maciej aand Geoff (oh my!)

2667

2668

Since CFTypeRef is really void*, a RetainPtr couldn't be used.

2669

RefType was "void", which doesn't actually exist as a type.

2670

Since RefType only existed for operator*(), and since that operator

2671

doesn't make any sense for RetainPtr, I removed them!

2672

2673

* kjs/nodes.cpp: Touch this to force a rebuild and (hopefully) help the

2674

compiler with dependencies

2675

* wtf/RetainPtr.h: Nuke RefType and operator*()

2676

2677

2007-03-26 Geoffrey Garen <ggaren@apple.com>

2678

2679

Touched a file to (hopefully) help the compiler with RetainPtr dependencies.

2680

2681

* kjs/nodes.cpp:

2682

(Node::deref):

2683

2684

2007-03-24 Brady Eidson <beidson@apple.com>

2685

2686

Reviewed by Adam

2687

2688

Whoops, RetainPtr should be in the WTF namespace

2689

2690

* wtf/RetainPtr.h:

2691

2692

2007-03-24 Brady Eidson <beidson@apple.com>

2693

2694

Reviewed by Adam

2695

2696

<rdar://problem/5086210> - Move RetainPtr to WTF

2697

2698

* wtf/RetainPtr.h: Added

2699

* JavaScriptCore.xcodeproj/project.pbxproj: Add it to the project file

2700

* JavaScriptCore.vcproj/WTF/WTF.vcproj: Ditto

2701

2702

2703

2007-03-23 Christopher Brichford <chrisb@adobe.com>

2704

2705

Reviewed/landed by Adam.

2706

2707

Bug 13175: Make apollo mac project files for JavaScriptCore actually

2708

build something

2709

http://bugs.webkit.org/show_bug.cgi?id=13175

2710

2711

- Changing apollo mac project files for JavaScriptCore such that they actually build

2712

JavaScriptCore source code.

2713

2714

* JavaScriptCore.apolloproj/ForwardingSources/grammar.cpp: Added.

2715

* JavaScriptCore.apolloproj/mac/JavaScriptCore/JavaScriptCore.xcconfig:

2716

* JavaScriptCore.apolloproj/mac/JavaScriptCore/JavaScriptCore.xcodeproj/project.pbxproj:

2717

2718

2007-03-24 Mark Rowe <mrowe@apple.com>

2719

2720

Rubber-stamped by Darin.

2721

2722

* Configurations/JavaScriptCore.xcconfig: Remove unnecessary INFOPLIST_PREPROCESS.

2723

2724

2007-03-22 Christopher Brichford <chrisb@adobe.com>

2725

2726

Reviewed/landed by Adam.

2727

2728

Bug 13164: Initial version of mac JavaScriptCore project files for

2729

apollo port

2730

http://bugs.webkit.org/show_bug.cgi?id=13164

2731

2732

- Adding mac project files for apollo port of JavaScriptCore. Currently project

2733

just builds dftables.

2734

2735

* JavaScriptCore.apolloproj/mac/JavaScriptCore/JavaScriptCore.Debug.xcconfig: Added.

2736

* JavaScriptCore.apolloproj/mac/JavaScriptCore/JavaScriptCore.Release.xcconfig: Added.

2737

* JavaScriptCore.apolloproj/mac/JavaScriptCore/JavaScriptCore.xcconfig: Added.

2738

* JavaScriptCore.apolloproj/mac/JavaScriptCore/JavaScriptCore.xcodeproj/project.pbxproj: Added.

2739

2740

2007-03-21 Timothy Hatcher <timothy@apple.com>

2741

2742

Reviewed by Darin.

2743

2744

<rdar://problem/5076599> JavaScriptCore has a weak export (vtable for KJS::JSCell)

2745

2746

* JavaScriptCore.exp: Remove __ZTVN3KJS6JSCellE.

2747

2748

2007-03-21 Adele Peterson <adele@apple.com>

2749

2750

Reviewed by Geoff.

2751

2752

* API/JSStringRef.cpp: (JSStringIsEqual): Added JSLock.

2753

2754

2007-03-21 Zack Rusin <zrusin@trolltech.com>

2755

2756

Fix the compile when USE(MULTIPLE_THREADS) isn't

2757

defined

2758

2759

* kjs/JSLock.cpp:

2760

(KJS::JSLock::currentThreadIsHoldingLock):

2761

2762

2007-03-20 Maciej Stachowiak <mjs@apple.com>

2763

2764

Reviewed by Geoff and Adam.

2765

2766

- make USE(MULTIPLE_THREADS) support more portable

2767

http://bugs.webkit.org/show_bug.cgi?id=13069

2768

2769

- fixed a threadsafety bug discovered by testing this

2770

2771

- enhanced threadsafety assertions in collector

2772

2773

* API/JSCallbackObject.cpp:

2774

(KJS::JSCallbackObject::~JSCallbackObject): This destructor can't

2775

DropAllLocks around the finalize callback, because it gets called

2776

from garbage collection and we can't let other threads collect!

2777

2778

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

2779

* kjs/JSLock.cpp:

2780

(KJS::JSLock::currentThreadIsHoldingLock): Added new function

2781

to allow stronger assertions than just that the lock is held

2782

by some thread (you can now assert that the current thread is

2783

holding it, given the new JSLock design).

2784

* kjs/JSLock.h:

2785

* kjs/collector.cpp: Refactored for portability plus added some

2786

stronger assertions.

2787

(KJS::Collector::allocate):

2788

(KJS::currentThreadStackBase):

2789

(KJS::Collector::registerAsMainThread):

2790

(KJS::onMainThread):

2791

(KJS::PlatformThread::PlatformThread):

2792

(KJS::getCurrentPlatformThread):

2793

(KJS::Collector::Thread::Thread):

2794

(KJS::destroyRegisteredThread):

2795

(KJS::Collector::registerThread):

2796

(KJS::Collector::markCurrentThreadConservatively):

2797

(KJS::suspendThread):

2798

(KJS::resumeThread):

2799

(KJS::getPlatformThreadRegisters):

2800

(KJS::otherThreadStackPointer):

2801

(KJS::otherThreadStackBase):

2802

(KJS::Collector::markOtherThreadConservatively):

2803

(KJS::Collector::markStackObjectsConservatively):

2804

(KJS::Collector::protect):

2805

(KJS::Collector::unprotect):

2806

(KJS::Collector::collectOnMainThreadOnly):

2807

(KJS::Collector::markMainThreadOnlyObjects):

2808

(KJS::Collector::collect):

2809

* kjs/collector.h:

2810

* wtf/FastMalloc.cpp:

2811

(WTF::fastMallocSetIsMultiThreaded):

2812

* wtf/FastMallocInternal.h:

2813

* wtf/Platform.h:

2814

2815

2007-03-19 Darin Adler <darin@apple.com>

2816

2817

* kjs/value.h: Roll ~JSValue change out. It was causing problems. I'll do it right later.

2818

2819

2007-03-19 Geoffrey Garen <ggaren@apple.com>

2820

2821

Reviewed by John Sullivan.

2822

2823

Fixed <rdar://problem/5073380> REGRESSION: Crash occurs at WTF::fastFree()

2824

when reloading liveconnect page (applet)

2825

2826

Best to use free when you use malloc, especially when malloc and delete

2827

use completely different libraries.

2828

2829

* bindings/jni/jni_runtime.cpp:

2830

(JavaMethod::~JavaMethod):

2831

2832

2007-03-19 Andrew Wellington <proton@wiretapped.net>

2833

2834

Reviewed by Maciej.

2835

2836

Really set Xcode editor to use 4 space indentation (http://webkit.org/coding/coding-style.html)

2837

2838

* JavaScriptCore.xcodeproj/project.pbxproj:

2839

2840

2007-03-19 Darin Adler <darin@apple.com>

2841

2842

Reviewed by Geoff.

2843

2844

- Changed list size threshold to 5 based on testing.

2845

2846

I was testing the i-Bench JavaScript with the list statistics

2847

dumping on, and discovered that there were many 5-element lists.

2848

The fast case for lists was for 4 elements and fewer. By changing

2849

the threshold to 5 elements we get a measurable speedup. I believe

2850

this will help real web pages too, not just the benchmark.

2851

2852

* kjs/list.cpp: Change constant from 4 to 5.

2853

2854

2007-03-19 Darin Adler <darin@apple.com>

2855

2856

* kjs/value.h: Oops, fix build.

2857

2858

2007-03-19 Darin Adler <darin@apple.com>

2859

2860

Reviewed by Geoff.

2861

2862

- remove ~JSValue; tiny low-risk performance boost

2863

2864

* kjs/value.h: Remove unneeded empty virtual destructor from JSValue.

2865

The only class derived from JSValue is JSCell and it already has a

2866

virtual destructor. Declaring an empty constructor in JSValue had one

2867

good effect: it marked the destructor private, making it a compile

2868

time error to try to destroy a JSValue; but that's not a likely

2869

mistake for someone to make. It had two bad effects: (1) it caused gcc,

2870

at least, to generate code to fix up the virtual table pointer to

2871

point to the JSValue version of the virtual table inside the destructor

2872

of all classes derived from JSValue directly or indirectly; (2) it

2873

caused JSValue to be a polymorphic class so required a virtual table for

2874

it. It's cleaner to not have either of those.

2875

2876

2007-03-18 Maciej Stachowiak <mjs@apple.com>

2877

2878

Reviewed by Mark.

2879

2880

- avoid static construction (and global variable access) in a smarter, more portable way,

2881

to later enable MUTLI_THREAD mode to work on other platforms and compilers.

2882

2883

* kjs/CommonIdentifiers.cpp: Added. New class to hold all the shared identifiers.

2884

(KJS::CommonIdentifiers::CommonIdentifiers):

2885

(KJS::CommonIdentifiers::shared):

2886

* kjs/CommonIdentifiers.h: Added.

2887

2888

* kjs/ExecState.h:

2889

(KJS::ExecState::propertyNames): Hand the CommonIdentifiers instance here for easy access.

2890

(KJS::ExecState::ExecState):

2891

2892

* API/JSObjectRef.cpp:

2893

(JSObjectMakeConstructor):

2894

* CMakeLists.txt:

2895

* JavaScriptCore.exp:

2896

* JavaScriptCore.pri:

2897

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

2898

* JavaScriptCore.xcodeproj/project.pbxproj:

2899

* JavaScriptCoreSources.bkl:

2900

* bindings/runtime_array.cpp:

2901

(RuntimeArray::getOwnPropertySlot):

2902

(RuntimeArray::put):

2903

* bindings/runtime_method.cpp:

2904

(RuntimeMethod::getOwnPropertySlot):

2905

* kjs/array_object.cpp:

2906

(ArrayInstance::getOwnPropertySlot):

2907

(ArrayInstance::put):

2908

(ArrayInstance::deleteProperty):

2909

(ArrayProtoFunc::ArrayProtoFunc):

2910

(ArrayProtoFunc::callAsFunction):

2911

(ArrayObjectImp::ArrayObjectImp):

2912

* kjs/bool_object.cpp:

2913

(BooleanPrototype::BooleanPrototype):

2914

(BooleanProtoFunc::BooleanProtoFunc):

2915

(BooleanProtoFunc::callAsFunction):

2916

(BooleanObjectImp::BooleanObjectImp):

2917

* kjs/completion.h:

2918

(KJS::Completion::Completion):

2919

* kjs/date_object.cpp:

2920

(KJS::DateProtoFunc::DateProtoFunc):

2921

(KJS::DateObjectImp::DateObjectImp):

2922

(KJS::DateObjectFuncImp::DateObjectFuncImp):

2923

* kjs/error_object.cpp:

2924

(ErrorPrototype::ErrorPrototype):

2925

(ErrorProtoFunc::ErrorProtoFunc):

2926

(ErrorProtoFunc::callAsFunction):

2927

(ErrorObjectImp::ErrorObjectImp):

2928

(ErrorObjectImp::construct):

2929

(NativeErrorPrototype::NativeErrorPrototype):

2930

(NativeErrorImp::NativeErrorImp):

2931

(NativeErrorImp::construct):

2932

(NativeErrorImp::callAsFunction):

2933

* kjs/function.cpp:

2934

(KJS::FunctionImp::getOwnPropertySlot):

2935

(KJS::FunctionImp::put):

2936

(KJS::FunctionImp::deleteProperty):

2937

(KJS::FunctionImp::getParameterName):

2938

(KJS::DeclaredFunctionImp::construct):

2939

(KJS::IndexToNameMap::unMap):

2940

(KJS::Arguments::Arguments):

2941

(KJS::ActivationImp::getOwnPropertySlot):

2942

(KJS::ActivationImp::deleteProperty):

2943

(KJS::GlobalFuncImp::GlobalFuncImp):

2944

* kjs/function_object.cpp:

2945

(FunctionPrototype::FunctionPrototype):

2946

(FunctionProtoFunc::FunctionProtoFunc):

2947

(FunctionProtoFunc::callAsFunction):

2948

(FunctionObjectImp::FunctionObjectImp):

2949

(FunctionObjectImp::construct):

2950

* kjs/grammar.y:

2951

* kjs/identifier.cpp:

2952

* kjs/identifier.h:

2953

* kjs/interpreter.cpp:

2954

(KJS::Interpreter::init):

2955

(KJS::Interpreter::initGlobalObject):

2956

* kjs/interpreter.h:

2957

* kjs/lookup.h:

2958

* kjs/math_object.cpp:

2959

(MathFuncImp::MathFuncImp):

2960

* kjs/nodes.cpp:

2961

(ArrayNode::evaluate):

2962

(FuncDeclNode::processFuncDecl):

2963

(FuncExprNode::evaluate):

2964

* kjs/number_object.cpp:

2965

(NumberPrototype::NumberPrototype):

2966

(NumberProtoFunc::NumberProtoFunc):

2967

(NumberObjectImp::NumberObjectImp):

2968

* kjs/object.cpp:

2969

(KJS::JSObject::put):

2970

(KJS::JSObject::defaultValue):

2971

(KJS::JSObject::hasInstance):

2972

* kjs/object.h:

2973

(KJS::JSObject::getOwnPropertySlot):

2974

* kjs/object_object.cpp:

2975

(ObjectPrototype::ObjectPrototype):

2976

(ObjectProtoFunc::ObjectProtoFunc):

2977

(ObjectObjectImp::ObjectObjectImp):

2978

* kjs/regexp_object.cpp:

2979

(RegExpPrototype::RegExpPrototype):

2980

(RegExpProtoFunc::RegExpProtoFunc):

2981

(RegExpObjectImp::RegExpObjectImp):

2982

* kjs/string_object.cpp:

2983

(KJS::StringInstance::getOwnPropertySlot):

2984

(KJS::StringInstance::put):

2985

(KJS::StringInstance::deleteProperty):

2986

(KJS::StringPrototype::StringPrototype):

2987

(KJS::StringProtoFunc::StringProtoFunc):

2988

(KJS::StringProtoFunc::callAsFunction):

2989

(KJS::StringObjectImp::StringObjectImp):

2990

(KJS::StringObjectFuncImp::StringObjectFuncImp):

2991

* kjs/testkjs.cpp:

2992

(TestFunctionImp::TestFunctionImp):

2993

2994

2007-03-18 Andrew Wellington <proton@wiretapped.net>

2995

2996

Reviewed by Mark Rowe

2997

2998

Set Xcode editor to use 4 space indentation (http://webkit.org/coding/coding-style.html)

2999

3000

* JavaScriptCore.xcodeproj/project.pbxproj:

3001

3002

2007-03-19 Mark Rowe <mrowe@apple.com>

3003

3004

Rubber-stamped by Brady.

3005

3006

Update references to bugzilla.opendarwin.org with bugs.webkit.org.

3007

3008

* bindings/c/c_utility.cpp:

3009

(KJS::Bindings::convertUTF8ToUTF16):

3010

* kjs/function.cpp:

3011

(KJS::FunctionImp::callAsFunction):

3012

* kjs/grammar.y:

3013

* kjs/keywords.table:

3014

* kjs/lexer.cpp:

3015

(KJS::Lexer::shift):

3016

3017

2007-03-18 Geoffrey Garen <ggaren@apple.com>

3018

3019

Reviewed by Oliver Hunt.

3020

3021

Exposed some extra toUInt32 functionality, as part of the fix for

3022

REGRESSION: Incomplete document.all implementation breaks abtelectronics.com

3023

(Style Change Through JavaScript Blanks Content)

3024

3025

* JavaScriptCore.exp:

3026

* kjs/identifier.h:

3027

(KJS::Identifier::toUInt32):

3028

3029

2007-03-18 Geoffrey Garen <ggaren@apple.com>

3030

3031

Removed duplicate export name.

3032

3033

* JavaScriptCore.exp:

3034

3035

2007-03-15 Geoffrey Garen <ggaren@apple.com>

3036

3037

Reviewed by Maciej Stachowiak.

3038

3039

Fixed <rdar://problem/5064964> Repro ASSERT failure in JS Bindings when

3040

closing window @ lowtrades.bptrade.com

3041

3042

Unfortunately, the bindings depend on UString and Identifier as string

3043

representations. So, they need to acquire the JSLock when doing something

3044

that will ref/deref their strings.

3045

3046

Layout tests, the original site, and Java, Flash, and Quicktime on the

3047

web work. No leaks reported. No automated test for this because testing

3048

the Java bindings, like math, is hard.

3049

3050

* bindings/runtime.h: Made Noncopyable, just to be sure.

3051

3052

* bindings/c/c_class.cpp:

3053

(KJS::Bindings::CClass::~CClass): Acquire the JSLock and explicitly clear the keys

3054

in our hashtable, since they're UString::Reps, and ref/deref aren't thread-safe.

3055

(KJS::Bindings::CClass::methodsNamed): Also acquire the JSLock when adding

3056

keys to the table, since the table ref's them.

3057

(KJS::Bindings::CClass::fieldNamed): ditto.

3058

3059

* bindings/c/c_utility.cpp: Removed dead function.

3060

(KJS::Bindings::convertValueToNPVariant): Acquire the JSLock because doing

3061

it recursively is pretty cheap, and it's just too confusing to tell whether

3062

all our callers do it for us.

3063

(KJS::Bindings::convertNPVariantToValue): ditto

3064

* bindings/c/c_utility.h:

3065

3066

* bindings/jni/jni_class.cpp: Same deal as c_class.cpp.

3067

(JavaClass::JavaClass):

3068

(JavaClass::~JavaClass):

3069

3070

* bindings/jni/jni_instance.cpp: Same deal as c_utility.cpp.

3071

(JavaInstance::stringValue):

3072

* bindings/jni/jni_jsobject.cpp:

3073

(JavaJSObject::convertValueToJObject):

3074

3075

* bindings/jni/jni_runtime.cpp:

3076

(JavaMethod::~JavaMethod): Moved from header, for clarity.

3077

(appendClassName): Made this static, so the set of callers is known, and

3078

we can assert that we hold the JSLock. Also changed it to take a UString

3079

reference, which makes the calling code simpler.

3080

(JavaMethod::signature): Store the ASCII value we care about instead of

3081

a UString, since UString is so much more hassle. Hold the JSLock while

3082

building up the temporary UString.

3083

3084

* bindings/jni/jni_runtime.h: Nixed dead code in JavaMethod.

3085

(KJS::Bindings::JavaString::JavaString): Hold a UString::Rep instead of

3086

a UString, so we can acquire the JSLock and explicitly release it.

3087

(KJS::Bindings::JavaString::_commonInit):

3088

(KJS::Bindings::JavaString::~JavaString):

3089

(KJS::Bindings::JavaString::UTF8String):

3090

(KJS::Bindings::JavaString::uchars):

3091

(KJS::Bindings::JavaString::length):

3092

(KJS::Bindings::JavaString::ustring):

3093

3094

* bindings/jni/jni_utility.cpp:

3095

(KJS::Bindings::convertArrayInstanceToJavaArray): Made this static, so

3096

the set of callers is known, and we can assert that we hold the JSLock.

3097

(KJS::Bindings::convertValueToJValue): Acquire the JSLock because doing

3098

it recursively is pretty cheap, and it's just too confusing to tell whether

3099

all our callers do it for us.

3100

3101

* bindings/objc/objc_runtime.h: Nixed some dead code.

3102

* bindings/objc/objc_utility.mm:

3103

(KJS::Bindings::convertNSStringToString): Same drill as above.

3104

3105

2007-03-18 Alexey Proskuryakov <ap@webkit.org>

3106

3107

Reviewed by Geoff.

3108

3109

http://bugs.webkit.org/show_bug.cgi?id=13105

3110

REGRESSION: an exception raised when calculating base value of a dot expression is not returned

3111

3112

Test: fast/js/dot-node-base-exception.html

3113

3114

* kjs/nodes.cpp:

3115

(FunctionCallDotNode::evaluate): Added the necessary KJS_CHECKEXCEPTIONVALUE.

3116

3117

2007-03-18 Steve Falkenburg <sfalken@apple.com>

3118

3119

Build fix.

3120

3121

* JavaScriptCore.vcproj/testkjs/testkjs.vcproj:

3122

3123

2007-03-17 Timothy Hatcher <timothy@apple.com>

3124

3125

Reviewed by Mark Rowe.

3126

3127

Made Version.xcconfig smarter when building for different configurations.

3128

Now uses the 522+ OpenSource version for Debug and Release, while using the

3129

full 522.4 version for Production builds. The system prefix is also computed

3130

based on the current system, so 4522.4 on Tiger and 5522.4 on Leopard.

3131

3132

* Configurations/JavaScriptCore.xcconfig:

3133

* Configurations/Version.xcconfig:

3134

3135

2007-03-15 Maciej Stachowiak <mjs@apple.com>

3136

3137

Not reviewed.

3138

3139

- build fix

3140

3141

* wtf/TCSystemAlloc.cpp:

3142

3143

2007-03-15 Maciej Stachowiak <mjs@apple.com>

3144

3145

Reviewed by Geoff and Steve.

3146

3147

- fix some portability issues with TCMalloc.

3148

3149

* JavaScriptCore.vcproj/WTF/WTF.vcproj:

3150

* kjs/config.h:

3151

* wtf/FastMalloc.cpp:

3152

(WTF::SizeClass):

3153

(WTF::InitSizeClasses):

3154

(WTF::TCMalloc_PageHeap::Split):

3155

(WTF::TCMalloc_PageHeap::RegisterSizeClass):

3156

(WTF::TCMalloc_Central_FreeList::length):

3157

(WTF::TCMalloc_ThreadCache::InitTSD):

3158

(WTF::TCMalloc_ThreadCache::CreateCacheIfNecessary):

3159

* wtf/TCSpinLock.h:

3160

* wtf/TCSystemAlloc.cpp:

3161

(TryVirtualAlloc):

3162

(TCMalloc_SystemAlloc):

3163

3164

2007-03-15 Timothy Hatcher <timothy@apple.com>

3165

3166

Reviewed by John.

3167

3168

* Factored out most of our common build settings into .xcconfig files. Anything that was common in

3169

each build configuration was factored out into the shared .xcconfig file.

3170

* Adds a Version.xcconfig file to define the current framework version, to be used in other places.

3171

* Use the new $(BUNDLE_VERSION) (defined in Version.xcconfig) in the preprocessed Info.plist.

3172

* Use the versions defined in Version.xcconfig to set $(DYLIB_CURRENT_VERSION).

3173

3174

* Configurations/Base.xcconfig: Added.

3175

* Configurations/DebugRelease.xcconfig: Added.

3176

* Configurations/JavaScriptCore.xcconfig: Added.

3177

* Configurations/Version.xcconfig: Added.

3178

* Info.plist:

3179

* JavaScriptCore.xcodeproj/project.pbxproj:

3180

3181

2007-03-16 Shrikant Gangoda <shrikant.gangoda@celunite.com>

3182

3183

Gdk build fix.

3184

3185

* kjs/DateMath.cpp: gettimeofday comes from <sys/time.h> on Linux.

3186

3187

2007-03-14 Kevin McCullough <kmccullough@apple.com>

3188

3189

Reviewed by .

3190

3191

- Fixed one more build breakage

3192

3193

* kjs/date_object.cpp:

3194

(KJS::formatLocaleDate):

3195

3196

2007-03-14 Kevin McCullough <kmccullough@apple.com>

3197

3198

Reviewed by .

3199

3200

- Fixed a build breakage.

3201

3202

* kjs/DateMath.cpp:

3203

* kjs/date_object.cpp:

3204

(KJS::formatLocaleDate):

3205

(KJS::DateObjectImp::construct):

3206

3207

2007-03-14 Kevin McCullough <kmccullough@apple.com>

3208

3209

Reviewed by Geoff.

3210

3211

- rdar://problem/5045720

3212

- DST changes in US affect JavaScript date calculations (12975)

3213

This fix was to ensure we properly test for the new changes to DST in the US.

3214

Also this fixes when we apply DST, now we correctly map most past years to current

3215

DST rules. We still have a small issue with years before 1900 or after 2100.

3216

rdar://problem/5055038

3217

3218

* kjs/DateMath.cpp: Fix DST to match spec better.

3219

(KJS::getCurrentUTCTime):

3220

(KJS::mimimumYearForDST):

3221

(KJS::maximumYearForDST):

3222

(KJS::equivalentYearForDST):

3223

(KJS::getDSTOffset):

3224

* kjs/DateMath.h: Consolodated common funtionality.

3225

* kjs/date_object.cpp: Consolodated common functionality.

3226

(KJS::formatLocaleDate):

3227

(KJS::DateObjectImp::construct):

3228

* tests/mozilla/ecma/jsref.js: Added functions for finding the correct days when DST starts and ends.

3229

* tests/mozilla/ecma/shell.js: Added back in the old DST functions for ease of merging with mozilla if needed.

3230

* tests/mozilla/ecma_2/jsref.js: Added functions for finding the correct days when DST starts and ends.

3231

* tests/mozilla/ecma_3/Date/shell.js: Added functions for finding the correct days when DST starts and ends.

3232

* tests/mozilla/expected.html: Updated to show all date tests passing.

3233

3234

=== Safari-5522.4 ===

3235

3236

2007-03-13 Kevin McCullough <kmccullough@apple.com>

3237

3238

Reviewed by .

3239

3240

- Adding expected failures until the are truly fixed.

3241

- rdar://problem/5060302

3242

3243

* tests/mozilla/expected.html:

3244

3245

2007-03-12 Kevin McCullough <kmccullough@apple.com>

3246

3247

Reviewed by .

3248

3249

- Actually update tests for new DST rules.

3250

3251

* tests/mozilla/ecma/Date/15.9.3.1-1.js:

3252

* tests/mozilla/ecma/Date/15.9.3.1-2.js:

3253

* tests/mozilla/ecma/Date/15.9.3.1-3.js:

3254

* tests/mozilla/ecma/Date/15.9.3.1-4.js:

3255

* tests/mozilla/ecma/Date/15.9.3.1-5.js:

3256

* tests/mozilla/ecma/Date/15.9.3.2-1.js:

3257

* tests/mozilla/ecma/Date/15.9.3.2-2.js:

3258

* tests/mozilla/ecma/Date/15.9.3.2-3.js:

3259

* tests/mozilla/ecma/Date/15.9.3.2-4.js:

3260

* tests/mozilla/ecma/Date/15.9.3.2-5.js:

3261

* tests/mozilla/ecma/Date/15.9.3.8-1.js:

3262

* tests/mozilla/ecma/Date/15.9.3.8-2.js:

3263

* tests/mozilla/ecma/Date/15.9.3.8-3.js:

3264

* tests/mozilla/ecma/Date/15.9.3.8-4.js:

3265

* tests/mozilla/ecma/Date/15.9.3.8-5.js:

3266

* tests/mozilla/ecma/Date/15.9.5.10-1.js:

3267

* tests/mozilla/ecma/Date/15.9.5.10-10.js:

3268

* tests/mozilla/ecma/Date/15.9.5.10-11.js:

3269

* tests/mozilla/ecma/Date/15.9.5.10-12.js:

3270

* tests/mozilla/ecma/Date/15.9.5.10-13.js:

3271

* tests/mozilla/ecma/Date/15.9.5.10-2.js:

3272

* tests/mozilla/ecma/Date/15.9.5.10-3.js:

3273

* tests/mozilla/ecma/Date/15.9.5.10-4.js:

3274

* tests/mozilla/ecma/Date/15.9.5.10-5.js:

3275

* tests/mozilla/ecma/Date/15.9.5.10-6.js:

3276

* tests/mozilla/ecma/Date/15.9.5.10-7.js:

3277

* tests/mozilla/ecma/Date/15.9.5.10-8.js:

3278

* tests/mozilla/ecma/Date/15.9.5.10-9.js:

3279

* tests/mozilla/ecma/jsref.js:

3280

* tests/mozilla/ecma_2/jsref.js:

3281

* tests/mozilla/ecma_3/Date/shell.js:

3282

3283

2007-03-12 Kevin McCullough <kmccullough@apple.com>

3284

3285

Reviewed by .

3286

3287

- Update tests for new DST rules.

3288

3289

* tests/mozilla/ecma/shell.js:

3290

3291

2007-03-11 Geoffrey Garen <ggaren@apple.com>

3292

3293

Reviewed by Oliver Hunt.

3294

3295

Fixed <rdar://problem/4681051> Installer crashes in KJS::Collector::

3296

markOtherThreadConservatively(KJS::Collector::Thread*) trying to install

3297

iLife 06 using Rosetta on an Intel Machine

3298

3299

The problem was that our thread-specific data destructor would modify the

3300

list of active JavaScript threads without holding the JSLock, corrupting

3301

the list. Corruption was especially likely if one JavaScript thread exited

3302

while another was starting up.

3303

3304

* JavaScriptCore.exp:

3305

* kjs/JSLock.cpp: Don't conflate locking the JSLock with registering a

3306

thread, since the thread-specific data destructor needs to lock

3307

without registering a thread. Instead, treat thread registration as a

3308

part of the convenience of the JSLock object, and whittle down JSLock::lock()

3309

to just the bits that actually do the locking.

3310

(KJS::JSLock::lock):

3311

(KJS::JSLock::registerThread):

3312

* kjs/JSLock.h: Updated comments to mention the new behavior above, and

3313

other recent changes.

3314

(KJS::JSLock::JSLock):

3315

* kjs/collector.cpp:

3316

(KJS::destroyRegisteredThread): Lock here.

3317

(KJS::Collector::registerThread): To match, assert that we're locked here.

3318

3319

2007-03-10 Geoffrey Garen <ggaren@apple.com>

3320

3321

Reviewed by Darin Adler.

3322

3323

Fixed <rdar://problem/4587763> PAC file: lock inversion between QT and

3324

JSCore causes a hang @ www.panoramas.dk

3325

3326

With a PAC file, run-webkit-tests --threaded passes, the reported site

3327

works, and all the Quicktime/JavaScript and Flash/JavaScript examples

3328

I found through Google work, too.

3329

3330

Any time JavaScript causes arbitrary non-JavaScript code to execute, it

3331

risks deadlock, because that code may block, trying to acquire a lock

3332

owned by a thread that is waiting to execute JavaScript. In this case,

3333

the thread was a networking thread that was waiting to interpret a PAC file.

3334

3335

Because non-JavaScript code may execute in response to, well, anything,

3336

a perfect solution to this problem is impossible. I've implemented an

3337

optimistic solution, instead: JavaScript will drop its lock whenever it

3338

makes a direct call to non-JavaScript code through a bridging/plug-in API,

3339

but will blissfully ignore the indirect ways it may cause non-JavaScript

3340

code to run (resizing a window, for example).

3341

3342

Unfortunately, this solution introduces significant locking overhead in

3343

the bridging APIs. I don't see a way around that.

3344

3345

This patch includes some distinct bug fixes I saw along the way:

3346

3347

* bindings/objc/objc_instance.mm: Fixed a bug where a nested begin() call

3348

would leak its autorelease pool, because it would NULL out _pool without

3349

draining it.

3350

3351

* bindings/runtime_object.cpp:

3352

(RuntimeObjectImp::methodGetter): Don't copy an Identifier to ASCII only

3353

to turn around and make an Identifier from the ASCII. In an earlier

3354

version of this patch, the copy caused an assertion failure. Now it's

3355

just unnecessary work.

3356

(RuntimeObjectImp::getOwnPropertySlot): ditto

3357

3358

* bindings/objc/objc_instance.h: Removed overrides of setVAlueOfField and

3359

getValueOfField, because they did exactly what the base class versions did.

3360

Removed overrides of Noncopyable declarations for the same reason.

3361

3362

* bindings/runtime.h: Inherit from Noncopyable instead of rolling our own.

3363

* bindings/c/c_instance.h: ditto

3364

3365

And the actual patch:

3366

3367

* API/JSCallbackConstructor.cpp: Drop all locks when calling out to C.

3368

(KJS::JSCallbackConstructor::construct):

3369

* API/JSCallbackFunction.cpp: ditto

3370

(KJS::JSCallbackFunction::callAsFunction):

3371

* API/JSCallbackObject.cpp: ditto

3372

(KJS::JSCallbackObject::init):

3373

(KJS::JSCallbackObject::~JSCallbackObject):

3374

(KJS::JSCallbackObject::getOwnPropertySlot):

3375

(KJS::JSCallbackObject::put):

3376

(KJS::JSCallbackObject::deleteProperty):

3377

(KJS::JSCallbackObject::construct):

3378

(KJS::JSCallbackObject::hasInstance):

3379

(KJS::JSCallbackObject::callAsFunction):

3380

(KJS::JSCallbackObject::getPropertyNames):

3381

(KJS::JSCallbackObject::toNumber):

3382

(KJS::JSCallbackObject::toString):

3383

(KJS::JSCallbackObject::staticValueGetter):

3384

(KJS::JSCallbackObject::callbackGetter):

3385

3386

* bindings/c/c_instance.cpp: Drop all locks when calling out to C.

3387

(KJS::Bindings::CInstance::invokeMethod):

3388

(KJS::Bindings::CInstance::invokeDefaultMethod):

3389

* bindings/c/c_runtime.cpp: Drop all locks when calling out to C.

3390

(KJS::Bindings::CField::valueFromInstance):

3391

(KJS::Bindings::CField::setValueToInstance):

3392

* bindings/jni/jni_objc.mm:

3393

(KJS::Bindings::dispatchJNICall): Drop all locks when calling out to Java.

3394

3395

* bindings/objc/objc_instance.mm: The changes here are to accomodate the

3396

fact that C++ unwinding of DropAllLocks goes crazy when you put it inside

3397

a @try block. I moved all JavaScript stuff outside of the @try blocks, and

3398

then prefixed the whole blocks with DropAllLocks objects. This required some

3399

supporting changes in other functions, which now acquire the JSLock for

3400

themselves, intead of relying on their callers to do so.

3401

(ObjcInstance::end):

3402

(ObjcInstance::invokeMethod):

3403

(ObjcInstance::invokeDefaultMethod):

3404

(ObjcInstance::setValueOfUndefinedField):

3405

(ObjcInstance::getValueOfUndefinedField):

3406

* bindings/objc/objc_runtime.mm: Same as above, except I didn't want to

3407

change throwError to acquire the JSLock for itself.

3408

(ObjcField::valueFromInstance):

3409

(ObjcField::setValueToInstance):

3410

* bindings/objc/objc_utility.mm: Supporting changes mentioned above.

3411

(KJS::Bindings::convertValueToObjcValue):

3412

(KJS::Bindings::convertObjcValueToValue):

3413

3414

* kjs/JSLock.cpp:

3415

(1) Fixed DropAllLocks to behave as advertised, and drop the JSLock only

3416

if the current thread actually acquired it in the first place. This is

3417

important because WebKit needs to ensure that the JSLock has been

3418

dropped before it makes a plug-in call, even though it doesn't know if

3419

the current thread actually acquired the JSLock. (We don't want WebKit

3420

to accidentally drop a lock belonging to *another thread*.)

3421

(2) Used the new per-thread code written for (1) to make recursive calls

3422

to JSLock very cheap. JSLock now knows to call pthread_mutext_lock/

3423

pthread_mutext_unlock only at nesting level 0.

3424

(KJS::createDidLockJSMutex):

3425

(KJS::JSLock::lock):

3426

(KJS::JSLock::unlock):

3427

(KJS::DropAllLocks::DropAllLocks):

3428

(KJS::DropAllLocks::~DropAllLocks):

3429

(KJS::JSLock::lockCount):

3430

* kjs/JSLock.h: Don't duplicate Noncopyable.

3431

(KJS::JSLock::~JSLock):

3432

3433

* wtf/Assertions.h: Blind attempt at helping the Windows build.

3434

3435

2007-03-08 MorganL <morganl.webkit@yahoo.com>

3436

3437

Reviewed by Darin.

3438

3439

http://bugs.webkit.org/show_bug.cgi?id=13018

3440

Bug 13018: allow embedders to override the definition of CRASH.

3441

3442

* wtf/Assertions.h: make it possible to override CRASH.

3443

3444

2007-03-07 Anrong Hu <huanr@yahoo.com>

3445

3446

Reviewed by Maciej.

3447

3448

Fix http://bugs.webkit.org/show_bug.cgi?id=12535

3449

Bug 12535: Stack-optimizing compilers can trick GC into freeing in-use objects

3450

3451

* kjs/internal.cpp:

3452

(KJS::StringImp::toObject): Copy val onto the stack so it is not subject to garbage collection.

3453

3454

2007-03-07 Geoffrey Garen <ggaren@apple.com>

3455

3456

Build fix for non-multiple-thread folks.

3457

3458

Use a shared global in the non-multiple-thread case.

3459

3460

* wtf/FastMalloc.cpp:

3461

(WTF::isForbidden):

3462

(WTF::fastMallocForbid):

3463

(WTF::fastMallocAllow):

3464

3465

2007-03-07 Geoffrey Garen <ggaren@apple.com>

3466

3467

Reviewed by Darin Adler.

3468

3469

Fixed ASSERT failure I just introduced.

3470

3471

Made the fastMalloc isForbidden flag per thread. (Oops!) We expect that

3472

other threads will malloc while we're marking -- we just want to prevent

3473

our own marking from malloc'ing.

3474

3475

* wtf/FastMalloc.cpp:

3476

(WTF::initializeIsForbiddenKey):

3477

(WTF::isForbidden):

3478

(WTF::fastMallocForbid):

3479

(WTF::fastMallocAllow):

3480

(WTF::fastMalloc):

3481

(WTF::fastCalloc):

3482

(WTF::fastFree):

3483

(WTF::fastRealloc):

3484

(WTF::do_malloc):

3485

3486

2007-03-07 Shrikant Gangoda <shrikant.gangoda@celunite.com>

3487

3488

Reviewed by Maciej.

3489

3490

http://bugs.webkit.org/show_bug.cgi?id=12997

3491

3492

Wrap pthread-specific assertion in #if USE(MULTIPLE_THREADS).

3493

3494

* kjs/collector.cpp:

3495

(KJS::Collector::markMainThreadOnlyObjects):

3496

3497

2007-03-06 Geoffrey Garen <ggaren@apple.com>

3498

3499

Reviewed by Maciej Stachowiak.

3500

3501

Fixed <rdar://problem/4576242> | http://bugs.webkit.org/show_bug.cgi?id=12586

3502

PAC file: malloc deadlock sometimes causes a hang @ www.apple.com/pro/profiles/ (12586)

3503

3504

This is a modified version of r14752 on the branch.

3505

3506

These changes just add debugging functionality. They ASSERT that we don't

3507

malloc during the mark phase of a garbage collection, which can cause a

3508

deadlock.

3509

3510

* kjs/collector.cpp:

3511

(KJS::Collector::collect):

3512

* wtf/FastMalloc.cpp:

3513

(WTF::fastMallocForbid):

3514

(WTF::fastMallocAllow):

3515

(WTF::fastMalloc):

3516

(WTF::fastCalloc):

3517

(WTF::fastFree):

3518

(WTF::fastRealloc):

3519

(WTF::do_malloc):

3520

* wtf/FastMalloc.h:

3521

3522

2007-03-06 Geoffrey Garen <ggaren@apple.com>

3523

3524

Reviewed by Maciej Stachowiak.

3525

3526

Fixed all known crashers exposed by run-webkit-tests --threaded. This covers:

3527

3528

<rdar://problem/4565394> | http://bugs.webkit.org/show_bug.cgi?id=12585

3529

PAC file: after closing a window that contains macworld.com, new window

3530

crashes (KJS::PropertyMap::mark()) (12585)

3531

<rdar://problem/4571215> | http://bugs.webkit.org/show_bug.cgi?id=9211

3532

PAC file: Crash occurs when clicking on the navigation tabs at http://www.businessweek.com/ (9211)

3533

<rdar://problem/4557926>

3534

PAC file: Crash occurs when attempting to view image in slideshow mode

3535

at http://d.smugmug.com/gallery/581716 ( KJS::IfNode::execute (KJS::

3536

ExecState*) + 312) if you use a PAC file

3537

3538

(1) Added some missing JSLocks, along with related ASSERTs.

3539

3540

(2) Fully implemented support for objects that can only be garbage collected

3541

on the main thread. So far, only WebCore uses this. We can add it to API

3542

later if we learn that it's needed.

3543

3544

The implementation uses a "main thread only" flag inside each object. When

3545

collecting on a secondary thread, the Collector does an extra pass through

3546

the heap to mark all flagged objects before sweeping. This solution makes

3547

the common case -- flag lots of objects, but never collect on a secondary

3548

thread -- very fast, even though the uncommon case of garbage collecting

3549

on a secondary thread isn't as fast as it could be. I left some notes

3550

about how to speed it up, if we ever care.

3551

3552

For posterity, here are some things I learned about GC while investigating:

3553

3554

* Each collect must either mark or delete every heap object. "Zombie"

3555

objects, which are neither marked nor deleted, raise these issues:

3556

3557

* On the next pass, the conservative marking algorithm might mark a

3558

zombie, causing it to mark freed objects.

3559

3560

* The client might try to use a zombie, which would seem live because

3561

its finalizer had not yet run.

3562

3563

* A collect on the main thread is free to delete any object. Presumably,

3564

objects allocated on secondary threads have thread-safe finalizers.

3565

3566

* A collect on a secondary thread must not delete thread-unsafe objects.

3567

3568

* The mark function must be thread-safe.

3569

3570

Line by line comments:

3571

3572

* API/JSObjectRef.h: Added comment specifying that the finalize callback

3573

may run on any thread.

3574

3575

* JavaScriptCore.exp: Nothing to see here.

3576

3577

* bindings/npruntime.cpp:

3578

(_NPN_GetStringIdentifier): Added JSLock.

3579

3580

* bindings/objc/objc_instance.h:

3581

* bindings/objc/objc_instance.mm:

3582

(ObjcInstance::~ObjcInstance): Use an autorelease pool. The other callers

3583

to CFRelease needed one, too, but they were dead code, so I removed them

3584

instead. (This fixes a leak seen while running run-webkit-tests --threaded,

3585

although I don't think it's specifically a threading issue.)

3586

3587

* kjs/collector.cpp:

3588

(KJS::Collector::collectOnMainThreadOnly): New function. Tells the collector

3589

to collect a value only if it's collecting on the main thread.

3590

(KJS::Collector::markMainThreadOnlyObjects): New function. Scans the heap

3591

for "main thread only" objects and marks them.

3592

3593

* kjs/date_object.cpp:

3594

(KJS::DateObjectImp::DateObjectImp): To make the new ASSERTs happy, allocate

3595

our globals on the heap, avoiding a seemingly unsafe destructor call at

3596

program exit time.

3597

* kjs/function_object.cpp:

3598

(FunctionPrototype::FunctionPrototype): ditto

3599

3600

* kjs/interpreter.cpp:

3601

(KJS::Interpreter::mark): Removed boolean parameter, which was an incomplete

3602

and arguably hackish way to implement markMainThreadOnlyObjects() inside WebCore.

3603

* kjs/interpreter.h:

3604

3605

* kjs/identifier.cpp:

3606

(KJS::identifierTable): Added some ASSERTs to check for thread safety

3607

problems.

3608

3609

* kjs/list.cpp: Added some ASSERTs to check for thread safety problems.

3610

(KJS::allocateListImp):

3611

(KJS::List::release):

3612

(KJS::List::append):

3613

(KJS::List::empty): Make the new ASSERTs happy.

3614

3615

* kjs/object.h:

3616

(KJS::JSObject::JSObject): "m_destructorIsThreadSafe" => "m_collectOnMainThreadOnly".

3617

I removed the constructor parameter because m_collectOnMainThreadOnly,

3618

like m_marked, is a Collector bit, so only the Collector should set or get it.

3619

3620

* kjs/object_object.cpp:

3621

(ObjectPrototype::ObjectPrototype): Make the ASSERTs happy.

3622

* kjs/regexp_object.cpp:

3623

(RegExpPrototype::RegExpPrototype): ditto

3624

3625

* kjs/ustring.cpp: Added some ASSERTs to check for thread safety problems.

3626

(KJS::UCharReference::ref):

3627

(KJS::UString::Rep::createCopying):

3628

(KJS::UString::Rep::create):

3629

(KJS::UString::Rep::destroy):

3630

(KJS::UString::null): Make the new ASSERTs happy.

3631

* kjs/ustring.h:

3632

(KJS::UString::Rep::ref): Added some ASSERTs to check for thread safety problems.

3633

(KJS::UString::Rep::deref):

3634

3635

* kjs/value.h:

3636

(KJS::JSCell::JSCell):

3637

3638

2007-03-06 Geoffrey Garen <ggaren@apple.com>

3639

3640

Reviewed by Maciej Stachowiak.

3641

3642

2% speedup on super accurate JS iBench.

3643

3644

(KJS::Collector::collect): Removed anti-optimization to call

3645

pthread_is_threaded_np() before calling pthread_main_np(). Almost all

3646

apps have more than one thread, so the extra call is actually worse.

3647

Interestingly, even the single-threaded testkjs shows a speed gain

3648

from removing the pthread_is_threaded_np() short-circuit. Not sure why.

3649

3650

2007-03-04 Don Gibson <dgibson77@gmail.com>

3651

3652

Reviewed by Nikolas Zimmermann.

3653

3654

- fix http://bugs.webkit.org/show_bug.cgi?id=12950

3655

Assertions.cpp should not #define macros that are already defined

3656

3657

* wtf/Assertions.cpp: Don't #define WINVER and _WIN32_WINNT if they

3658

are already defined.

3659

3660

2007-03-02 Steve Falkenburg <sfalken@apple.com>

3661

3662

Reviewed by Anders.

3663

3664

Add unsigned int hash traits (matches existing unsigned long version)

3665

3666

* wtf/HashTraits.h:

3667

(WTF::):

3668

3669

2007-03-02 Adam Roben <aroben@apple.com>

3670

3671

Reviewed by Kevin M.

3672

3673

Try to fix the Qt build.

3674

3675

* kjs/DateMath.cpp:

3676

(KJS::msToGregorianDateTime): Removed unnecessary "struct" keyword.

3677

* kjs/DateMath.h: Moved forward declarations to the top of the file

3678

before they are used.

3679

* kjs/date_object.cpp:

3680

(KJS::formatLocaleDate): Changed to take a const GregorianDateTime&

3681

since GregorianDateTime is Noncopyable.

3682

3683

2007-03-02 Darin Adler <darin@apple.com>

3684

3685

Reviewed by Kevin McCullough.

3686

3687

- fix http://bugs.webkit.org/show_bug.cgi?id=12867

3688

REGRESSION: BenchJS test 7 (dates) is 220% slower than in Safari 2.0.4

3689

3690

* kjs/DateMath.h: Marked GregorianDateTime as noncopyable, since it has a non-trivial

3691

destructor and not the correspoding copy constructor or assignment operator.

3692

Changed the GregorianDateTime constructor to use member initialization syntax.

3693

Fixed the destructor to use the array delete operator, since timeZone is an array.

3694

3695

* kjs/DateMath.cpp:

3696

(KJS::daysInYear): Changed to call isLeapYear so the rule is not repeated twice.

3697

(KJS::getUTCOffset): Added caching on PLATFORM(DARWIN), since we can rely on the

3698

notify_check function and "com.apple.system.timezone" to let us know when the

3699

offset has changed.

3700

3701

2007-02-27 Geoffrey Garen <ggaren@apple.com>

3702

3703

Reviewed by Darin Adler.

3704

3705

Follow-up to fixing http://bugs.webkit.org/show_bug.cgi?id=12659 | <rdar://problem/4954306>

3706

JS objects not collected after closing window @ ebay.com/maps.google.com

3707

3708

Changed Interpreter cache of global constructors and prototypes from

3709

ProtectedPtrs to bare, marked pointers. ProtectedPtrs are inefficient,

3710

and they increase the risk of reference cycles. Also, Darin said something

3711

about ProtectedPtrs giving him warts.

3712

3713

Also changed data members to precise types from generic JSObject*'s.

3714

3715

Layout tests and JS tests pass.

3716

3717

* kjs/SavedBuiltins.h:

3718

* kjs/interpreter.cpp:

3719

(KJS::Interpreter::init):

3720

(KJS::Interpreter::~Interpreter):

3721

(KJS::Interpreter::initGlobalObject): Moved Identifier::init() call to

3722

constructor, for clarity.

3723

(KJS::Interpreter::mark):

3724

* kjs/interpreter.h:

3725

3726

2007-02-27 Geoffrey Garen <ggaren@apple.com>

3727

3728

Reviewed by Maciej Stachowiak.

3729

3730

Fixed http://bugs.webkit.org/show_bug.cgi?id=12659 | <rdar://problem/4954306>

3731

JS objects not collected after closing window @ ebay.com/maps.google.com

3732

3733

Don't GC in the Interpreter destructor. For that to work, the Interpreter

3734

would have to NULL out all of its ProtectedPtrs before calling collect(). But

3735

we've decided that we don't want things to work that way, anyway. We want the

3736

client to be in charge of manual GC so that it can optimize cases when

3737

it will be destroying many interpreters at once

3738

(e.g., http://bugs.webkit.org/show_bug.cgi?id=12900).

3739

3740

Also removed Interpreter::collect() because it was redundant with

3741

Collector::collect().

3742

3743

* JavaScriptCore.exp:

3744

* kjs/interpreter.cpp:

3745

(KJS::Interpreter::~Interpreter):

3746

* kjs/testkjs.cpp:

3747

(TestFunctionImp::callAsFunction):

3748

3749

2007-02-26 Krzysztof Kowalczyk <kkowalczyk@gmail.com>

3750

3751

Reviewed by Adam Roben.

3752

3753

Rename *_SUPPORT defines to ENABLE_*.

3754

3755

* jscore.bkl:

3756

3757

2007-02-26 Maciej Stachowiak <mjs@apple.com>

3758

3759

Reviewed by Lars.

3760

3761

- <rdar://problem/5021698> Disable experimental SVG features (12883)

3762

3763

* wtf/Platform.h: Add ENABLE() macro similar to HAVE() and USE(), to

3764

allow nicer handling of optional WebKit features.

3765

3766

2007-02-22 George Staikos <staikos@kde.org>

3767

3768

Reviewed by Lars.

3769

3770

Add return values

3771

3772

* wtf/unicode/qt4/UnicodeQt4.h:

3773

(WTF::Unicode::toLower):

3774

(WTF::Unicode::toUpper):

3775

3776

2007-02-22 Oscar Cwajbaum <public@oscarc.net>

3777

3778

Reviewed by Maciej.

3779

3780

Fix ARM-specific alignment problem in FastMalloc

3781

http://bugs.webkit.org/show_bug.cgi?id=12841

3782

3783

* wtf/FastMalloc.cpp:

3784

Modify how pageheap_memory is declared to ensure proper alignment

3785

on architectures such as ARM

3786

3787

2007-02-20 Zack Rusin <zrusin@trolltech.com>

3788

3789

Reviewed by Lars

3790

3791

Make sure that non-void methods always return something.

3792

3793

* wtf/unicode/qt4/UnicodeQt4.h:

3794

(WTF::Unicode::toLower):

3795

(WTF::Unicode::toUpper):

3796

(WTF::Unicode::foldCase):

3797

3798

2007-02-18 Kevin Ollivier <kevino@theolliviers.com>

3799

3800

Reviewed by Adam Roben.

3801

3802

Fix cases where MSVC-specific code was identified as Win32 platform

3803

code. (as it should be compiled for e.g. wx port when using MSVC too)

3804

3805

* wtf/Assertions.h:

3806

* wtf/MathExtras.h:

3807

* wtf/StringExtras.h:

3808

changed PLATFORM(WIN) sections to COMPILER(MSVC) as necessary

3809

3810

2007-02-17 Krzysztof Kowalczyk <kkowalczyk@gmail.com>

3811

3812

Reviewed by Adam Roben.

3813

3814

Fix crashes on ARM due to different struct packing. Based on a patch

3815

by Mike Emmel.

3816

* kjs/ustring.cpp: compile-time assert to make sure sizeof(UChar) == 2

3817

* kjs/ustring.h: pack UChar struct to ensure that sizeof(UChar) == 2

3818

* wtf/Assertions.h: add COMPILE_ASSERT macro for compile-time assertions

3819

3820

2007-02-16 George Staikos <staikos@kde.org>

3821

3822

Reviewed by Maciej.

3823

3824

Fix uninitialized variable

3825

3826

* bindings/testbindings.cpp:

3827

(myAllocate):

3828

3829

2007-02-16 Anders Carlsson <acarlsson@apple.com>

3830

3831

Reviewed by Mitz.

3832

3833

http://bugs.webkit.org/show_bug.cgi?id=12788

3834

REGRESSION: Going back one page in history has a noticeable delay

3835

3836

Um...if all elements in two vectors are equal, then I guess we could say that

3837

the two vectors are equal too.

3838

3839

* wtf/Vector.h:

3840

(WTF::):

3841

3842

2007-02-14 Anders Carlsson <acarlsson@apple.com>

3843

3844

Reviewed by Darin.

3845

3846

Add new canCompareWithMemcmp vector trait and use it to determine whether

3847

operator== can use memcmp.

3848

3849

* wtf/Vector.h:

3850

(WTF::):

3851

(WTF::VectorTypeOperations::compare):

3852

(WTF::operator==):

3853

* wtf/VectorTraits.h:

3854

(WTF::):

3855

3856

2007-02-13 Brady Eidson <beidson@apple.com>

3857

3858

Reviewed by Darin

3859

3860

Tweaked vector a bit

3861

3862

* wtf/Vector.h:

3863

(WTF::operator==):

3864

3865

2007-02-13 Dex Deacon <occupant4@gmail.com>

3866

3867

Reviewed by Darin.

3868

3869

- fix for http://bugs.webkit.org/show_bug.cgi?id=12750

3870

Vector operator== was not defined correctly. It returned void,

3871

did not accept const Vectors, and used an int instead of size_t.

3872

3873

* wtf/Vector.h: fixed comparison operators

3874

(WTF::operator==):

3875

(WTF::operator!=):

3876

3877

2007-02-10 David Carson <dacarson@gmail.com>

3878

3879

Reviewed by Maciej.

3880

3881

- fix for http://bugs.webkit.org/show_bug.cgi?id=12636

3882

Corrected the generation of method signatures when the parameter

3883

is an Array.

3884

Added support for converting a Javascript array to a Java array.

3885

3886

* bindings/jni/jni_utility.h: added new type for array, array_type

3887

* bindings/jni/jni_runtime.cpp: add support for new array type

3888

(JavaField::valueFromInstance):

3889

(JavaField::setValueToInstance):

3890

(JavaMethod::JavaMethod):

3891

(JavaMethod::signature):

3892

* bindings/jni/jni_utility.cpp: add support for new array type

3893

(KJS::Bindings::callJNIMethod):

3894

(KJS::Bindings::callJNIStaticMethod):

3895

(KJS::Bindings::callJNIMethodIDA):

3896

(KJS::Bindings::JNITypeFromClassName):

3897

(KJS::Bindings::signatureFromPrimitiveType):

3898

(KJS::Bindings::JNITypeFromPrimitiveType):

3899

(KJS::Bindings::getJNIField):

3900

(KJS::Bindings::convertArrayInstanceToJavaArray): new method

3901

converts the Javascript array to the requested Java array.

3902

(KJS::Bindings::convertValueToJValue):

3903

3904

2007-02-08 Anders Carlsson <acarlsson@apple.com>

3905

3906

Reviewed by Geoff.

3907

3908

<rdar://problem/4930614>

3909

Safari complains about "Slow Script" if GMail is left open and machine is busy

3910

3911

<rdar://problem/4649516>

3912

Turn off slow script dialog or crank up time that makes it come up

3913

3914

<rdar://problem/4963589>

3915

Slow script warning is displayed after closing of PROMPT or PRINT dialog

3916

3917

Re-do the way script timeouts are handled. No longer use a unix timer that sends signals. Instead, add a

3918

tick count and increment it in loop bodies. If the tick count reaches a threshold, do a timeout check. If the total time executing

3919

is higher than the timeout value, (possibly) interrupt the script. The timeout checker also adjusts the threshold dynamically

3920

to prevent doing the timeout check too often.

3921

3922

* JavaScriptCore.exp:

3923

Remove pause and resume calls.

3924

3925

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

3926

Add winmm.lib.

3927

3928

* kjs/interpreter.cpp:

3929

(KJS::Interpreter::init):

3930

(KJS::Interpreter::~Interpreter):

3931

(KJS::Interpreter::startTimeoutCheck):

3932

(KJS::Interpreter::stopTimeoutCheck):

3933

(KJS::Interpreter::resetTimeoutCheck):

3934

(KJS::getCurrentTime):

3935

(KJS::Interpreter::checkTimeout):

3936

* kjs/interpreter.h:

3937

(KJS::Interpreter::timedOut):

3938

* kjs/nodes.cpp:

3939

(DoWhileNode::execute):

3940

(WhileNode::execute):

3941

(ForNode::execute):

3942

3943

2007-02-07 Darin Adler <darin@apple.com>

3944

3945

* JavaScriptCore.vcproj/JavaScriptCore.sln: Reenable testkjs.

3946

3947

2007-02-07 Darin Adler <darin@apple.com>

3948

3949

Reviewed by Geoff.

3950

3951

- another build fix; this time for sure

3952

3953

* pcre/pcre_exec.c: (match):

3954

The compiler caught an incorrect use of the othercase variable across

3955

a call to RMATCH in character repeat processing. Local variables can

3956

change in the crazy NO_RECURSE mode that we use, so we instead need

3957

the value in othercase to be in one of the special stack frame variables.

3958

Added a new stack frame variable for this purpose named repeat_othercase.

3959

Also noted a similar error in the non-UTF-16 side of the #ifdef, but

3960

didn't try to fix that one. Also removed a SUPPORT_UCP #ifdef from the

3961

PCRE_UTF16 side; that code doesn't work without the Unicde properties

3962

table, and we don't try to use it that way.

3963

3964

2007-02-06 Steve Falkenburg <sfalken@apple.com>

3965

3966

Disable testkjs in sln until we figure out mysterious compiler warning.

3967

3968

* JavaScriptCore.vcproj/JavaScriptCore.sln:

3969

3970

2007-02-06 Steve Falkenburg <sfalken@apple.com>

3971

3972

Build fix by ggaren

3973

3974

* pcre/pcre_exec.c:

3975

(match):

3976

3977

2007-02-06 Darin Adler <darin@apple.com>

3978

3979

Reviewed by Geoff.

3980

3981

- fix <rdar://problem/4979089> PCRE should avoid setjmp/longjmp even when compiler

3982

is not GCC

3983

3984

Added a new code path that's slower and way uglier but doesn't rely on GCC's

3985

computed gotos.

3986

3987

* pcre/pcre_exec.c: Added a numeric parameter to the RMATCH function. It must be

3988

different at every RMATCH call site. Changed the non-GCC NO_RECURSE version of

3989

the macro to use a label incorporating the number. Changed the RRETURN macro to

3990

use a goto instead of longjmp.

3991

(match): Added a different number at each callsite, using a perl script for the

3992

first-time task. Going forward it should be easy to maintain by hand. Added a

3993

switch statement at the bottom of the function. We'll get compile time errors

3994

if we have anything in the switch statement that's never used in an RMATCH,

3995

but errors in the other direction are silent except at runtime.

3996

3997

2007-02-06 Darin Adler <darin@apple.com>

3998

3999

Reviewed by John.

4000

4001

- fix <rdar://problem/4687840> 9A241: JavaScript RegExp 25-30x slower than on 10.4.7

4002

4003

I used Shark to figure out what to do. The test case is now 15% faster than with

4004

stock Safari. Some other regular expression cases might still be a few % slower

4005

than before, but the >10x slowdown is now completely gone.

4006

4007

1) Fix slowness caused by setjmp/longjmp by using computed goto instead.

4008

4009

Use GCC extensions - locally declared labels, labels as values, and computed goto -

4010

instead of using setjmp/longjmp to implemement non-recursive version of the regular

4011

expression system. We could probably make this even faster if we reduced the use

4012

of malloc a bit too.

4013

4014

2) Fix slowness caused by allocating heapframe objects by allocating the first

4015

16 of them from the stack.

4016

4017

3) Speed up use of malloc and free in PCRE by making it use fastMalloc and fastFree.

4018

4019

4) Speed up the test case by adding a special case to a UString function.

4020

4021

5) Made a small improvement to the innermost hottest loop of match by hoisting

4022

the conversion from int to pcre_uchar out of the loop.

4023

4024

* JavaScriptCore.xcodeproj/project.pbxproj: Compile FastMallocPCRE.cpp, and don't

4025

compile pcre_globals.c.

4026

4027

* wtf/FastMallocPCRE.cpp: Added. A copy of pcre_globals.c that uses FastMalloc.h.

4028

This is better than code that sets the PCRE allocation globals because by doing it

4029

this way there's guaranteed to be no problem with order of initialization.

4030

4031

* kjs/ustring.cpp: (KJS::UString::spliceSubstringsWithSeparators): Add a fast

4032

special case when this is called for only one subrange and no seaprators. This

4033

was happening a lot in the test case and it seems quite reasonable to optimize this.

4034

4035

* pcre/pcre_exec.c: Create a copy of the RMATCH and RRETURN macros that use goto

4036

instead of setjmp/longjmp. Change code that calls pcre_stack_malloc to first use

4037

storage on the stack inside the match function.

4038

(match): Move initialization of utf8 up a couple lines to avoid "possibly used

4039

uninitialized" warning. Use a local variable so we compare with pcre_uchar instead

4040

of with int inside the inner "find a character" loop.

4041

4042

2007-02-03 George Staikos <staikos@kde.org>

4043

4044

Reviewed by Alexey.

4045

4046

-1 is not a valid point. We can't handle anything > 0xffff anyway.

4047

Fixes crash on cases like eval("x");

4048

4049

* wtf/unicode/qt4/UnicodeQt4.h:

4050

(WTF::Unicode::category):

4051

4052

2007-02-02 Darin Adler <darin@apple.com>

4053

4054

Reviewed by Anders.

4055

4056

- fix copying and assigning a ListHashSet

4057

4058

No test because the code path with bugs I am fixing is not used yet.

4059

4060

* wtf/ListHashSet.h: Tweaked ListHashSetNodeAllocator a little bit for clarity.

4061

Changed m_allocator to be an OwnPtr instead of doing an explicit delete.

4062

Fixed bug in copy constructor where we'd have an uninitialized m_allocator.

4063

Fixed bug in assignment operator where it would swap only the hash table, and

4064

not the head, tail, and allocator pointers.

4065

4066

2007-02-02 Geoffrey Garen <ggaren@apple.com>

4067

4068

Reviewed by Maciej Stachowiak.

4069

4070

Use WTFLog instead of fprintf for logging KJS::Node leaks.

4071

4072

* kjs/nodes.cpp:

4073

(NodeCounter::~NodeCounter): Changed count to unsigned, updated

4074

to match style guidelines.

4075

4076

2007-02-02 Maciej Stachowiak <mjs@apple.com>

4077

4078

- not reviewed, build fix

4079

4080

* wtf/ListHashSet.h:

4081

(WTF::ListHashSetNodeAllocator::ListHashSetNodeAllocator): ummm, use union correctly

4082

4083

2007-02-01 Maciej Stachowiak <mjs@apple.com>

4084

4085

Reviewed by Darin.

4086

4087

- use a custom allocator for ListHashSet, to fix ~1% perf regression using it for form control

4088

4089

* wtf/ListHashSet.h:

4090

(WTF::ListHashSetNodeAllocator::ListHashSetNodeAllocator):

4091

(WTF::ListHashSetNodeAllocator::allocate):

4092

(WTF::ListHashSetNodeAllocator::deallocate):

4093

(WTF::ListHashSetNode::operator new):

4094

(WTF::ListHashSetNode::operator delete):

4095

(WTF::ListHashSetNode::destroy):

4096

(WTF::ListHashSetTranslator::translate):

4097

(WTF::::ListHashSet):

4098

(WTF::::~ListHashSet):

4099

(WTF::::add):

4100

(WTF::::unlinkAndDelete):

4101

(WTF::::deleteAllNodes):

4102

4103

2007-01-31 Maciej Stachowiak <mjs@apple.com>

4104

4105

Reviewed by Adam.

4106

4107

- fix sporadic crash

4108

4109

* wtf/ListHashSet.h:

4110

(WTF::::remove): remove before deleting

4111

4112

2007-01-31 Maciej Stachowiak <mjs@apple.com>

4113

4114

Reviewed by Mark with help from Lars.

4115

4116

- added new ListHashSet class, which combines a hashtable and a linked list to provide a set

4117

that keeps elements in inserted order

4118

4119

This is to assist in fixing the following:

4120

<rdar://problem/4751164> REGRESSION: Safari places text on incorrect button when returning to a page via back [10541]

4121

http://bugs.webkit.org/show_bug.cgi?id=10541

4122

4123

* JavaScriptCore.vcproj/WTF/WTF.vcproj:

4124

* JavaScriptCore.xcodeproj/project.pbxproj:

4125

* wtf/HashTable.h:

4126

(WTF::HashTable::find):

4127

(WTF::HashTable::contains):

4128

(WTF::::find):

4129

(WTF::::contains):

4130

* wtf/ListHashSet.h: Added.

4131

(WTF::ListHashSetNode::ListHashSetNode):

4132

(WTF::ListHashSetNodeHashFunctions::hash):

4133

(WTF::ListHashSetNodeHashFunctions::equal):

4134

(WTF::ListHashSetIterator::ListHashSetIterator):

4135

(WTF::ListHashSetIterator::get):

4136

(WTF::ListHashSetIterator::operator*):

4137

(WTF::ListHashSetIterator::operator->):

4138

(WTF::ListHashSetIterator::operator++):

4139

(WTF::ListHashSetIterator::operator--):

4140

(WTF::ListHashSetIterator::operator==):

4141

(WTF::ListHashSetIterator::operator!=):

4142

(WTF::ListHashSetIterator::operator const_iterator):

4143

(WTF::ListHashSetIterator::node):

4144

(WTF::ListHashSetConstIterator::ListHashSetConstIterator):

4145

(WTF::ListHashSetConstIterator::get):

4146

(WTF::ListHashSetConstIterator::operator*):

4147

(WTF::ListHashSetConstIterator::operator->):

4148

(WTF::ListHashSetConstIterator::operator++):

4149

(WTF::ListHashSetConstIterator::operator--):

4150

(WTF::ListHashSetConstIterator::operator==):

4151

(WTF::ListHashSetConstIterator::operator!=):

4152

(WTF::ListHashSetConstIterator::node):

4153

(WTF::ListHashSetTranslator::hash):

4154

(WTF::ListHashSetTranslator::equal):

4155

(WTF::ListHashSetTranslator::translate):

4156

(WTF::::ListHashSet):

4157

(WTF::::operator):

4158

(WTF::::~ListHashSet):

4159

(WTF::::size):

4160

(WTF::::capacity):

4161

(WTF::::isEmpty):

4162

(WTF::::begin):

4163

(WTF::::end):

4164

(WTF::::find):

4165

(WTF::::contains):

4166

(WTF::::add):

4167

(WTF::::remove):

4168

(WTF::::clear):

4169

(WTF::::unlinkAndDelete):

4170

(WTF::::appendNode):

4171

(WTF::::deleteAllNodes):

4172

(WTF::::makeIterator):

4173

(WTF::::makeConstIterator):

4174

(WTF::deleteAllValues):

4175

4176

2007-01-30 Darin Adler <darin@apple.com>

4177

4178

* kjs/DateMath.cpp: Fix license header to reflect LGPL as the first license

4179

mentioned. We still mention the option of using under MPL or GPL since some

4180

of this code came from the Mozilla project with those license terms.

4181

4182

2007-01-30 Simon Hausmann <hausmann@kde.org>

4183

4184

Reviewed by Zack.

4185

4186

Turned JavaScriptCore from a separate library into an includable

4187

project, to combine it all into libWebKitQt.

4188

4189

* JavaScriptCore.pri: Added.

4190

* JavaScriptCore.pro: Removed.

4191

* kjs/testkjs.pro:

4192

4193

2007-01-29 Geoffrey Garen <ggaren@apple.com>

4194

4195

Reviewed by Maciej Stachowiak.

4196

4197

Fixed <rdar://problem/4485644> REGRESSION: JavaScriptCore has init routines

4198

4199

The TCMalloc module now initializes, if needed, inside GetCache() and

4200

fastMallocSetIsMultiThreaded(). We leverage the same synchronization

4201

technique used for enabling / disabling the single-threaded optimization

4202

to synchronize initialization of the library without requiring a lock

4203

for every malloc.

4204

4205

1,251 runs of tcmalloc_unittest, 2 runs of a custom, massively multi-threaded

4206

tcmalloc_unittest, and my custom version of the PLT show no regressions.

4207

Super-accurate JS iBench reports a .24% regression, which is right at the

4208

limit of its error range, so I'm declaring victory.

4209

4210

* wtf/FastMalloc.cpp:

4211

(WTF::fastMallocSetIsMultiThreaded): Initialize, if needed. (InitModule()

4212

checks the "if needed" part.)

4213

(WTF::TCMalloc_ThreadCache::GetCache): Restored original TCMalloc code

4214

inside #ifdef, for posterity. Added new initialization logic.

4215

(WTF::TCMalloc_ThreadCache::InitModule): Call InitTSD(), since we don't

4216

have a static initializer to call it for us, now. This means that fastMalloc

4217

is not usable as a general libc allocator, but it never was, and if it were

4218

the general libc allocator, we wouldn't be here in the first place, so whatever.

4219

(WTF::TCMalloc_ThreadCache::InitTSD): Don't try to take the pageheap_lock,

4220

since InitModule already has it.

4221

4222

2007-01-29 Kevin McCullough <KMcCullough@apple.com>

4223

4224

Reviewed by Geoff and Oliver.

4225

4226

- rdar://problem/4955561

4227

- missusing JavaScript shouldn't crash webkit. Now it doesn't, in this case.

4228

4229

* bindings/objc/objc_runtime.mm:

4230

(ObjcFallbackObjectImp::callAsFunction):

4231

* bindings/runtime_method.cpp:

4232

(RuntimeMethod::callAsFunction):

4233

* bindings/runtime_object.cpp:

4234

(RuntimeObjectImp::callAsFunction):

4235

4236

2007-01-28 Geoffrey Garen <ggaren@apple.com>

4237

4238

Reviewed by Maciej Stachowiak.

4239

4240

First step in fixing <rdar://problem/4485644> REGRESSION: JavaScriptCore

4241

has init routines

4242

4243

Don't rely on a static initializer to store the main thread's ID (which

4244

we would use to detect allocations on secondary threads). Instead, require

4245

the caller to notify fastMalloc if it might allocate on a secondary thread.

4246

4247

Also fixed what seemed like a race condition in do_malloc.

4248

4249

tcmalloc_unittest and my custom versions of JS iBench and PLT show no

4250

regressions.

4251

4252

* wtf/FastMalloc.cpp:

4253

(WTF::fastMallocSetIsMultiThreaded):

4254

(1) Renamed from "fastMallocRegisterThread", which was a misleading name because

4255

not all threads need to register with fastMalloc -- only secondary threads

4256

need to, and only for the purpose of disabling its single-threaded optimization.

4257

4258

(2) Use the pageheap_lock instead of a custom one, since we need to synchronize

4259

with the read of isMultiThreaded inside CreateCacheIfNecessary. This is a new

4260

requirement, now that we can't guarantee that the first call to CreateCacheIfNecessary

4261

will occur on the main thread at init time, before any other threads have been created.

4262

4263

(WTF::TCMalloc_ThreadCache::CreateCacheIfNecessary):

4264

(WTF::do_malloc): Reverted WTF change only to call GetCache() if size <= kMaxSize.

4265

The WTF code would read phinited without holding the pageheap_lock, which

4266

seemed like a race condition. Regardless, calling GetCache reduces the number

4267

of code paths to module initialization, which will help in writing the

4268

final fix for this bug.

4269

4270

2007-01-28 David Kilzer <ddkilzer@webkit.org>

4271

4272

Reviewed by Darin.

4273

4274

- fix http://bugs.webkit.org/show_bug.cgi?id=9815

4275

JavaScript TypeError loading Dean Edwards' JS compressor/obfuscator

4276

4277

Creating a function using 'new Function()' was not setting its prototype with the

4278

same flags as 'function() { }'.

4279

4280

Test: fast/js/function-prototype.html

4281

4282

* kjs/function_object.cpp:

4283

(FunctionObjectImp::construct): Change flags from DontEnum|DontDelete|ReadOnly to

4284

Internal|DontDelete to match FuncDeclNode::processFuncDecl() and

4285

FuncExprNode::evaluate() in kjs/nodes.cpp.

4286

4287

2007-01-27 Geoffrey Garen <ggaren@apple.com>

4288

4289

Reviewed by Beth Dakin.

4290

4291

Added some missing JSLocks, which might fix <rdar://problem/4889707>.

4292

4293

We need to lock whenever we might allocate memory because our FastMalloc

4294

implementation requires clients to register their threads, which we do

4295

through JSLock.

4296

4297

We also need to lock whenever modifying ref-counts because they're not

4298

thread-safe.

4299

4300

* API/JSObjectRef.cpp:

4301

(JSClassCreate): Allocates memory

4302

(JSClassRetain): Modifies a ref-count

4303

(JSClassRelease): Modifies a ref-count

4304

(JSPropertyNameArrayRetain): Modifies a ref-count

4305

(JSPropertyNameArrayRelease): Modifies a ref-count

4306

* API/JSStringRef.cpp:

4307

(JSStringRetain): Modifies a ref-count

4308

* API/JSValueRef.cpp:

4309

(JSValueIsInstanceOfConstructor): Might allocate memory if an exception

4310

is thrown.

4311

4312

2007-01-27 Lars Knoll <lars@trolltech.com>

4313

4314

Fix the Qt build.

4315

4316

* bindings/qt/qt_instance.h:

4317

4318

2007-01-25 Geoffrey Garen <ggaren@apple.com>

4319

4320

Reviewed by Maciej Stachowiak.

4321

4322

Fixed <rdar://problem/4608404> WebScriptObject's _rootObject lack

4323

of ownership policy causes crashes (e.g., in Dashcode)

4324

4325

The old model for RootObject ownership was either to (1) leak them or (2) assign

4326

them to a single owner -- the WebCore::Frame -- which would destroy them

4327

when it believed that all of its plug-ins had unloaded.

4328

4329

This model was broken because of (1) and also because plug-ins are not the only

4330

RootObject clients. All Bindings clients are RootObjects clients, including

4331

applications, which outlive any particular WebCore::Frame.

4332

4333

The new model for RootObject ownership is to reference-count them, with a

4334

throw-back to the old model: The WebCore::Frame tracks the RootObjects

4335

it creates, and invalidates them when it believes that all of its plug-ins

4336

have unloaded.

4337

4338

We maintain this throw-back to avoid plug-in leaks, particularly from Java.

4339

Java is completely broken when it comes to releasing JavaScript objects.

4340

Comments in our code allege that Java does not always call finalize when

4341

collecting objects. Moreoever, my own testing reveals that, when Java does

4342

notify JavaScript of a finalize, the data it provides is totally bogus.

4343

4344

This setup is far from ideal, but I don't think we can do better without

4345

completely rewriting the bindings code, and possibly part of the Java

4346

plug-in / VM.

4347

4348

Layout tests pass. No additional leaks reported. WebCore/manual-tests/*liveconnect*

4349

and a few LiveConnect demos on the web also run without a hitch.

4350

4351

const RootObject* => RootObject*, since we need to ref/deref

4352

4353

* bindings/NP_jsobject.cpp:

4354

(jsDeallocate): deref our RootObjects. Also unprotect or JSObject, instead

4355

of just relying on the RootObject to do it for us when it's invalidated.

4356

(_isSafeScript): Check RootObject validity.

4357

(_NPN_CreateScriptObject): ditto

4358

(_NPN_Invoke): ditto

4359

(_NPN_Evaluate): ditto

4360

(_NPN_GetProperty): ditto

4361

(_NPN_SetProperty): ditto

4362

(_NPN_RemoveProperty): ditto

4363

(_NPN_HasProperty): ditto

4364

(_NPN_HasMethod): ditto

4365

(_NPN_SetException): ditto

4366

4367

* bindings/runtime_root.cpp:

4368

Revived bit-rotted LIAR LIAR LIAR comment.

4369

4370

LOOK: Added support for invalidating RootObjects without deleting them,

4371

which is the main goal of this patch.

4372

4373

Moved protect counting into the RootObject class, to emphasize that

4374

the RootObject protects the JSObject, and unprotects it upon being invalidated.

4375

addNativeReference => RootObject::gcProtect

4376

removeNativeReference => RootObject::gcUnprotect

4377

ProtectCountSet::contains => RootObject::gcIsProtected

4378

4379

I know we'll all be sad to see the word "native" go.

4380

4381

* bindings/runtime_root.h: Added ref-counting support to RootObject, with

4382

all the standard accoutrements.

4383

4384

* bindings/c/c_utility.cpp:

4385

(KJS::Bindings::convertValueToNPVariant): If we can't find a valid RootObject,

4386

return void instead of just leaking.

4387

4388

* bindings/jni/jni_instance.cpp:

4389

(JavaInstance::JavaInstance): Don't take a RootObject in our constructor;

4390

be like other Instances and require the caller to call setRootObject. This

4391

reduces the number of ownership code paths.

4392

(JavaInstance::invokeMethod): Check RootObject for validity.

4393

* bindings/jni/jni_instance.h: Removed private no-arg constructor. Having

4394

an arg constructor accomplishes the same thing.

4395

4396

* bindings/jni/jni_jsobject.cpp:

4397

(JavaJSObject::invoke): No need to call findProtectCountSet, because finalize()

4398

checks for RootObject validity.

4399

(JavaJSObject::JavaJSObject): check RootObject for validity

4400

(JavaJSObject::call): ditto

4401

(JavaJSObject::eval): ditto

4402

(JavaJSObject::getMember): ditto

4403

(JavaJSObject::setMember): ditto

4404

(JavaJSObject::removeMember): ditto

4405

(JavaJSObject::getSlot): ditto

4406

(JavaJSObject::setSlot): ditto

4407

(JavaJSObject::toString): ditto

4408

(JavaJSObject::finalize): ditto

4409

(JavaJSObject::createNative): No need to tell the RootObject to protect

4410

the global object, since the RootObject already owns the interpreter.

4411

4412

* bindings/jni/jni_runtime.cpp:

4413

(JavaArray::JavaArray): Removed copy construcutor becaue it was unused.

4414

Dead code is dangerous code.

4415

4416

* bindings/objc/objc_runtime.mm: Added WebUndefined protocol. Previous use

4417

of WebScriptObject was bogus, because WebUndefined is not a subclass of

4418

WebScriptObject.

4419

(convertValueToObjcObject): If we can't find a valid RootObject,

4420

return nil instead of just leaking.

4421

4422

* bindings/objc/objc_utility.mm:

4423

(KJS::Bindings::convertValueToObjcValue): If we can't find a valid RootObject,

4424

return nil instead of just leaking.

4425

4426

2007-01-27 Andrew Wellington <proton@wiretapped.net>

4427

4428

Reviewed by Maciej.

4429

4430

Fix for Repeated string concatenation results in OOM crash

4431

http://bugs.webkit.org/show_bug.cgi?id=11131

4432

4433

* kjs/operations.cpp:

4434

(KJS::add): Throw exception if string addition result is null

4435

* kjs/ustring.cpp:

4436

(KJS::UString::UString): Don't call memcpy when malloc failed

4437

4438

2007-01-25 Jan Kraemer <camel@gmx.de>

4439

4440

Reviewed by Maciej

4441

4442

Fix for http://bugs.webkit.org/show_bug.cgi?id=12382

4443

4444

Fix crash on architectures with 32 bit ints and

4445

64 bit longs (For example Linux on AMD64)

4446

4447

* kjs/dtoa.cpp: #define Long int as suggested in comment

4448

4449

2007-01-24 Geoffrey Garen <ggaren@apple.com>

4450

4451

Fixed up #include order for style. No review necessary.

4452

4453

* API/JSStringRef.cpp:

4454

4455

2007-01-24 Geoffrey Garen <ggaren@apple.com>

4456

4457

Reviewed by Maciej Stachowiak.

4458

4459

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

4460

Copy JSStringRefCF, in case anybody wants to use it. (I just added

4461

it recently.)

4462

4463

2007-01-24 Maciej Stachowiak <mjs@apple.com>

4464

4465

Not reviewed, trivial property change.

4466

4467

* JavaScriptCore.vcproj/JavaScriptCore.sln: remove svn:mime-type

4468

property which made this binary.

4469

4470

2007-01-25 Mark Rowe <mrowe@apple.com>

4471

4472

Reviewed by Darin.

4473

4474

* Info.plist: Update copyright string.

4475

4476

2007-01-24 Darin Adler <darin@apple.com>

4477

4478

Reviewed by Mark Rowe.

4479

4480

* JavaScriptCore.xcodeproj/project.pbxproj: Changed to /usr/sbin/sysctl

4481

so we don't rely on people's paths.

4482

4483

2007-01-23 Alice Liu <alice.liu@apple.com>

4484

4485

release build fix

4486

4487

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

4488

Copy APICasts.h

4489

4490

2007-01-23 Geoffrey Garen <ggaren@apple.com>

4491

4492

build fix

4493

4494

* API/JSStringRef.h:

4495

* JavaScriptCore.xcodeproj/project.pbxproj:

4496

4497

2007-01-24 Mark Rowe <mrowe@apple.com>

4498

4499

Build fix for DumpRenderTree.

4500

4501

* JavaScriptCore.xcodeproj/project.pbxproj: Make JSStringRefCF.h public so it's copied into built framework.

4502

4503

2007-01-23 Anders Carlsson <acarlsson@apple.com>

4504

4505

Reviewed by Darin.

4506

4507

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

4508

Copy APICasts.h

4509

4510

2007-01-23 Geoffrey Garen <ggaren@apple.com>

4511

4512

Reviewed by Maciej Stachowiak.

4513

4514

Fixed <rdar://problem/4885131> Move CFString function declarations from

4515

JSStringRef.h to JSStringRefCF.h

4516

4517

Also removed remaining API FIXMEs and changed them into Radars.

4518

4519

* API/JSClassRef.cpp:

4520

(OpaqueJSClass::OpaqueJSClass): Added Radar numbers for UTF8 conversion.

4521

4522

* API/JSContextRef.cpp:

4523

(JSGlobalContextCreate): Replaced FIXME for NULL JSContextRef with Radar number.

4524

4525

* API/JSObjectRef.h: Removed FIXME, which is unprofessional in a public header.

4526

4527

* API/JSStringRef.cpp: Moved CF related implementations to JSStringRefCF.cpp.

4528

(JSStringCreateWithUTF8CString): Replaced FIXME with Radar number.

4529

* API/JSStringRef.h: Moved CF related declarations to JSStringRefCF.h. Added

4530

#include of JSStringRefCF.h as a stopgap until clients start #including

4531

it as needed by themselves.

4532

4533

* API/JSStringRefCF.cpp: Added.

4534

(JSStringCreateWithCFString):

4535

(JSStringCopyCFString): Replaced JSChar cast with UniChar cast, which is

4536

more appropriate for a CF call.

4537

* API/JSStringRefCF.h: Added.

4538

* JavaScriptCore.xcodeproj/project.pbxproj:

4539

4540

2007-01-18 Sanjay Madhav <sanjay12@gmail.com>

4541

4542

Reviewed by Darin.

4543

4544

Add JavaScriptCore define to help with tracing of when objects are marked.

4545

4546

* kjs/object.cpp:

4547

(KJS::JSObject::mark):

4548

4549

2007-01-18 Simon Hausmann <hausmann@kde.org>

4550

4551

Reviewed by Zack.

4552

4553

* JavaScriptCore.pro: Remove generated files on make clean.

4554

* pcre/pcre.pri:

4555

4556

2007-01-16 Alexey Proskuryakov <ap@webkit.org>

4557

4558

Reviewed by Maciej.

4559

4560

http://bugs.webkit.org/show_bug.cgi?id=12268

4561

Give object prototypes their own names

4562

4563

* kjs/lookup.h: Append "Prototype" to ClassName in KJS_IMPLEMENT_PROTOTYPE.

4564

4565

2007-01-16 Geoffrey Garen <ggaren@apple.com>

4566

4567

Reviewed by Darin Adler.

4568

4569

Added re-entrency checking to GC allocation and collection. It is an error

4570

to allocate or collect from within a collection. We've had at least one

4571

case of each bug in the past.

4572

4573

Added a comment to the API header, explaining that API clients must not

4574

make this mistake, either.

4575

4576

Layout tests and JS tests pass.

4577

4578

* API/JSObjectRef.h:

4579

* kjs/collector.cpp:

4580

(KJS::GCLock::GCLock):

4581

(KJS::GCLock::~GCLock):

4582

(KJS::Collector::allocate):

4583

(KJS::Collector::collect):

4584

4585

2007-01-14 Mark Rowe <mrowe@apple.com>

4586

4587

Reviewed by Mitz.

4588

4589

Minor fixes to JavaScript pretty-printing.

4590

4591

* JavaScriptCore.exp:

4592

* kjs/Parser.cpp:

4593

(KJS::Parser::prettyPrint): Return line number and error message if parsing fails.

4594

* kjs/Parser.h:

4595

* kjs/nodes2string.cpp:

4596

(ElementNode::streamTo): Include comma delimiters in array literals.

4597

(PropertyNameNode::streamTo): Quote property names in object literals to handle the case when the property name is not a valid identifier.

4598

* kjs/testkjs.cpp:

4599

(doIt): Print any errors encountered while pretty-printing.

4600

4601

2007-01-12 Anders Carlsson <acarlsson@apple.com>

4602

4603

Reviewed by Darin.

4604

4605

* wtf/HashTraits.h:

4606

Add hash traits for unsigned long and unsigned long long.

4607

4608

2007-01-12 Geoffrey Garen <ggaren@apple.com>

4609

4610

RS by Brady Eidson.

4611

4612

Rolling back in r18786 with leaks fixed, and these renames slightly reworked:

4613

4614

Because they can return 0:

4615

rootObjectForImp => findRootObject (overloaded for JSObject* and Interpreter*)

4616

rootObjectForInterpreter => findRootObject (ditto)

4617

findReferenceSet => findProtectCountSet

4618

4619

2007-01-11 Geoffrey Garen <ggaren@apple.com>

4620

4621

RS by Brady Eidson.

4622

4623

Rolling out r18786 because it caused leaks.

4624

4625

2007-01-11 Geoffrey Garen <ggaren@apple.com>

4626

4627

Reviewed by Anders Carlsson.

4628

4629

Even more cleanup in preparation for fixing <rdar://problem/4608404>

4630

WebScriptObject's _executionContext lack of ownership policy causes

4631

crashes (e.g., in Dashcode)

4632

4633

Layout tests pass.

4634

4635

Renames:

4636

ReferencesSet | ProtectCounts => ProtectCountSet (because it's a typename for a set of GC protect counts)

4637

ReferencesByRootMap => RootObjectMap (because RootObjectToProtectCountSetMap would have been confusing)

4638

pv => protectedValues

4639

rootObjectForImp => getRootObject (overloaded for JSObject* and Interpreter*)

4640

rootObjectForInterpreter => getRootObject (ditto)

4641

findReferenceSet => getProtectCountSet

4642

imp => jsObject

4643

4644

(KJS::Bindings::getRootObjectMap): Changed to take advantage of built-in

4645

facility for initializing static variables.

4646

4647

(KJS::Bindings::getProtectCountSet):

4648

(KJS::Bindings::destroyProtectCountSet): Added. Helps encapsulate the fact

4649

that getting a ProtectCountSet entails adding a RootObject to a hash table,

4650

and destroying one entails the reverse.

4651

4652

(KJS::Bindings::getRootObject): Removed spurious NULL check.

4653

4654

(KJS::Bindings::findReferenceSet): Renamed. Changed to use getRootObject()

4655

instead of iterating on its own.

4656

4657

(KJS::Bindings::addNativeReference): Changed to use an early return instead

4658

of indenting the whole function.

4659

(KJS::Bindings::removeNativeReference): Ditto.

4660

4661

2007-01-11 Geoffrey Garen <ggaren@apple.com>

4662

4663

Reviewed by Anders Carlsson.

4664

4665

Even more cleanup in preparation for fixing <rdar://problem/4608404>

4666

WebScriptObject's _executionContext lack of ownership policy causes

4667

crashes (e.g., in Dashcode)

4668

4669

Layout tests pass.

4670

4671

Renames:

4672

findRootObjectForNativeHandleFunction => createRootObject

4673

FindRootObjectForNativeHandleFunctionPtr => CreateRootObjectFunction

4674

4675

Also removed unnecessary use of "Bindings::" prefix.

4676

4677

* JavaScriptCore.exp:

4678

* bindings/jni/jni_jsobject.cpp:

4679

(JavaJSObject::createNative):

4680

(JavaJSObject::convertValueToJObject):

4681

(JavaJSObject::convertJObjectToValue):

4682

* bindings/runtime_root.cpp:

4683

(KJS::Bindings::RootObject::setCreateRootObject):

4684

* bindings/runtime_root.h:

4685

(KJS::Bindings::RootObject::createRootObject):

4686

4687

2007-01-11 George Staikos <staikos@kde.org>

4688

4689

Reviewed by Maciej

4690

4691

Appears to be Mac specific right now.

4692

4693

* kjs/config.h:

4694

4695

2007-01-10 Lars Knoll <lars@trolltech.com>

4696

4697

Reviewed by Zack

4698

4699

Use the new functionality in Qt 4.3, to make

4700

the methods closer compliant with the Unicode

4701

spec.

4702

4703

Keep the old code so that it still compiles against

4704

Qt 4.2.

4705

4706

* wtf/unicode/qt4/UnicodeQt4.h:

4707

(WTF::Unicode::toLower):

4708

(WTF::Unicode::toUpper):

4709

(WTF::Unicode::toTitleCase):

4710

(WTF::Unicode::foldCase):

4711

(WTF::Unicode::isFormatChar):

4712

(WTF::Unicode::isPrintableChar):

4713

(WTF::Unicode::isSeparatorSpace):

4714

(WTF::Unicode::isPunct):

4715

(WTF::Unicode::isDigit):

4716

(WTF::Unicode::isLower):

4717

(WTF::Unicode::isUpper):

4718

(WTF::Unicode::digitValue):

4719

(WTF::Unicode::mirroredChar):

4720

(WTF::Unicode::combiningClass):

4721

(WTF::Unicode::decompositionType):

4722

(WTF::Unicode::umemcasecmp):

4723

(WTF::Unicode::direction):

4724

(WTF::Unicode::category):

4725

4726

2007-01-09 Darin Adler <darin@apple.com>

4727

4728

- update 2007 Apple copyright for the new company name

4729

4730

* kjs/DateMath.cpp:

4731

4732

2007-01-09 Darin Adler <darin@apple.com>

4733

4734

- fix build

4735

4736

* kjs/string_object.cpp: (KJS::StringProtoFunc::callAsFunction):

4737

Actually compile it this time.

4738

4739

2007-01-09 Darin Adler <darin@apple.com>

4740

4741

- fix build

4742

4743

* kjs/string_object.cpp: (KJS::StringProtoFunc::callAsFunction):

4744

Change types.

4745

4746

2007-01-09 Darin Adler <darin@apple.com>

4747

4748

- fix build on platforms where Unicode::UChar is != uint16_t

4749

4750

* kjs/string_object.cpp: (KJS::StringProtoFunc::callAsFunction):

4751

Change types.

4752

4753

2007-01-09 Mitz Pettel <mitz@webkit.org>

4754

4755

Reviewed by Darin.

4756

4757

- changes for http://bugs.webkit.org/show_bug.cgi?id=11078

4758

Forms Don't Submit (ASP Pages)

4759

4760

* JavaScriptCore.exp:

4761

* kjs/value.cpp:

4762

(KJS::JSValue::toInt32): Folded toInt32Inline into this method, which was its

4763

only caller.

4764

(KJS::JSValue::toUInt32): Added a variant that reports if the conversion has

4765

succeeded.

4766

* kjs/value.h:

4767

4768

2007-01-09 Darin Adler <darin@apple.com>

4769

4770

Reviewed by Maciej.

4771

4772

- fix http://bugs.webkit.org/show_bug.cgi?id=12174

4773

improve Unicode use (less WTF::Unicode:: prefix, centralized character names)

4774

4775

* wtf/unicode/icu/UnicodeIcu.h: Change parameter and return types

4776

to UChar32 and UChar. Removed unneeded type casts and added some

4777

const to functions that lacked it. Removed WTF::Unicode::memcmp.

4778

(WTF::Unicode::umemcasecmp): Renamed from strcasecmp since this

4779

doesn't work on 0-terminated strings as the str functions do.

4780

* wtf/unicode/qt4/UnicodeQt4.h: Ditto.

4781

4782

- got rid of namespace prefixes from most uses of WTF::Unicode

4783

4784

* kjs/function.cpp:

4785

(KJS::isStrWhiteSpace):

4786

(KJS::escapeStringForPrettyPrinting):

4787

* kjs/lexer.cpp:

4788

(KJS::Lexer::isWhiteSpace):

4789

(KJS::Lexer::isIdentStart):

4790

(KJS::Lexer::isIdentPart):

4791

* kjs/string_object.cpp:

4792

(KJS::StringProtoFunc::callAsFunction):

4793

4794

2007-01-07 David Kilzer <ddkilzer@webkit.org>

4795

4796

Reviewed by Darin.

4797

4798

- fix http://bugs.webkit.org/show_bug.cgi?id=11917

4799

setlocale() can return null

4800

4801

* kjs/date_object.cpp:

4802

(KJS::DateProtoFunc::callAsFunction): Removed dead code.

4803

4804

2007-01-07 David Carson <dacarson@gmail.com>

4805

4806

Reviewed by Darin.

4807

4808

- fix http://bugs.webkit.org/show_bug.cgi?id=12100

4809

JNI bindings should be available to non-Mac platforms that have JNI

4810

4811

Change JNI so that it is not wrapped in the PLATFORM(MAC) ifdef, enabling

4812

other platforms who have JNI to use it.

4813

4814

* bindings/jni/jni_instance.h:

4815

Removed unnecessary include of <CoreFoundation/CoreFoundation.h>

4816

* bindings/jni/jni_utility.cpp:

4817

(KJS::Bindings::setJavaVM):

4818

* bindings/jni/jni_utility.h:

4819

Added new method for clients to set the JavaVM

4820

* bindings/runtime.cpp:

4821

(KJS::Bindings::Instance::createBindingForLanguageInstance):

4822

Changed code to utilize new #if HAVE(JNI)

4823

* kjs/config.h:

4824

Added new #define for JNI, ie HAVE_JNI

4825

4826

2007-01-07 David Carson <dacarson@gmail.com>

4827

4828

Reviewed by Darin.

4829

4830

Fix http://bugs.webkit.org/show_bug.cgi?id=11431

4831

ARM platform has some byte alignment issues

4832

4833

Fix for NaN being 4 bytes and it must start on a byte boundary

4834

for ARM architectures.

4835

4836

* kjs/fpconst.cpp:

4837

(KJS::):

4838

4839

2007-01-04 David Kilzer <ddkilzer@webkit.org>

4840

4841

Reviewed by Kevin McCullough.

4842

4843

- fix http://bugs.webkit.org/show_bug.cgi?id=12070

4844

REGRESSION: KJS::getUTCOffset() caches UTC offset but ignores time zone changes

4845

4846

* kjs/DateMath.cpp:

4847

(KJS::getUTCOffset): Don't cache UTC offset.

4848

4849

2007-01-02 Darin Adler <darin@apple.com>

4850

4851

- minor tweak (hope this doesn't re-break Windows)

4852

4853

* pcre/pcre_compile.c: Removed use of const pcre_uchar const * -- Mitz probably

4854

meant const pcre_uchar *const, but I think we can do without the explicit const here.

4855

4856

* pcre/pcre_internal.h: Re-enabled warning C4114.

4857

4858

2007-01-02 David Kilzer <ddkilzer@webkit.org>

4859

4860

Reviewed by NOBODY (Windows build fix).

4861

4862

The MSVC compiler requires variables to be declared at the top of the enclosing block in C source.

4863

4864

Disable this warning to prevent MSVC from complaining about the 'const pcre_uchar const *' type:

4865

warning C4114: same type qualifier used more than once

4866

4867

* pcre/pcre_compile.c:

4868

(pcre_compile2): Moved variable declarations to top of their respective enclosing blocks.

4869

* pcre/pcre_internal.h: Added pragma to disable compiler warning.

4870

4871

2007-01-01 Mitz Pettel <mitz@webkit.org>

4872

4873

Reviewed by Darin.

4874

4875

- fix http://bugs.webkit.org/show_bug.cgi?id=11849

4876

REGRESSION (r18182): Google Calendar is broken (a regular expression containing a null character is not parsed correctly)

4877

4878

Modified pcre_compile() (and the functions that it calls) to work with patterns

4879

containing null characters.

4880

4881

Covered by JavaScriptCore tests ecma_3/RegExp/octal-002.js and ecma_3/RegExp/regress-85721.js

4882

4883

* kjs/regexp.cpp:

4884

(KJS::RegExp::RegExp): Changed to not null-terminate the pattern string and instead

4885

pass its length to pcre_compile.

4886

* pcre/pcre.h:

4887

* pcre/pcre_compile.c:

4888

(check_escape):

4889

(get_ucp):

4890

(is_counted_repeat):

4891

(check_posix_syntax):

4892

(compile_branch):

4893

(compile_regex):

4894

(pcre_compile): Added a parameter specifying the length of the pattern, which

4895

is no longer required to be null-terminated and may contain null characters.

4896

(pcre_compile2):

4897

* pcre/pcre_internal.h:

4898

* tests/mozilla/expected.html: Updated for the two tests that this patch

4899

fixes. Also updated failing results for ecma_3/RegExp/regress-100199.js

4900

which were not updated after bug 6257 was fixed.

4901

4902

2007-01-01 David Kilzer <ddkilzer@webkit.org>

4903

4904

Reviewed by Darin.

4905

4906

- fix http://bugs.webkit.org/show_bug.cgi?id=12057

4907

REGRESSION: JavaScript Date Is One Day In The Future in GMT time zone

4908

4909

Because Mac OS X returns geographically and historically accurate time zone information,

4910

converting Jan 02, 1970 12:00:00 AM to local time then subtracting 24 hours did not work

4911

in GMT (London - England) since it was in BST (+0100) all year in 1970[1]. Instead, the

4912

UTC offset is calculated by converting Jan 01, 2000 12:00:00 AM to local time then

4913

subtracting that from the same date in UTC.

4914

4915

[1] http://en.wikipedia.org/wiki/British_Summer_Time

4916

4917

* kjs/DateMath.cpp:

4918

(KJS::getUTCOffset): Updated UTC offset calculation.

4919

(KJS::getDSTOffset): Improved comment.

4920

4921

2006-12-31 David Kilzer <ddkilzer@webkit.org>

4922

4923

Reviewed by Geoff.

4924

4925

Update embedded pcre library from version 6.2 to 6.4. Changes from pcre 6.2 to 6.3

4926

did not include any files in JavaScriptCore/pcre.

4927

4928

All changes include renaming EXPORT to PCRE_EXPORT, renaming of ucp_findchar() to

4929

_pcre_ucp_findchar(), or comment changes. Additional changes noted below.

4930

4931

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: Updated source file list.

4932

* JavaScriptCore.xcodeproj/project.pbxproj: Renamed pcre_printint.c to pcre_printint.src

4933

and changed it from a source file to a header file.

4934

* JavaScriptCoreSources.bkl: Updated source file list.

4935

* pcre/CMakeLists.txt: Updated source file list.

4936

* pcre/pcre-config.h:

4937

* pcre/pcre.h: Updated version.

4938

* pcre/pcre.pri: Updated source file list.

4939

* pcre/pcre_compile.c: Include pcre_printint.src #if DEBUG.

4940

(pcre_compile2):

4941

* pcre/pcre_config.c:

4942

* pcre/pcre_exec.c:

4943

(match):

4944

* pcre/pcre_fullinfo.c:

4945

* pcre/pcre_info.c:

4946

* pcre/pcre_internal.h: Added header guard. Removed export of _pcre_printint().

4947

* pcre/pcre_ord2utf8.c:

4948

* pcre/pcre_printint.c: Renamed to pcre_printint.src.

4949

* pcre/pcre_printint.src: Added. Renamed _pcre_printint() to pcre_printint().

4950

* pcre/pcre_refcount.c:

4951

* pcre/pcre_study.c:

4952

* pcre/pcre_tables.c:

4953

* pcre/pcre_try_flipped.c:

4954

* pcre/pcre_ucp_findchar.c: Added contents of ucp_findchar.c.

4955

* pcre/pcre_version.c:

4956

* pcre/pcre_xclass.c:

4957

(_pcre_xclass):

4958

* pcre/ucp.h: Removed export of ucp_findchar().

4959

* pcre/ucp_findchar.c: Removed. Contents moved to pcre_ucp_findchar.c.

4960

4961

2006-12-29 David Kilzer <ddkilzer@webkit.org>

4962

4963

Reviewed by Geoff.

4964

4965

Update embedded pcre library from version 6.1 to 6.2. From the pcre ChangeLog:

4966

4967

3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like

4968

operating environments where this matters.

4969

4970

5. Named capturing subpatterns were not being correctly counted when a pattern

4971

was compiled. This caused two problems: (a) If there were more than 100

4972

such subpatterns, the calculation of the memory needed for the whole

4973

compiled pattern went wrong, leading to an overflow error. (b) Numerical

4974

back references of the form \12, where the number was greater than 9, were

4975

not recognized as back references, even though there were sufficient

4976

previous subpatterns.

4977

4978

* pcre/dftables.c: Item 3.

4979

(main):

4980

* pcre/pcre.h: Updated version.

4981

* pcre/pcre_compile.c: Item 5.

4982

(read_repeat_counts):

4983

(pcre_compile2):

4984

4985

2006-12-29 Geoffrey Garen <ggaren@apple.com>

4986

4987

Reviewed by Brian Dash... err... Mark Rowe.

4988

4989

More cleanup in preparation for fixing <rdar://problem/4608404>

4990

WebScriptObject's _executionContext lack of ownership policy causes

4991

crashes (e.g., in Dashcode)

4992

4993

The key change here is to RootObject::RootObject().

4994

4995

* JavaScriptCore.exp:

4996

4997

* bindings/c/c_utility.cpp:

4998

(KJS::Bindings::convertValueToNPVariant): Changed to use new constructor.

4999

5000

* bindings/jni/jni_jsobject.cpp:

5001

(JavaJSObject::createNative): Changed to use new constructor. Replaced

5002

large 'if' followed by default condition with "if !" and explicit default

5003

condition.

5004

5005

* bindings/objc/objc_runtime.mm:

5006

(convertValueToObjcObject): Changed to use new constructor.

5007

5008

* bindings/runtime_root.cpp:

5009

(KJS::Bindings::RootObject::destroy): "removeAllNativeReferences" => "destroy"

5010

because this function actually destroys the RootObject.

5011

5012

* bindings/runtime_root.h: Changed Interpreter* to RefPtr<Interpreter>

5013

to prevent a RootObject from holding a stale Interperter*.

5014

5015

(KJS::Bindings::RootObject::RootObject): Changed constructor to take an

5016

Interpreter*, since it's pointless to create a RootObject without one.

5017

Removed setRootObjectImp() and rootObjectImp() because they were just

5018

a confusing way of setting and getting the Interpreter's global object.

5019

5020

(KJS::Bindings::RootObject::nativeHandle): "_nativeHandle" => "m_nativeHandle"

5021

(KJS::Bindings::RootObject::interpreter): "_interpreter" => "m_interpreter"

5022

5023

2006-12-28 George Staikos <staikos@kde.org>

5024

5025

Reviewed by Olliej.

5026

5027

* bindings/qt/qt_instance.cpp: build

5028

(KJS::Bindings::QtInstance::QtInstance):

5029

5030

2006-12-28 Geoffrey Garen <ggaren@apple.com>

5031

5032

Reviewed by Oliver Hunt.

5033

5034

More cleanup. Layout tests pass.

5035

5036

Use a helper function to initialize and access WebUndefined and WebScriptObject.

5037

5038

* bindings/objc/objc_runtime.h:

5039

* bindings/objc/objc_runtime.mm:

5040

(KJS::Bindings::webScriptObjectClass):

5041

(KJS::Bindings::webUndefinedClass):

5042

(convertValueToObjcObject):

5043

* bindings/objc/objc_utility.mm:

5044

(KJS::Bindings::convertValueToObjcValue):

5045

(KJS::Bindings::convertObjcValueToValue):

5046

5047

2006-12-28 Geoffrey Garen <ggaren@apple.com>

5048

5049

Reviewed by Brady Eidson.

5050

5051

Some cleanup in preparation for fixing <rdar://problem/4608404>

5052

WebScriptObject's _executionContext lack of ownership policy causes

5053

crashes (e.g., in Dashcode)

5054

5055

I'm just trying to make heads or tails of this baffling code.

5056

5057

Renamed "root" | "execContext" | "executionContext" => "rootObject", because

5058

that's the object's (admittedly vague) type name.

5059

5060

* bindings/runtime.cpp: Removed createLanguageInstanceForValue

5061

because I'll give you a dollar if you can explain to me what it actually did.

5062

5063

* bindings/runtime_root.cpp: Put everything in the KJS::Bindings namespace,

5064

removing the KJS::Bindings prefix from individual functions and datatypes.

5065

This matches the header and eliminates a lot of syntax cruft.

5066

5067

* bindings/c/c_utility.cpp:

5068

(KJS::Bindings::convertValueToNPVariant): Replaced use of createLanguageInstanceForValue

5069

with call to _NPN_CreateScriptObject because that's what createLanguageInstanceForValue

5070

actually did (but don't ask me for that dollar now; that's cheating.)

5071

5072

* bindings/objc/objc_utility.h:

5073

* bindings/objc/objc_utility.mm:

5074

(KJS::Bindings::convertValueToObjcValue): Removed. Its only purpose was

5075

to call a single function for WebKit, which WebKit can do on its own.

5076

5077

* kjs/interpreter.h: Removed rtti() because it was unused, and this class

5078

is scheduled for demolition anyway.

5079

5080

* kjs/interpreter.cpp: Removed createLanguageInstanceForValue because it had

5081

nothing to do with the Interpreter, and nothing makes Chuck Norris more mad

5082

than a function whose sole purpose is to call another function of the same

5083

name. (Really, I asked him.)

5084

5085

2006-12-26 Geoffrey Garen <ggaren@apple.com>

5086

5087

Reviewed by Eric Seidel.

5088

5089

Some cleanup in preparation for fixing <rdar://problem/4740328> Safari

5090

crash on quit in _NPN_ReleaseObject from KJS::Bindings::CInstance::~CInstance

5091

5092

* bindings/c/c_instance.cpp:

5093

* bindings/c/c_instance.h: Removed unused copy constructor and assignment

5094

operator. They made tracking data flow more difficult. Unused code is also

5095

dangerous because it can succumb to bit rot with the stealth of a Ninja.

5096

5097

Replaced #include with forward declaration to reduce header dependency.

5098

5099

* bindings/npruntime.cpp: Sorted #includes.

5100

(_NPN_GetStringIdentifier): Replaced assert with ASSERT.

5101

(_NPN_GetStringIdentifiers): ditto

5102

(_NPN_ReleaseVariantValue): ditto

5103

(_NPN_CreateObject): ditto

5104

(_NPN_RetainObject): ditto

5105

(_NPN_ReleaseObject): ditto

5106

(_NPN_DeallocateObject): ditto

5107

5108

2006-12-20 Anders Carlsson <acarlsson@apple.com>

5109

5110

* kjs/string_object.cpp:

5111

(localeCompare):

5112

Another speculative Win32 fix.

5113

5114

2006-12-20 Anders Carlsson <acarlsson@apple.com>

5115

5116

* kjs/string_object.cpp:

5117

(localeCompare):

5118

Speculative Win32 fix.

5119

5120

2006-12-20 Anders Carlsson <acarlsson@apple.com>

5121

5122

Reviewed by Darin.

5123

5124

<rdar://problem/4235733>

5125

<http://bugs.webkit.org/?show_bug.cgi?id=10193>

5126

support String.localeCompare.

5127

5128

Implement localeCompare.

5129

5130

* JavaScriptCore.xcodeproj/project.pbxproj:

5131

* kjs/string_object.cpp:

5132

(localeCompare):

5133

(StringProtoFunc::callAsFunction):

5134

* kjs/string_object.h:

5135

(KJS::StringProtoFunc::):

5136

5137

2006-12-20 Timothy Hatcher <timothy@apple.com>

5138

5139

Reviewed by Mark Rowe.

5140

5141

* JavaScriptCore.xcodeproj/project.pbxproj: use GCC 4.0 for all the other test targets

5142

5143

2006-12-20 Timothy Hatcher <timothy@apple.com>

5144

5145

Reviewed by Mark Rowe.

5146

5147

<rdar://problem/4871613> JavaScriptCore-421.31's dftables target needs to override default compiler and use gcc-4.0

5148

5149

* JavaScriptCore.xcodeproj/project.pbxproj:

5150

5151

2006-12-20 Lars Knoll <lars@trolltech.com>

5152

5153

Reviewed by David Hyatt

5154

5155

Added support to bind QObject's to

5156

JavaScript.

5157

5158

* JavaScriptCore.pro:

5159

* bindings/qt/qt_class.cpp: Added.

5160

(KJS::Bindings::QtClass::QtClass):

5161

(KJS::Bindings::QtClass::~QtClass):

5162

(KJS::Bindings::QtClass::classForObject):

5163

(KJS::Bindings::QtClass::name):

5164

(KJS::Bindings::QtClass::methodsNamed):

5165

(KJS::Bindings::QtClass::fieldNamed):

5166

* bindings/qt/qt_class.h: Added.

5167

(KJS::Bindings::QtClass::constructorAt):

5168

(KJS::Bindings::QtClass::numConstructors):

5169

* bindings/qt/qt_instance.cpp: Added.

5170

(KJS::Bindings::QtInstance::QtInstance):

5171

(KJS::Bindings::QtInstance::~QtInstance):

5172

(KJS::Bindings::QtInstance::operator=):

5173

(KJS::Bindings::QtInstance::getClass):

5174

(KJS::Bindings::QtInstance::begin):

5175

(KJS::Bindings::QtInstance::end):

5176

(KJS::Bindings::QtInstance::implementsCall):

5177

(KJS::Bindings::QtInstance::invokeMethod):

5178

(KJS::Bindings::QtInstance::invokeDefaultMethod):

5179

(KJS::Bindings::QtInstance::defaultValue):

5180

(KJS::Bindings::QtInstance::stringValue):

5181

(KJS::Bindings::QtInstance::numberValue):

5182

(KJS::Bindings::QtInstance::booleanValue):

5183

(KJS::Bindings::QtInstance::valueOf):

5184

* bindings/qt/qt_instance.h: Added.

5185

(KJS::Bindings::QtInstance::getObject):

5186

* bindings/qt/qt_runtime.cpp: Added.

5187

(KJS::Bindings::convertValueToQVariant):

5188

(KJS::Bindings::convertQVariantToValue):

5189

(KJS::Bindings::QtField::name):

5190

(KJS::Bindings::QtField::valueFromInstance):

5191

(KJS::Bindings::QtField::setValueToInstance):

5192

* bindings/qt/qt_runtime.h: Added.

5193

(KJS::Bindings::QtField::QtField):

5194

(KJS::Bindings::QtField::type):

5195

(KJS::Bindings::QtMethod::QtMethod):

5196

(KJS::Bindings::QtMethod::name):

5197

(KJS::Bindings::QtMethod::numParameters):

5198

* bindings/runtime.cpp:

5199

(KJS::Bindings::Instance::createBindingForLanguageInstance):

5200

* bindings/runtime.h:

5201

(KJS::Bindings::Instance::):

5202

* bindings/testbindings.pro: Added.

5203

* bindings/testqtbindings.cpp: Added.

5204

(MyObject::MyObject):

5205

(MyObject::setTestString):

5206

(MyObject::setTestInt):

5207

(MyObject::testString):

5208

(MyObject::testInt):

5209

(MyObject::foo):

5210

(Global::className):

5211

(main):

5212

5213

2006-12-19 Anders Carlsson <acarlsson@apple.com>

5214

5215

Reviewed by Geoff.

5216

5217

Add -p option to testkjs which pretty prints the files instead of executing them.

5218

5219

* JavaScriptCore.exp:

5220

* JavaScriptCore.xcodeproj/project.pbxproj:

5221

* kjs/Parser.cpp:

5222

(KJS::Parser::prettyPrint):

5223

* kjs/Parser.h:

5224

* kjs/testkjs.cpp:

5225

(doIt):

5226

5227

2006-12-19 Brady Eidson <beidson@apple.com>

5228

5229

Rubberstamped by Lou

5230

5231

Removed unneccessary "else"

5232

5233

* wtf/Assertions.cpp:

5234

5235

2006-12-19 Timothy Hatcher <timothy@apple.com>

5236

5237

Reviewed by Darin.

5238

5239

<rdar://problem/4891774> Local WebCore/WebBrowser builds fail in 9A328 due to warning about ObjC-2.0 language features

5240

5241

* JavaScriptCore.xcodeproj/project.pbxproj:

5242

5243

2006-12-17 Simon Hausmann <hausmann@kde.org>

5244

5245

Reviewed by Zack.

5246

5247

* kjs/testkjs.pro: Oops, make it also build on machines other than

5248

mine :)

5249

5250

2006-12-17 Simon Hausmann <hausmann@kde.org>

5251

5252

Reviewed by Rob Buis.

5253

5254

* kjs/testkjs.pro: Added .pro file to build testkjs.

5255

5256

2006-12-16 Alexey Proskuryakov <ap@webkit.org>

5257

5258

Reviewed by Rob.

5259

5260

A deleted object was accessed to prepare RegExp construction error messages.

5261

5262

* kjs/regexp_object.cpp:

5263

(RegExpObjectImp::construct): Wrap the RegExp into an OwnPtr.

5264

5265

2006-12-16 Mitz Pettel <mitz@webkit.org>

5266

5267

Reviewed by Alexey.

5268

5269

- fix http://bugs.webkit.org/show_bug.cgi?id=11814

5270

REGRESSION(r18098): Find does not work with capital letters

5271

5272

Test: editing/execCommand/findString-3.html

5273

5274

* wtf/unicode/icu/UnicodeIcu.h:

5275

(WTF::Unicode::foldCase): Changed to not return an error if the result fits

5276

in the buffer without a null terminator.

5277

5278

2006-12-13 Maciej Stachowiak <mjs@apple.com>

5279

5280

Reviewed by Anders.

5281

5282

- added equality and inequality operations for HashMap and Vector, useful for comparing more complex types

5283

5284

* wtf/HashMap.h:

5285

(WTF::operator==):

5286

(WTF::operator!=):

5287

* wtf/Vector.h:

5288

(WTF::operator==):

5289

(WTF::operator!=):

5290

5291

2006-12-12 Alexey Proskuryakov <ap@webkit.org>

5292

5293

Reviewed by Geoff. Based on a patch by Maks Orlovich.

5294

5295

http://bugs.webkit.org/show_bug.cgi?id=6257

5296

Throw errors on invalid expressions (KJS merge)

5297

5298

* kjs/regexp.cpp:

5299

(KJS::RegExp::RegExp):

5300

(KJS::RegExp::~RegExp):

5301

(KJS::RegExp::match):

5302

* kjs/regexp.h:

5303

(KJS::RegExp::flags):

5304

(KJS::RegExp::isValid):

5305

(KJS::RegExp::errorMessage):

5306

(KJS::RegExp::subPatterns):

5307

Remember and report RegExp construction failures. Renamed data members not to start with underscores.

5308

5309

* kjs/regexp_object.cpp:

5310

(RegExpObjectImp::construct): Raise an exception if RegExp construction fails.

5311

(RegExpObjectImp::callAsFunction): Removed an obsolete comment.

5312

5313

* tests/mozilla/ecma_3/RegExp/regress-119909.js: Reduced the number of nested parentheses to

5314

a value supported by PCRE.

5315

5316

2006-12-11 Alexey Proskuryakov <ap@webkit.org>

5317

5318

Reviewed by Darin.

5319

5320

http://bugs.webkit.org/show_bug.cgi?id=9673

5321

Add support for window.atob() and window.btoa()

5322

5323

* JavaScriptCore.exp: Export UString::is8Bit().

5324

* JavaScriptCore.xcodeproj/project.pbxproj: Added StringExtras.h as

5325

a private header.

5326

5327

2006-12-11 Darin Adler <darin@apple.com>

5328

5329

Reviewed by Brady.

5330

5331

* JavaScriptCore.xcodeproj/project.pbxproj: Let Xcode update this

5332

(I think Hyatt is using an old Xcode).

5333

5334

2006-12-11 David Hyatt <hyatt@apple.com>

5335

5336

Fix the failing layout test. Just remove Unicode::isSpace and

5337

revert StringImpl to do the same thing it was doing before.

5338

5339

Reviewed by darin

5340

5341

* wtf/unicode/icu/UnicodeIcu.h:

5342

* wtf/unicode/qt4/UnicodeQt4.h:

5343

5344

2006-12-09 George Staikos <staikos@kde.org>

5345

5346

Reviewed by Zack.

5347

5348

Fix bison again on qmake build.

5349

5350

* JavaScriptCore.pro:

5351

5352

2006-12-09 Lars Knoll <lars@trolltech.com>

5353

5354

Reviewed by Zack

5355

5356

Make it possible to build WebKit with qmake.

5357

5358

* JavaScriptCore.pro: Added.

5359

* kjs/kjs.pro: Removed.

5360

* pcre/pcre.pri: Added.

5361

5362

2006-12-09 Zack Rusin <zack@kde.org>

5363

5364

Fixing the compilation with platform kde after the icu changes.

5365

5366

* CMakeLists.txt:

5367

5368

2006-12-09 Adam Roben <aroben@apple.com>

5369

5370

Reviewed by Darin.

5371

5372

Some updates in reaction to r18098.

5373

5374

* wtf/unicode/icu/UnicodeIcu.h: Use !! to convert UBool to bool in all

5375

cases.

5376

(WTF::Unicode::toLower):

5377

(WTF::Unicode::toUpper):

5378

(WTF::Unicode::isDigit):

5379

(WTF::Unicode::isSpace):

5380

(WTF::Unicode::isPunct):

5381

(WTF::Unicode::isLower):

5382

(WTF::Unicode::isUpper):

5383

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

5384

* JavaScriptCore.vcproj/WTF/WTF.vcproj:

5385

5386

2006-12-09 George Staikos <staikos@kde.org>

5387

5388

Patch by Lars Knoll, comment out ICU dependency on Qt platform (unused code).

5389

5390

Reviewed by Darin.

5391

5392

* bindings/c/c_utility.cpp:

5393

(KJS::Bindings::convertUTF8ToUTF16):

5394

5395

2006-12-08 David Hyatt <hyatt@apple.com>

5396

5397

Land the new ICU abstraction layer. Patch by Lars.

5398

5399

Reviewed by me

5400

5401

* JavaScriptCore.xcodeproj/project.pbxproj:

5402

* wtf/Platform.h:

5403

* wtf/unicode/UnicodeCategory.h: Removed.

5404

* wtf/unicode/UnicodeDecomposition.h: Removed.

5405

* wtf/unicode/UnicodeDirection.h: Removed.

5406

* wtf/unicode/icu/UnicodeIcu.h:

5407

(WTF::Unicode::):

5408

(WTF::Unicode::foldCase):

5409

(WTF::Unicode::toLower):

5410

(WTF::Unicode::toUpper):

5411

(WTF::Unicode::toTitleCase):

5412

(WTF::Unicode::isDigit):

5413

(WTF::Unicode::isSpace):

5414

(WTF::Unicode::isPunct):

5415

(WTF::Unicode::mirroredChar):

5416

(WTF::Unicode::category):

5417

(WTF::Unicode::direction):

5418

(WTF::Unicode::isLower):

5419

(WTF::Unicode::isUpper):

5420

(WTF::Unicode::digitValue):

5421

(WTF::Unicode::combiningClass):

5422

(WTF::Unicode::decompositionType):

5423

(WTF::Unicode::strcasecmp):

5424

(WTF::Unicode::memset):

5425

* wtf/unicode/qt4/UnicodeQt4.cpp: Removed.

5426

* wtf/unicode/qt4/UnicodeQt4.h:

5427

(WTF::Unicode::):

5428

(WTF::Unicode::toLower):

5429

(WTF::Unicode::toUpper):

5430

(WTF::Unicode::toTitleCase):

5431

(WTF::Unicode::foldCase):

5432

(WTF::Unicode::isPrintableChar):

5433

(WTF::Unicode::isLower):

5434

(WTF::Unicode::isUpper):

5435

(WTF::Unicode::digitValue):

5436

(WTF::Unicode::combiningClass):

5437

(WTF::Unicode::decompositionType):

5438

(WTF::Unicode::strcasecmp):

5439

(WTF::Unicode::memset):

5440

(WTF::Unicode::direction):

5441

(WTF::Unicode::category):

5442

5443

=== Safari-521.32 ===

5444

5445

2006-12-08 Adam Roben <aroben@apple.com>

5446

5447

Reviewed by Anders.

5448

5449

This is a mo' better fix for ensuring we don't use macro definitions

5450

of min/max.

5451

5452

* kjs/config.h:

5453

* wtf/Vector.h:

5454

5455

2006-12-07 Kevin Fyure <digdog@macports.org>

5456

5457

Reviewed by Darin.

5458

5459

http://bugs.webkit.org/show_bug.cgi?id=11545

5460

Disable the testcases do not follow the ECMA-262v3 specification.

5461

5462

* tests/mozilla/expected.html: Update Results.

5463

* tests/mozilla/js1_2/String/concat.js:

5464

4 tests disabled. The result of concat Array object is not followinig

5465

ECMA 15.5.4.6

5466

* tests/mozilla/js1_2/function/Number.js:

5467

1 test disabled. The result of Array object to Number object conversion

5468

is not following ECMA 9.3. And the test was duplicated in

5469

ecma/TypeConversion/9.3-1.js

5470

* tests/mozilla/js1_2/function/String.js:

5471

2 tests disabled. The result of Object/Array object to String object

5472

conversion is not following ECMA 15.5.1.1 and ECMA 9.8

5473

5474

2006-11-30 Steve Falkenburg <sfalken@apple.com>

5475

5476

Reviewed by Oliver.

5477

5478

Move WTF from JavaScriptCore project into a new WTF project.

5479

5480

* JavaScriptCore.vcproj/JavaScriptCore.sln: Add WTF.vcproj to sln

5481

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: Remove WTF source files

5482

* JavaScriptCore.vcproj/WTF/WTF.vcproj: Added.

5483

* JavaScriptCore.vcproj/testkjs/testkjs.vcproj: Add dependency on WTF.lib

5484

5485

2006-11-30 Geoffrey Garen <ggaren@apple.com>

5486

5487

Reviewed by Beth Dakin.

5488

5489

Fixed up garbage collection at window close time.

5490

5491

* kjs/interpreter.cpp:

5492

(KJS::Interpreter::~Interpreter): Garbage collect here, since

5493

destroying the interpreter frees the global object and

5494

therefore creates a lot of garbage.

5495

5496

2006-11-20 W. Andy Carrel <wac@google.com>

5497

5498

Reviewed by Maciej.

5499

5500

http://bugs.webkit.org/show_bug.cgi?id=11501

5501

REGRESSION: \u no longer escapes metacharacters in RegExps

5502

http://bugs.webkit.org/show_bug.cgi?id=11502

5503

Serializing RegExps doesn't preserve Unicode escapes

5504

5505

* kjs/lexer.cpp:

5506

(Lexer::Lexer):

5507

(Lexer::setCode):

5508

(Lexer::shift):

5509

(Lexer::scanRegExp):

5510

Push \u parsing back down into the RegExp object rather than in the

5511

parser. This backs out r17354 in favor of a new fix that better

5512

matches the behavior of other browsers.

5513

5514

* kjs/lexer.h:

5515

* kjs/regexp.cpp:

5516

(KJS::RegExp::RegExp):

5517

(KJS::sanitizePattern):

5518

(KJS::isHexDigit):

5519

(KJS::convertHex):

5520

(KJS::convertUnicode):

5521

* kjs/regexp.h:

5522

Translate \u escaped unicode characters for the benefit of pcre.

5523

5524

* kjs/ustring.cpp:

5525

(KJS::UString::append):

5526

Fix failure to increment length on the first UChar appended to a

5527

UString that was copy-on-write.

5528

5529

* tests/mozilla/ecma_2/RegExp/properties-001.js:

5530

Adjust tests back to the uniform standards.

5531

5532

2006-11-20 Samuel Weinig <sam@webkit.org>

5533

5534

Reviewed by Maciej.

5535

5536

Fix for http://bugs.webkit.org/show_bug.cgi?id=11647

5537

Fix Win32 build

5538

5539

* kjs/config.h: define NOMINMAX instead of min/max

5540

as themselves.

5541

* wtf/Vector.h: put back hack to ensure that min/max

5542

are not defined as macros.

5543

5544

2006-11-19 Simon Hausmann <hausmann@kde.org>

5545

5546

Reviewed by Zack.

5547

5548

http://bugs.webkit.org/show_bug.cgi?id=11649

5549

Fix CMake Qt-only build without KDE CMake files

5550

5551

* CMakeLists.txt:

5552

* pcre/CMakeLists.txt:

5553

5554

2006-11-17 Anders Carlsson <acarlsson@apple.com>

5555

5556

Reviewed by Adam.

5557

5558

Make sure that we always use std::min and std::max instead of macros.

5559

5560

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

5561

* kjs/config.h:

5562

* wtf/Vector.h:

5563

5564

=== Safari-521.31 ===

5565

5566

2006-11-12 Geoffrey Garen <ggaren@apple.com>

5567

5568

Reviewed by Beth Dakin.

5569

5570

Added project-wide setting to disable Microsoft's made-up deprecation

5571

warnings related to std:: functions. (Doesn't have any affect yet,

5572

since we currently disable all deprecation warnings.)

5573

5574

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

5575

5576

2006-11-12 Mark Rowe <bdash@webkit.org>

5577

5578

Reviewed by Mitz.

5579

5580

Clean up of JavaScriptCore bakefiles.

5581

5582

* JavaScriptCoreSources.bkl:

5583

* jscore.bkl:

5584

5585

2006-11-11 Alexey Proskuryakov <ap@webkit.org>

5586

5587

Reviewed by Maciej.

5588

5589

http://bugs.webkit.org/show_bug.cgi?id=11508

5590

Undisable some warnings for JSImmediate.h

5591

5592

Fix suggested by Don Gibson.

5593

5594

* kjs/JSImmediate.h:

5595

Re-enable all MSVC warnings, move the remaining runtime checks

5596

to compile-time.

5597

5598

2006-11-10 Zalan Bujtas <zalan.bujtas@nokia.com>

5599

5600

Reviewed by Maciej.

5601

5602

Added s60/symbian platform defines.

5603

http://bugs.webkit.org/show_bug.cgi?id=11540

5604

5605

* wtf/Platform.h:

5606

5607

=== Safari-521.30 ===

5608

5609

2006-11-08 Ada Chan <adachan@apple.com>

5610

5611

Reviewed by darin.

5612

5613

Added a method to delete all the keys in a HashMap.

5614

5615

* wtf/HashMap.h:

5616

(WTF::deleteAllPairFirsts):

5617

(WTF::deleteAllKeys):

5618

5619

2006-11-07 Anders Carlsson <acarlsson@apple.com>

5620

5621

Reviewed by Geoff.

5622

5623

* API/JSClassRef.cpp:

5624

(OpaqueJSClass::OpaqueJSClass):

5625

Initialize cachedPrototype to 0.

5626

5627

2006-11-06 Krzysztof Kowalczyk <kkowalczyk@gmail.com>

5628

5629

Reviewed by Maciej.

5630

5631

Remove warning about garbage after #else. #else clause applies for all

5632

non-mac platforms, not only win.

5633

5634

* kjs/date_object.cpp:

5635

5636

2006-11-06 Mark Rowe <bdash@webkit.org>

5637

5638

Reviewed by the wonderful Mitz Pettel.

5639

5640

http://bugs.webkit.org/show_bug.cgi?id=11524

5641

Bug 11524: REGRESSION(r9842): Array.prototype.join should use ToString operator rather than calling toString on each element

5642

5643

* kjs/array_object.cpp:

5644

(ArrayProtoFunc::callAsFunction): Use ToString operator on each element rather than calling their toString method.

5645

5646

2006-11-03 Steve Falkenburg <sfalken@apple.com>

5647

5648

Fix build

5649

5650

* kjs/JSImmediate.h:

5651

5652

2006-11-03 Alexey Proskuryakov <ap@nypop.com>

5653

5654

Reviewed by Darin.

5655

5656

http://bugs.webkit.org/show_bug.cgi?id=11504

5657

Fix warnings on non 32 bit platforms

5658

5659

* kjs/JSImmediate.h:

5660

(KJS::JSImmediate::NanAsBits):

5661

(KJS::JSImmediate::oneAsBits):

5662

Rewrite in a way that moves runtime checks to compile-time.

5663

5664

(KJS::):

5665

(KJS::JSImmediate::fromDouble):

5666

(KJS::JSImmediate::toDouble):

5667

5668

2006-11-02 George Staikos <staikos@kde.org>

5669

5670

Reviewed by Maciej.

5671

5672

* collector.cpp:

5673

Remove a deprecated pthreads call.

5674

5675

2006-11-02 Anders Carlsson <acarlsson@apple.com>

5676

5677

Reviewed by Maciej, landed by Anders.

5678

5679

* CMakeLists.txt:

5680

Make KDE support optional.

5681

5682

2006-11-01 Kevin McCullough <KMcCullough@apple.com>

5683

5684

Reviewed by Brady.

5685

5686

- Fixes many JavaScriptCore tests in other timezones. The root problem is that on mac localtime() returns historically accurate information for DST, but the JavaScript spec explicitly states to not take into account historical information but rather to interpolate from valid years.

5687

5688

* kjs/DateMath.cpp:

5689

(KJS::equivalentYearForDST):

5690

(KJS::getDSTOffsetSimple):

5691

(KJS::getDSTOffset):

5692

5693

2006-10-31 Geoffrey Garen <ggaren@apple.com>

5694

5695

Reviewed by Beth.

5696

5697

Fixed http://bugs.webkit.org/show_bug.cgi?id=11477

5698

REGRESSION: GMail crashes in KJS::FunctionImp::callerGetter

5699

5700

* kjs/function.cpp:

5701

(KJS::FunctionImp::argumentsGetter): Removed unnecessary braces.

5702

(KJS::FunctionImp::callerGetter): More logical NULL checking.

5703

5704

2006-10-31 Oliver Hunt <oliver@apple.com>

5705

5706

Reviewed by Geoff.

5707

5708

Adding definition for PLATFORM(CI)

5709

5710

* wtf/Platform.h:

5711

5712

2006-10-31 Vladimir Olexa <vladimir.olexa@gmail.com>

5713

5714

Reviewed by Geoff.

5715

5716

http://bugs.webkit.org/show_bug.cgi?id=4166

5717

Function object does not support caller property

5718

5719

Test: fast/js/caller-property.html

5720

5721

* kjs/function.cpp:

5722

(KJS::FunctionImp::callerGetter): added

5723

(KJS::FunctionImp::getOwnPropertySlot): added if statement to handle callerGetter()

5724

* kjs/function.h: added callerGetter() declaration

5725

* kjs/identifier.h: added caller property macro

5726

* tests/mozilla/expected.html:

5727

5728

2006-10-30 Kevin McCullough <KMcCullough@apple.com>

5729

5730

Reviewed by Adam.

5731

5732

- Fix some timezone issues and JavaScriptCore date tests. Addresses bugzilla 4930.

5733

5734

* kjs/DateMath.h:

5735

(KJS::GregorianDateTime::GregorianDateTime): Here's the fix, to add parenthesis for order of precedence.

5736

* kjs/date_object.cpp:

5737

(KJS::DateProtoFunc::callAsFunction):

5738

(KJS::DateObjectImp::construct): memset not needed as GregorianDateTime initializes itself.

5739

5740

2006-10-30 Darin Adler <darin@apple.com>

5741

5742

Reviewed by John Sullivan.

5743

5744

* kjs/SavedBuiltins.h: Added needed include.

5745

* wtf/OwnPtr.h: (WTF::OwnPtr::set): Fixed mistake in assertion.

5746

5747

2006-10-28 Darin Adler <darin@apple.com>

5748

5749

Reviewed by Maciej.

5750

5751

- renamed PassRefPtr::release to releaseRef to make it clearer that

5752

it's the counterpart of adoptRef, and to make it harder to confuse

5753

it with the safer-to-use RefPtr::release

5754

5755

* kjs/identifier.cpp:

5756

(KJS::CStringTranslator::translate):

5757

(KJS::UCharBufferTranslator::translate):

5758

* kjs/ustring.cpp:

5759

(KJS::UString::Rep::create):

5760

* wtf/PassRefPtr.h:

5761

(WTF::PassRefPtr::PassRefPtr):

5762

(WTF::PassRefPtr::~PassRefPtr):

5763

(WTF::PassRefPtr::get):

5764

(WTF::PassRefPtr::releaseRef):

5765

(WTF::PassRefPtr::operator->):

5766

(WTF::PassRefPtr::operator=):

5767

(WTF::adoptRef):

5768

(WTF::static_pointer_cast):

5769

(WTF::const_pointer_cast):

5770

* wtf/RefPtr.h:

5771

(WTF::RefPtr::RefPtr):

5772

(WTF::RefPtr::operator=):

5773

5774

2006-10-28 Darin Adler <darin@apple.com>

5775

5776

Reviewed by Steve.

5777

5778

* kjs/grammar.y: Add definitions of YYMALLOC and YYFREE to fix

5779

a warning some people see (not sure why others don't see it).

5780

5781

* JavaScriptCore.vcproj/JavaScriptCore/grammarWrapper.cpp: Touch

5782

this file to force it to re-build grammar.cpp.

5783

5784

2006-10-28 Darin Adler <darin@apple.com>

5785

5786

Reviewed by Geoff.

5787

5788

- made changes so the code compiles with the highest warning level

5789

under MSVC (disabling some warnings, making some code fixes)

5790

5791

* API/JSCallbackConstructor.cpp:

5792

(KJS::JSCallbackConstructor::construct):

5793

* API/JSCallbackFunction.cpp:

5794

(KJS::JSCallbackFunction::callAsFunction):

5795

* API/JSCallbackObject.cpp:

5796

(KJS::JSCallbackObject::init):

5797

(KJS::JSCallbackObject::construct):

5798

(KJS::JSCallbackObject::callAsFunction):

5799

* API/JSObjectRef.cpp:

5800

(JSPropertyNameArrayGetNameAtIndex):

5801

* API/JSStringRef.cpp:

5802

(JSStringCreateWithCharacters):

5803

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

5804

* bindings/c/c_utility.cpp:

5805

(KJS::Bindings::convertUTF8ToUTF16):

5806

(KJS::Bindings::coerceValueToNPVariantStringType):

5807

(KJS::Bindings::convertValueToNPVariant):

5808

* kjs/DateMath.h:

5809

(KJS::GregorianDateTime::GregorianDateTime):

5810

* kjs/ExecState.h:

5811

(KJS::ExecState::hadException):

5812

* kjs/JSImmediate.h:

5813

(KJS::JSImmediate::fromDouble):

5814

(KJS::JSImmediate::toDouble):

5815

(KJS::JSImmediate::NanAsBits):

5816

(KJS::JSImmediate::oneAsBits):

5817

* kjs/Parser.h:

5818

* kjs/PropertyNameArray.h:

5819

(KJS::PropertyNameArray::size):

5820

* kjs/array_object.cpp:

5821

(ArrayObjectImp::callAsFunction):

5822

* kjs/bool_object.cpp:

5823

(BooleanObjectImp::callAsFunction):

5824

* kjs/collector.cpp:

5825

(KJS::Collector::allocate):

5826

(KJS::Collector::markCurrentThreadConservatively):

5827

(KJS::Collector::collect):

5828

* kjs/completion.h:

5829

(KJS::Completion::isValueCompletion):

5830

* kjs/date_object.cpp:

5831

(KJS::findMonth):

5832

* kjs/debugger.cpp:

5833

(Debugger::sourceParsed):

5834

(Debugger::sourceUnused):

5835

(Debugger::exception):

5836

(Debugger::atStatement):

5837

(Debugger::callEvent):

5838

(Debugger::returnEvent):

5839

* kjs/dtoa.cpp:

5840

* kjs/error_object.cpp:

5841

(ErrorObjectImp::callAsFunction):

5842

(NativeErrorImp::callAsFunction):

5843

* kjs/function.cpp:

5844

(KJS::FunctionImp::processVarDecls):

5845

(KJS::GlobalFuncImp::callAsFunction):

5846

* kjs/function_object.cpp:

5847

(FunctionPrototype::callAsFunction):

5848

* kjs/grammar.y:

5849

* kjs/identifier.cpp:

5850

(KJS::CStringTranslator::translate):

5851

(KJS::Identifier::add):

5852

* kjs/internal.h:

5853

* kjs/lexer.cpp:

5854

(Lexer::lex):

5855

(Lexer::isIdentStart):

5856

(Lexer::isIdentPart):

5857

(isDecimalDigit):

5858

(Lexer::isHexDigit):

5859

(Lexer::isOctalDigit):

5860

(Lexer::matchPunctuator):

5861

(Lexer::singleEscape):

5862

(Lexer::convertOctal):

5863

(Lexer::convertHex):

5864

(Lexer::convertUnicode):

5865

(Lexer::record8):

5866

* kjs/lexer.h:

5867

* kjs/math_object.cpp:

5868

(MathFuncImp::callAsFunction):

5869

* kjs/number_object.cpp:

5870

(integer_part_noexp):

5871

(intPow10):

5872

(NumberProtoFunc::callAsFunction):

5873

(NumberObjectImp::callAsFunction):

5874

* kjs/object.cpp:

5875

(KJS::JSObject::deleteProperty):

5876

(KJS::JSObject::callAsFunction):

5877

(KJS::JSObject::toBoolean):

5878

(KJS::JSObject::toObject):

5879

* kjs/object.h:

5880

(KJS::JSObject::getPropertySlot):

5881

* kjs/property_map.cpp:

5882

(KJS::isValid):

5883

(KJS::PropertyMap::put):

5884

(KJS::PropertyMap::insert):

5885

(KJS::PropertyMap::containsGettersOrSetters):

5886

* kjs/property_map.h:

5887

(KJS::PropertyMap::hasGetterSetterProperties):

5888

* kjs/property_slot.h:

5889

* kjs/string_object.cpp:

5890

(StringInstance::getPropertyNames):

5891

(StringObjectImp::callAsFunction):

5892

(StringObjectFuncImp::callAsFunction):

5893

* kjs/ustring.cpp:

5894

(KJS::UString::Rep::computeHash):

5895

(KJS::UString::UString):

5896

(KJS::UString::from):

5897

(KJS::UString::append):

5898

(KJS::UString::ascii):

5899

(KJS::UString::operator=):

5900

(KJS::UString::find):

5901

(KJS::UString::rfind):

5902

* kjs/ustring.h:

5903

(KJS::UChar::high):

5904

(KJS::UChar::low):

5905

(KJS::UCharReference::low):

5906

(KJS::UCharReference::high):

5907

* kjs/value.cpp:

5908

(KJS::JSValue::toUInt16):

5909

* kjs/value.h:

5910

* pcre/pcre_compile.c:

5911

(get_othercase_range):

5912

* pcre/pcre_exec.c:

5913

(match):

5914

* pcre/pcre_internal.h:

5915

* wtf/HashFunctions.h:

5916

(WTF::intHash):

5917

(WTF::PtrHash::hash):

5918

* wtf/MathExtras.h:

5919

(isnan):

5920

(lround):

5921

(lroundf):

5922

* wtf/StringExtras.h:

5923

(strncasecmp):

5924

* wtf/unicode/icu/UnicodeIcu.h:

5925

(WTF::Unicode::isPrintableChar):

5926

5927

2006-10-26 W. Andy Carrel <wac@google.com>

5928

5929

Reviewed by Maciej.

5930

5931

- Fix http://bugs.webkit.org/show_bug.cgi?id=7445 /

5932

<rdar://problem/4614195> (and 7253 / <rdar://4694011>) by changing

5933

inline regexps so that they can have \u escaped Unicode sequences and

5934

still work properly.

5935

5936

* kjs/lexer.cpp:

5937

(Lexer::Lexer):

5938

(Lexer::setCode):

5939

(Lexer::shift): Looking ahead one additional character for the benefit

5940

of scanRegExp

5941

(Lexer::scanRegExp): Change code to support unicode escapes in inline

5942

regexps.

5943

* kjs/lexer.h: Extra lookahead added.

5944

5945

=== Safari-521.29 ===

5946

5947

2006-10-26 Nikolas Zimmermann <zimmermann@kde.org>

5948

5949

Reviewed by Darin.

5950

5951

Fix build with older gcc 3.3.4.

5952

5953

* kjs/DateMath.cpp: Remove inline prefix.

5954

(KJS::equivalentYearForDST):

5955

5956

2006-10-26 Darin Adler <darin@apple.com>

5957

5958

Reviewed by John.

5959

5960

- fix iteration of properties of string objects (found because of a warning

5961

emitted by the MSVC compiler)

5962

5963

* kjs/string_object.cpp: (StringInstance::getPropertyNames): Change code that

5964

wants to format a number as a string to use UString::from. Before it was using

5965

the UString constructor that makes a string from a character!

5966

5967

* kjs/ustring.h:

5968

* kjs/ustring.cpp: Remove the dangerous and not all that helpful UString(char)

5969

constructor.

5970

5971

* kjs/grammar.y: Change code to not depend on the UString(char) constructor.

5972

This is potentially more efficient anyway because we could overload the + operator

5973

some day to handle char* directly instead of creating a UString.

5974

5975

* kjs/nodes2string.cpp: (SourceStream::operator<<): Change code to not depend on

5976

the UString(char) constructor.

5977

5978

2006-10-25 Kevin McCullough <KMcCullough@apple.com>

5979

5980

Reviewed by Steve (rubber stamp).

5981

5982

- Link against your local build of JavaScriptCore.lib first, this fixes some errors on release builds of testkjs.

5983

5984

* JavaScriptCore.vcproj/testkjs/testkjs.vcproj:

5985

5986

2006-10-25 Geoffrey Garen <ggaren@apple.com>

5987

5988

Reviewed by Lou.

5989

5990

Removed duplicate symbol declaration.

5991

5992

* JavaScriptCore.xcodeproj/project.pbxproj:

5993

* kjs/grammar.y:

5994

5995

2006-10-24 Steve Falkenburg <sfalken@apple.com>

5996

5997

Build config change

5998

5999

* JavaScriptCore.vcproj/testkjs/testkjs.vcproj:

6000

6001

2006-10-24 Kevin McCullough <KMcCullough@apple.com>

6002

6003

Reviewed by Brady.

6004

6005

- Fixes a date formatting issue on win. Specifically strftime cannot handle some ranges of time so we shift time call strftime and then manipulate the returned string, if needed.

6006

6007

* kjs/date_object.cpp:

6008

(KJS::):

6009

(KJS::formatLocaleDate):

6010

(KJS::DateProtoFunc::callAsFunction):

6011

6012

2006-10-23 Kevin McCullough <KMcCullough@apple.com>

6013

6014

Reviewed by

6015

6016

- Build fix

6017

6018

* JavaScriptCore.xcodeproj/project.pbxproj:

6019

* kjs/grammar.y:

6020

6021

2006-10-23 Kevin McCullough <KMcCullough@apple.com>

6022

6023

Reviewed by Maciej.

6024

6025

- Makes the toTM function an operator. Was going to piggy back on a patch but the patch needs more work.

6026

6027

* kjs/DateMath.cpp:

6028

(KJS::equivalentYearForDST):

6029

* kjs/DateMath.h:

6030

(KJS::GregorianDateTime::operator tm):

6031

* kjs/date_object.cpp:

6032

(KJS::formatTime):

6033

(KJS::DateProtoFunc::callAsFunction):

6034

6035

2006-10-23 Kevin McCullough <KMcCullough@apple.com>

6036

6037

Reviewed by Maciej.

6038

6039

- Fixes two regressions on win. Both are stack overflows. For one the number of recursions is capped at 100, and for the other, nested parenthesis pairs are not evaluated (since they would evaluate to whatever is in them anyway).

6040

6041

* kjs/grammar.y:

6042

* kjs/object.cpp:

6043

6044

2006-10-21 Steve Falkenburg <sfalken@apple.com>

6045

6046

Reviewed by Adam.

6047

6048

Add minimal compatibility with MSVCRT leak checker

6049

6050

* wtf/FastMalloc.h:

6051

6052

2006-10-23 Kevin McCullough <KMcCullough@apple.com>

6053

6054

Reviewed by Geof.

6055

6056

- Sets the lowercase range correctly in the test and consolidates a variable to make the test more readable.

6057

6058

* tests/mozilla/ecma/String/15.5.4.11-2.js:

6059

6060

2006-10-21 Darin Adler <darin@apple.com>

6061

6062

Reviewed by Anders.

6063

6064

- http://bugs.webkit.org/show_bug.cgi?id=11377

6065

swap(Vector, Vector) should be O(1) instead of O(n)

6066

6067

* wtf/Vector.h:

6068

(WTF::VectorBuffer::swap): Added.

6069

(WTF::Vector::swap): Added.

6070

(WTF::swap): Added overload that takes two Vector objects.

6071

6072

2006-10-21 Darin Adler <darin@apple.com>

6073

6074

Reviewed by Adam.

6075

6076

- http://bugs.webkit.org/show_bug.cgi?id=11376

6077

build scripts should invoke make with "-j" option for multiple processors

6078

6079

* JavaScriptCore.xcodeproj/project.pbxproj: Pass -j `sysctl -n hw.ncpu` to make.

6080

6081

2006-10-19 Kevin McCullough <KMcCullough@apple.com>

6082

6083

Reviewed by Geof.

6084

6085

Changed test to make us pass Georgian case changing for Unicode 4.0 and 5.0. This incorporates changes from the 1.4 revision of the same mozilla test.

6086

On Tiger we are still using Unicode 4.0 but on win and Leopard we are using Unicode 5.0, so this test currently allows for either answer.

6087

6088

* tests/mozilla/ecma/String/15.5.4.11-2.js:

6089

6090

2006-10-18 Maciej Stachowiak <mjs@apple.com>

6091

6092

Reviewed by Geoff.

6093

6094

- remove vestiges of KXMLCore name (former name of WTF).

6095

6096

* wtf/Assertions.h:

6097

* wtf/FastMalloc.h:

6098

(operator new):

6099

(operator delete):

6100

(operator new[]):

6101

(operator delete[]):

6102

* wtf/FastMallocInternal.h:

6103

* wtf/Forward.h:

6104

* wtf/GetPtr.h:

6105

* wtf/HashCountedSet.h:

6106

* wtf/HashFunctions.h:

6107

* wtf/HashMap.h:

6108

* wtf/HashSet.h:

6109

* wtf/HashTable.h:

6110

* wtf/HashTraits.h:

6111

* wtf/ListRefPtr.h:

6112

* wtf/MathExtras.h:

6113

* wtf/Noncopyable.h:

6114

* wtf/OwnArrayPtr.h:

6115

* wtf/OwnPtr.h:

6116

* wtf/PassRefPtr.h:

6117

* wtf/Platform.h:

6118

* wtf/RefPtr.h:

6119

* wtf/StringExtras.h:

6120

(snprintf):

6121

* wtf/UnusedParam.h:

6122

* wtf/Vector.h:

6123

* wtf/VectorTraits.h:

6124

6125

2006-10-17 Steve Falkenburg <sfalken@apple.com>

6126

6127

Reviewed by Maciej.

6128

6129

Adjust include paths

6130

6131

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

6132

6133

2006-10-17 Kevin McCullough <KMcCullough@apple.com>

6134

6135

Reviewed by Darin.

6136

6137

Fixed a date issue where the UTC offset was not set in win.

6138

6139

* kjs/DateMath.cpp:

6140

(KJS::getDSTOffsetSimple):

6141

(KJS::getDSTOffset):

6142

(KJS::msToGregorianDateTime):

6143

* kjs/DateMath.h:

6144

(KJS::):

6145

(KJS::GregorianDateTime::GregorianDateTime):

6146

6147

2006-10-17 Kevin McCullough <KMcCullough@apple.com>

6148

6149

Reviewed by Brady.

6150

6151

Fixes a JavaScriptCore math issue on win.

6152

6153

* kjs/math_object.cpp:

6154

(MathFuncImp::callAsFunction):

6155

* wtf/MathExtras.h:

6156

(wtf_atan2):

6157

6158

2006-10-16 Kevin McCullough <kmccullough@apple.com>

6159

6160

Reviewed by Geof.

6161

6162

Removed unecessary global specifiers.

6163

6164

* kjs/math_object.cpp:

6165

(MathFuncImp::callAsFunction):

6166

6167

2006-10-16 Kevin McCullough <KMcCullough@apple.com>

6168

6169

Reviewed by John.

6170

6171

Fixes a compile order issue for testkjs on win.

6172

6173

* JavaScriptCore.vcproj/testkjs/testkjs.vcproj:

6174

6175

2006-10-15 Krzysztof Kowalczyk <kkowalczyk@gmail.com>

6176

6177

Reviewed by Anders.

6178

6179

Remove junk (as gcc calls it) after #else clause.

6180

6181

* wtf/FastMalloc.cpp:

6182

(WTF::do_free):

6183

6184

2006-10-14 Krzysztof Kowalczyk <kkowalczyk@gmail.com>

6185

6186

Reviewed by Maciej.

6187

6188

Define KXMLCORE_USE_CURL for platforms that wish to use CURL as

6189

networking, and set it for GDK build

6190

6191

* wtf/Platform.h:

6192

6193

2006-10-13 Marvin Decker <marv.decker@gmail.com>

6194

6195

Reviewed by Kevin McCullough.

6196

6197

Fixes http://bugs.webkit.org/show_bug.cgi?id=11283

6198

Fixes Qt/Linux and Windows build

6199

6200

* kjs/DateMath.cpp:

6201

* kjs/DateMath.h:

6202

* kjs/date_object.cpp:

6203

(KJS::DateProtoFunc::callAsFunction):

6204

6205

2006-10-13 Kevin McCullough <KMcCullough@apple.com>

6206

6207

Reviewed by Adam, Geoff, Darin.

6208

6209

Fixed displaying the UTC offset and time zone string, as well as renamed the GregorianDateTime structure and clean up.

6210

6211

* ChangeLog:

6212

* kjs/DateMath.cpp:

6213

(KJS::getUTCOffset):

6214

(KJS::getDSTOffsetSimple):

6215

(KJS::gregorianDateTimeToMS):

6216

(KJS::msToGregorianDateTime):

6217

* kjs/DateMath.h:

6218

(KJS::GregorianDateTime::GregorianDateTime):

6219

(KJS::GregorianDateTime::~GregorianDateTime):

6220

(KJS::GregorianDateTime::toTM):

6221

* kjs/date_object.cpp:

6222

(KJS::gmtoffset):

6223

(KJS::formatDate):

6224

(KJS::formatDateUTCVariant):

6225

(KJS::formatTime):

6226

(KJS::fillStructuresUsingTimeArgs):

6227

(KJS::fillStructuresUsingDateArgs):

6228

(KJS::DateInstance::getTime):

6229

(KJS::DateInstance::getUTCTime):

6230

(KJS::DateProtoFunc::callAsFunction):

6231

(KJS::DateObjectImp::construct):

6232

(KJS::DateObjectImp::callAsFunction):

6233

(KJS::DateObjectFuncImp::callAsFunction):

6234

(KJS::parseDate):

6235

* kjs/date_object.h:

6236

6237

2006-10-13 Kevin McCullough <KMcCullough@apple.com>

6238

6239

Reviewed by Adam.

6240

6241

Gets JavaScripCore tests running on windows.

6242

6243

* Scripts/run-javascriptcore-tests:

6244

* Scripts/webkitdirs.pm:

6245

6246

2006-10-12 Geoffrey Garen <ggaren@apple.com>

6247

6248

Reviewed by Maciej.

6249

6250

Removed JSObjectMakeWithPrototype, clarified some comments. We really

6251

don't want people to manage their own prototypes, so we don't want an

6252

extra function in the API devoted to just that. People can still manage

6253

their own prototypes if they really want by using JSObjectSetPrototype.

6254

6255

* API/JSClassRef.cpp:

6256

(OpaqueJSClass::createNoAutomaticPrototype):

6257

(OpaqueJSClass::create):

6258

* API/JSClassRef.h:

6259

* API/JSObjectRef.cpp:

6260

(JSClassCreate):

6261

(JSObjectMake):

6262

* API/JSObjectRef.h:

6263

* API/testapi.c:

6264

(main):

6265

* JavaScriptCore.exp:

6266

6267

2006-10-12 Kevin McCullough <KMcCullough@apple.com>

6268

6269

Reviewed by Adam.

6270

6271

Build breakage fix

6272

6273

* kjs/DateMath.cpp:

6274

(KJS::msToTM):

6275

* kjs/date_object.cpp:

6276

(KJS::gmtoffset):

6277

6278

2006-10-11 Kevin McCullough <KMcCullough@apple.com>

6279

6280

Reviewed by Geoff.

6281

6282

Added our own tm struct to have a consistent set of fields, which lets us display the DST offset and timezone strings correctly. Also there is some code cleanup.

6283

6284

* kjs/DateMath.cpp:

6285

(KJS::timeToMS):

6286

(KJS::getUTCOffset):

6287

(KJS::getDSTOffsetSimple):

6288

(KJS::dateToMS):

6289

(KJS::msToTM):

6290

(KJS::tmToKJStm):

6291

(KJS::KJStmToTm):

6292

* kjs/DateMath.h:

6293

* kjs/date_object.cpp:

6294

(KJS::gmtoffset):

6295

(KJS::formatTime):

6296

(KJS::DateProtoFunc::callAsFunction):

6297

(KJS::DateObjectImp::construct):

6298

(KJS::DateObjectImp::callAsFunction):

6299

(KJS::DateObjectFuncImp::callAsFunction):

6300

(KJS::parseDate):

6301

* kjs/date_object.h:

6302

6303

2006-10-09 Krzysztof Kowalczyk <kkowalczyk@gmail.com>

6304

6305

Reviewed by Geoff.

6306

6307

Improve gdk build compiler flags (show warning, no rtti and exceptions).

6308

6309

* jscore.bkl:

6310

6311

2006-10-06 Kevin McCullough <KMcCullough@apple.com>

6312

6313

Reviewed by Brady.

6314

6315

DST and TimeZones were wrong in some cases, specifically on some of the dates where DST changes.

6316

6317

* kjs/DateMath.cpp:

6318

(KJS::equivalentYearForDST):

6319

(KJS::getUTCOffset):

6320

(KJS::getDSTOffsetSimple):

6321

(KJS::getDSTOffset):

6322

(KJS::dateToMseconds):

6323

(KJS::msToTM):

6324

* kjs/DateMath.h:

6325

* kjs/date_object.cpp:

6326

(KJS::gmtoffset):

6327

6328

2006-10-05 Darin Adler <darin@apple.com>

6329

6330

Reviewed by Kevin McCullough.

6331

6332

* wtf/Assertions.cpp: Fix build when _DEBUG is not defined.

6333

6334

2006-10-04 Kevin McCullough <KMcCullough@apple.com>

6335

6336

Reviewed by Adam.

6337

6338

- Removed an unnecessary assert that was stopping many pages. tm_gmtoff was not set for UTC time in mozilla but is always set for us.

6339

6340

* kjs/DateMath.cpp:

6341

(KJS::getUTCOffset):

6342

(KJS::msToTM):

6343

* kjs/date_object.cpp:

6344

(KJS::gmtoffset):

6345

(KJS::formatTime):

6346

6347

2006-10-04 Geoffrey Garen <ggaren@apple.com>

6348

6349

Patch by Darin and me, reviewed by Maciej.

6350

6351

Fixed <rdar://problem/4518397> REGRESSION(?): Oft-seen but unrepro crash

6352

in JavaScript garbage collection (KJS::Collector::collect())

6353

<rdar://problem/4752492> Crash in KJS::collect

6354

6355

The issue here was allocating one garbage-collected object in the midst

6356

of allocating a second garbage-collected object. In such a case, the

6357

zeroIfFree word lies.

6358

6359

* kjs/collector.cpp:

6360

(KJS::Collector::allocate):

6361

(KJS::Collector::collect):

6362

6363

2006-10-04 Kevin McCullough <KMcCullough@apple.com>

6364

6365

Reviewed by Adam.

6366

6367

- Layout test fix

6368

6369

* kjs/DateMath.cpp:

6370

(KJS::dateToDayInYear): accept and correctly handle negative months

6371

6372

2006-10-05 Kevin McCullough <KMcCullough@apple.com>

6373

6374

build fix

6375

6376

* kjs/DateMath.cpp:

6377

(KJS::dateToDayInYear):

6378

6379

2006-10-05 Mark Rowe <bdash@webkit.org>

6380

6381

Reviewed by maculloch.

6382

6383

Gdk build fix.

6384

6385

* JavaScriptCoreSources.bkl: Add DateMath.cpp to file list.

6386

6387

2006-10-05 Kevin McCullough <KMcCullough@apple.com>

6388

6389

Reviewed by aroben

6390

6391

- build fix

6392

6393

* JavaScriptCore.xcodeproj/project.pbxproj:

6394

6395

2006-10-04 Nikolas Zimmermann <zimmermann@kde.org>

6396

6397

Reviewed by Mitz.

6398

6399

Fix Qt/Linux build by adding DateMath.cpp to compilation.

6400

6401

* CMakeLists.txt: Also replace tabs with spaces.

6402

6403

2006-10-04 Kevin McCullough <KMcCullough@apple.com>

6404

6405

Reviewed by DethBakin.

6406

6407

- Apparently the build bot uses an older version of XCode which warns about conversions and the newest version does not. I hope this fixes the build but I cann't be sure on my system.

6408

6409

* kjs/DateMath.cpp:

6410

(KJS::msToYear):

6411

(KJS::dayInYear):

6412

(KJS::dateToDayInYear):

6413

6414

2006-10-05 Darin Adler <darin@apple.com>

6415

6416

Reviewed by Adam.

6417

6418

* wtf/Assertions.cpp: Changed assertion formatting to omit the "======"

6419

lines so you can see more assertions in less space. Also improved format

6420

of file/line information so it works with more development environments.

6421

6422

2006-10-04 Kevin McCullough <KMcCullough@apple.com>

6423

6424

Reviewed by Tim H.

6425

6426

- The build machine is more sensitive about automatic conversions. These fixes exp

6427

licitly cast or change the input and return types of functions to avoid conversions.

6428

6429

* JavaScriptCore.xcodeproj/project.pbxproj:

6430

* kjs/DateMath.cpp:

6431

(KJS::):

6432

(KJS::msToDays):

6433

(KJS::msToYear):

6434

(KJS::dayInYear):

6435

(KJS::monthToDayInYear):

6436

(KJS::dateToDayInYear):

6437

(KJS::getDSTOffsetSimple):

6438

(KJS::getDSTOffset):

6439

(KJS::dateToMseconds):

6440

(KJS::msToTM):

6441

6442

2006-10-04 Kevin McCullough <KMcCullough@apple.com>

6443

6444

Reviewed by GGaren

6445

6446

- This is a big makeover for our Date implemenetation. This solves many platform specific issues, specifically dates before 1970, and simplifies some ugly code. The purpose of this was to get us to pass many of the JavaScriptCore tests on windows.

6447

6448

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

6449

* JavaScriptCore.xcodeproj/project.pbxproj:

6450

* kjs/DateMath.cpp: Added.

6451

(KJS::):

6452

(KJS::daysInYear):

6453

(KJS::daysFrom1970ToYear):

6454

(KJS::msFrom1970ToYear):

6455

(KJS::msToDays):

6456

(KJS::msToYear):

6457

(KJS::isLeapYear):

6458

(KJS::isInLeapYear):

6459

(KJS::dayInYear):

6460

(KJS::msToMilliseconds):

6461

(KJS::msToWeekDay):

6462

(KJS::msToSeconds):

6463

(KJS::msToMinutes):

6464

(KJS::msToHours):

6465

(KJS::msToMonth):

6466

(KJS::msToDayInMonth):

6467

(KJS::monthToDayInYear):

6468

(KJS::timeToMseconds):

6469

(KJS::dateToDayInYear):

6470

(KJS::equivalentYearForDST):

6471

(KJS::getUTCOffset):

6472

(KJS::getDSTOffsetSimple):

6473

(KJS::getDSTOffset):

6474

(KJS::localTimeToUTC):

6475

(KJS::UTCToLocalTime):

6476

(KJS::dateToMseconds):

6477

(KJS::msToTM):

6478

(KJS::isDST):

6479

* kjs/DateMath.h: Added.

6480

(KJS::):

6481

* kjs/date_object.cpp:

6482

(KJS::gmtoffset):

6483

(KJS::formatTime):

6484

(KJS::DateInstance::getTime):

6485

(KJS::DateInstance::getUTCTime):

6486

(KJS::DateProtoFunc::callAsFunction):

6487

(KJS::DateObjectImp::construct):

6488

(KJS::DateObjectFuncImp::callAsFunction):

6489

(KJS::parseDate):

6490

* kjs/testkjs.cpp:

6491

* os-win32/stdint.h:

6492

6493

2006-10-02 Nikolas Zimmermann <zimmermann@kde.org>

6494

6495

Reviewed/landed by Adam.

6496

6497

Build testkjs on Qt/Linux.

6498

6499

* CMakeLists.txt:

6500

6501

2006-10-02 Nikolas Zimmermann <zimmermann@kde.org>

6502

6503

Reviewed by eseidel. Landed by eseidel.

6504

6505

Fix win32 build, which has no inttypes.h

6506

6507

* wtf/Assertions.h:

6508

6509

2006-10-02 Nikolas Zimmermann <zimmermann@kde.org>

6510

6511

Reviewed by eseidel & mjs. Landed by eseidel.

6512

6513

Fix Qt/Linux build with older gcc 3.3.4.

6514

http://bugs.webkit.org/show_bug.cgi?id=11116

6515

6516

* kjs/lookup.h: Move cacheGlobalObject into KJS namespace.

6517

(KJS::cacheGlobalObject): Also remove GCC_ROOT_NS_HACK.

6518

* wtf/Assertions.h: Include inttypes.h for uintptr_t.

6519

6520

2006-09-28 Steve Falkenburg <sfalken@apple.com>

6521

6522

Reviewed by Maciej.

6523

6524

Use $(ConfigSuffix) set via vsprops files to add _debug

6525

to end of debug filenames.

6526

6527

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

6528

* JavaScriptCore.vcproj/debug.vsprops: Added.

6529

* JavaScriptCore.vcproj/dftables/dftables.vcproj:

6530

* JavaScriptCore.vcproj/release.vsprops: Added.

6531

* JavaScriptCore.vcproj/testkjs/testkjs.vcproj:

6532

6533

2006-09-28 Darin Adler <darin@apple.com>

6534

6535

Reviewed by Alice.

6536

6537

- support for change that should fix <rdar://problem/4733044>

6538

REGRESSION: XML iBench shows 10% perf. regression (copying

6539

strings while decoding)

6540

6541

* wtf/Vector.h: Changed VectorBuffer so that the general case

6542

contains an instance of the 0 case, since deriving from it

6543

was violating the Liskov Substitution Principle.

6544

(WTF::VectorBuffer::releaseBuffer): Added. Releases the buffer so it can

6545

be adopted by another data structure that uses the FastMalloc.h allocator.

6546

Returns 0 if the internal buffer was being used.

6547

(WTF::Vector::releaseBuffer): Added. Releases the buffer as above or creates

6548

a new one in the case where the internal buffer was being used.

6549

6550

2006-09-28 Maciej Stachowiak <mjs@apple.com>

6551

6552

Reviewed by Geoff.

6553

6554

- change garbage collection to happen at increments proportional to number of live objects, not always

6555

every 1000 allocations

6556

6557

* kjs/collector.cpp:

6558

(KJS::Collector::allocate):

6559

6560

2006-09-28 Maciej Stachowiak <mjs@apple.com>

6561

6562

Reviewed by Mitz.

6563

6564

- fixed REGRESSION (r16606): javascriptCore Crash on website load

6565

6566

Plus style fixes.

6567

6568

- fixed some possible off-by-one bugs

6569

- use indexing, not iterators, for Vectors

6570

- store Vector by pointer instead of by value to avoid blowing out FunctionImp size

6571

6572

* kjs/function.cpp:

6573

(KJS::FunctionImp::addParameter):

6574

(KJS::FunctionImp::parameterString):

6575

(KJS::FunctionImp::processParameters):

6576

(KJS::FunctionImp::lengthGetter):

6577

(KJS::FunctionImp::getParameterName):

6578

* kjs/function.h:

6579

6580

2006-09-27 Steve Falkenburg <sfalken@apple.com>

6581

6582

Reviewed by Maciej.

6583

6584

More build tweaks

6585

6586

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

6587

* JavaScriptCore.vcproj/JavaScriptCore/dstroot-to-sdk.cmd: Removed.

6588

6589

2006-09-27 John Sullivan <sullivan@apple.com>

6590

6591

* kjs/function.cpp:

6592

(KJS::FunctionImp::getParameterName):

6593

removed assertion that displeased gcc 4.0.1 (build 5420):

6594

ASSERT(static_cast<size_t>(index) == index);

6595

6596

2006-09-27 Kevin McCullough <KMcCullough@apple.com>

6597

6598

Reviewed by GGaren.

6599

6600

Cleanup of previous fix which was to address Radar: 4752492

6601

6602

* kjs/function.cpp:

6603

(KJS::FunctionImp::addParameter):

6604

(KJS::FunctionImp::parameterString):

6605

(KJS::FunctionImp::processParameters):

6606

(KJS::FunctionImp::lengthGetter):

6607

(KJS::FunctionImp::getParameterName):

6608

* kjs/function.h:

6609

6610

2006-09-27 Kevin McCullough <KMcCullough@apple.com>

6611

6612

Reviewed by Adele.

6613

6614

Fixes a GC stack overflow crash.

6615

The change is to move from a linked list implementation of Parameters to a Vector.

6616

The problem with the linked list is that each one creates it's own stack frame when being destroyed and in extreme cases this caused the stack to overflow.

6617

6618

* kjs/function.cpp:

6619

(KJS::Parameter::Parameter):

6620

(KJS::FunctionImp::addParameter):

6621

(KJS::FunctionImp::parameterString):

6622

(KJS::FunctionImp::processParameters):

6623

(KJS::FunctionImp::lengthGetter):

6624

(KJS::FunctionImp::getParameterName):

6625

* kjs/function.h:

6626

6627

2006-09-27 Steve Falkenburg <sfalken@apple.com>

6628

6629

Fix last path fix.

6630

6631

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

6632

6633

2006-09-27 Steve Falkenburg <sfalken@apple.com>

6634

6635

Set path before build.

6636

6637

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

6638

6639

2006-09-27 Sean Gies <seangies@apple.com>

6640

6641

Reviewed by Adam Roben.

6642

6643

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: Debug config should link to debug runtime.

6644

* JavaScriptCore.vcproj/dftables/dftables.vcproj: Debug config should link to debug runtime.

6645

6646

2006-09-27 Don Melton <gramps@apple.com>

6647

6648

Reviewed by Adam Roben.

6649

6650

Changed line ending from DOS to UNIX format so it doesn't die running

6651

on my machine. ;)

6652

6653

* JavaScriptCore.vcproj/JavaScriptCore/build-generated-files.sh:

6654

6655

2006-09-23 Alexey Proskuryakov <ap@nypop.com>

6656

6657

Reviewed by Maciej.

6658

6659

http://bugs.webkit.org/show_bug.cgi?id=10183

6660

REGRESSION: obfuscated JS decoding breaks because of soft hyphen removal

6661

(Fanfiction.net author pages not listing stories)

6662

6663

Rolled out the fix for bug 4139.

6664

6665

* kjs/lexer.cpp:

6666

(Lexer::setCode):

6667

(Lexer::shift):

6668

* tests/mozilla/ecma/Array/15.4.5.1-1.js:

6669

* tests/mozilla/expected.html:

6670

6671

2006-09-22 Steve Falkenburg <sfalken@apple.com>

6672

6673

Build fix.

6674

6675

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

6676

6677

2006-09-22 Darin Adler <darin@apple.com>

6678

6679

Reviewed by Alice.

6680

6681

* wtf/Vector.h: Add an append that takes a pointer and length.

6682

Generalize the existing Vector append to work on vectors with

6683

any value for inlineCapacity. Change the append algorithm so

6684

it doesn't check capacity each time through the loop.

6685

6686

2006-09-22 Steve Falkenburg <sfalken@apple.com>

6687

6688

Fix release build.

6689

6690

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

6691

6692

2006-09-21 Geoffrey Garen <ggaren@apple.com>

6693

6694

Reviewed by Maciej.

6695

6696

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

6697

Updated to include the right path.

6698

* wtf/FastMalloc.h: #include Platform.h, since we use Platform macros.

6699

6700

=== Safari-521.27 ===

6701

6702

2006-09-20 Anders Carlsson <acarlsson@apple.com>

6703

6704

Reviewed by Dave Hyatt.

6705

6706

* wtf/MathExtras.h:

6707

Get rid of lrint.

6708

6709

2006-09-20 Sean Gies <seangies@apple.com>

6710

6711

Reviewed by Steve Falkenburg.

6712

6713

* wtf/Assertions.cpp: Debug messages should go into debugger console.

6714

6715

2006-09-20 David Hyatt <hyatt@apple.com>

6716

6717

Add an implementation of lrint for Win32.

6718

6719

Reviewed by anders

6720

6721

* wtf/MathExtras.h:

6722

(lrint):

6723

6724

2006-09-15 Krzysztof Kowalczyk <kkowalczyk@gmail.com>

6725

6726

Reviewed by Adam.

6727

6728

http://bugs.webkit.org/show_bug.cgi?id=10864

6729

Bug 10864: Linux\GDK build fixes

6730

6731

* JavaScriptCoreSources.bkl:

6732

* jscore.bkl:

6733

6734

2006-09-15 Adam Roben <aroben@apple.com>

6735

6736

Windows build fix.

6737

6738

* JavaScriptCore.vcproj/JavaScriptCore/build-generated-files.sh:

6739

6740

2006-09-15 Anders Carlsson <acarlsson@apple.com>

6741

6742

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

6743

Fix the release build.

6744

6745

2006-09-15 Anders Carlsson <acarlsson@apple.com>

6746

6747

Reviewed by Steve.

6748

6749

Add JavaScriptCore API to the build.

6750

* API/JSBase.cpp:

6751

* API/JSCallbackConstructor.cpp:

6752

* API/JSCallbackFunction.cpp:

6753

* API/JSCallbackObject.cpp:

6754

* API/JSClassRef.cpp:

6755

* API/JSContextRef.cpp:

6756

* API/JSObjectRef.cpp:

6757

* API/JSStringRef.cpp:

6758

* API/JSValueRef.cpp:

6759

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

6760

* os-win32/stdbool.h: Added.

6761

6762

2006-09-12 Steve Falkenburg <sfalken@apple.com>

6763

6764

Reviewed by Ada.

6765

6766

Build tweaks (doing JavaScriptCore now since it doesn't have

6767

dependencies).

6768

6769

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

6770

* JavaScriptCore.vcproj/JavaScriptCore/build-generated-files.sh:

6771

* JavaScriptCore.vcproj/JavaScriptCore/dstroot-to-sdk.cmd: Added.

6772

* JavaScriptCore.vcproj/dftables/dftables.vcproj:

6773

* JavaScriptCore.vcproj/testkjs/testkjs.vcproj:

6774

6775

2006-09-11 Brady Eidson <beidson@apple.com>

6776

6777

Build fix - I think Tim's last checkin wasn't tested on Tiger, possibly. I simply

6778

commented out the undefined constants until he can have a chance to make the right call

6779

6780

* bindings/objc/objc_utility.mm:

6781

(KJS::Bindings::objcValueTypeForType): Commented out undefined symbols

6782

6783

2006-09-11 Timothy Hatcher <timothy@apple.com>

6784

6785

Reviewed by Tim O. and Darin.

6786

6787

Add support for more method signatures affecting ObjC methods called from JavaScript:

6788

- Added unsigned types and long long.

6789

- Allow methods that use const, oneway, bycopy and byref type modifiers.

6790

6791

* bindings/objc/objc_instance.mm:

6792

(ObjcInstance::invokeMethod):

6793

* bindings/objc/objc_utility.h:

6794

(KJS::Bindings::):

6795

* bindings/objc/objc_utility.mm:

6796

(KJS::Bindings::convertValueToObjcValue):

6797

(KJS::Bindings::convertObjcValueToValue):

6798

(KJS::Bindings::objcValueTypeForType):

6799

6800

2006-09-05 Timothy Hatcher <timothy@apple.com>

6801

6802

Reviewed by Tim O.

6803

6804

<rdar://problem/4715840> SEL is not char*

6805

6806

* bindings/objc/objc_class.mm:

6807

(KJS::Bindings::ObjcClass::methodsNamed): use sel_getName instead of a char* cast.

6808

* bindings/objc/objc_runtime.mm:

6809

(ObjcFallbackObjectImp::callAsFunction): ditto

6810

6811

2006-09-03 Alexey Proskuryakov <ap@nypop.com>

6812

6813

Reviewed by Tim H.

6814

6815

http://bugs.webkit.org/show_bug.cgi?id=10693

6816

Convert JavaScript arrays to AppleScript lists

6817

6818

* JavaScriptCore.exp: Export ArrayInstance::info and ArrayInstance::getItem().

6819

* kjs/array_instance.h:

6820

* kjs/array_object.cpp:

6821

(ArrayInstance::getItem): Added a method to access array items from C++.

6822

6823

2006-09-02 Krzysztof Kowalczyk <kkowalczyk@gmail.com>

6824

6825

Reviewed by Tim H.

6826

6827

Bug 10454: Unix bakefile fixes

6828

http://bugs.webkit.org/show_bug.cgi?id=10454

6829

6830

* JavaScriptCoreSources.bkl:

6831

6832

2006-09-01 Nikolas Zimmermann <zimmermann@kde.org>

6833

6834

Reviewed by hyatt. Landed by eseidel.

6835

6836

Fix build on Linux.

6837

6838

* pcre/CMakeLists.txt: Add wtf/ include.

6839

6840

2006-09-01 Nikolas Zimmermann <zimmermann@kde.org>

6841

6842

Reviewed and landed by ap.

6843

6844

Fix build on Linux (C89 without gcc extensions enabled).

6845

6846

* pcre/pcre_internal.h: Use C style comments.

6847

* wtf/Assertions.h: Use C style comments.

6848

* wtf/Platform.h: Use C style comments.

6849

6850

2006-09-01 Steve Falkenburg <sfalken@apple.com>

6851

6852

Fix build.

6853

6854

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

6855

* JavaScriptCore.vcproj/dftables/dftables.vcproj:

6856

6857

2006-08-31 Anders Carlsson <acarlsson@apple.com>

6858

6859

Reviewed by Darin.

6860

6861

Add new portability functions to MathExtras.h and add StringExtras.h which is for

6862

string portability functions.

6863

6864

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

6865

* bindings/c/c_instance.cpp:

6866

* kjs/date_object.cpp:

6867

* wtf/MathExtras.h:

6868

(copysign):

6869

(isfinite):

6870

* wtf/StringExtras.h: Added.

6871

(snprintf):

6872

(strncasecmp):

6873

6874

2006-08-31 Anders Carlsson <acarlsson@apple.com>

6875

6876

Reviewed by Tim H.

6877

6878

Fix Windows build.

6879

6880

* JavaScriptCore.vcproj/dftables/dftables.vcproj:

6881

* pcre/pcre_internal.h:

6882

6883

2006-08-31 Timothy Hatcher <timothy@apple.com>

6884

6885

Reviewed by Geoff.

6886

6887

Band-aid fix for PCRE to compile for ppc64 and x86_64 now that

6888

we use -Wshorten-64-to-32. Adds an INT_CAST macro that ASSERTs

6889

the value <= INT_MAX.

6890

6891

I filed <rdar://problem/4712064> to track the need to verify

6892

PCRE's 64-bit compliance.

6893

6894

* pcre/pcre_compile.c:

6895

(complete_callout):

6896

(compile_branch):

6897

(compile_regex):

6898

(pcre_compile2):

6899

* pcre/pcre_exec.c:

6900

(match):

6901

(pcre_exec):

6902

* pcre/pcre_get.c:

6903

(pcre_get_substring_list):

6904

* pcre/pcre_internal.h:

6905

* pcre/pcre_tables.c:

6906

* pcre/pcre_try_flipped.c:

6907

(_pcre_try_flipped):

6908

6909

2006-08-30 Darin Adler <darin@apple.com>

6910

6911

Reviewed by Tim Hatcher.

6912

6913

- add WTF::getPtr, a function template that makes it possible to write

6914

generic code that gets a raw pointer out of any of our pointer types

6915

6916

* JavaScriptCore.xcodeproj/project.pbxproj:

6917

* wtf/GetPtr.h: Added.

6918

* wtf/ListRefPtr.h: (WTF::getPtr): Added.

6919

* wtf/OwnArrayPtr.h: (WTF::getPtr): Added.

6920

* wtf/OwnPtr.h: (WTF::getPtr): Added.

6921

* wtf/PassRefPtr.h: (WTF::getPtr): Added.

6922

* wtf/RefPtr.h: (WTF::getPtr): Added.

6923

6924

2006-08-29 waylonis <waylonis@google.com>

6925

6926

Reviewed, tweaked by ggaren.

6927

6928

- Added storage and accessor functions for ExecState as a fix for

6929

http://bugs.webkit.org/show_bug.cgi?id=10114

6930

6931

* kjs/ExecState.cpp:

6932

(KJS::ExecState::ExecState):

6933

* kjs/ExecState.h:

6934

* kjs/context.h:

6935

(KJS::Context::setExecState):

6936

(KJS::Context::execState):

6937

6938

2006-08-30 Nikolas Zimmermann <zimmermann@kde.org>

6939

6940

Reviewed by Tim H.

6941

6942

Commit KDE related tweaks, to be able to

6943

differentiate between a Qt-only or a KDE build.

6944

6945

* CMakeLists.txt: Install wtf-unity library.

6946

* wtf/Platform.h: Add define for the KDE platform.

6947

6948

2006-08-28 Darin Adler <darin@apple.com>

6949

6950

Reviewed by Geoff.

6951

6952

* kjs/list.h: Use explicit in constructor (as appropriate).

6953

6954

2006-08-24 Nikolas Zimmermann <zimmermann@kde.org>

6955

6956

Reviewed, tweaked and landed by ap

6957

6958

http://bugs.webkit.org/show_bug.cgi?id=10467

6959

WebKit should have Qt platform support (Part II)

6960

6961

* CMakeLists.txt: Adjust to Anders' build fixes.

6962

* wtf/Platform.h: Fix define for the Qt platform (we don't use/need Cairo.)

6963

6964

2006-08-23 David Hyatt <hyatt@apple.com>

6965

6966

Fix Platform.h to include #defines for graphics features.

6967

6968

Reviewed by darin

6969

6970

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

6971

* wtf/Platform.h:

6972

6973

2006-08-23 Anders Carlsson <acarlsson@apple.com>

6974

6975

Reviewed by Darin.

6976

6977

Make the bindings compile without CoreFoundation.

6978

6979

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

6980

* bindings/c/c_instance.cpp:

6981

* bindings/c/c_utility.cpp:

6982

(KJS::Bindings::convertUTF8ToUTF16):

6983

* bindings/npapi.h:

6984

* bindings/runtime.cpp:

6985

(KJS::Bindings::Instance::createBindingForLanguageInstance):

6986

(KJS::Bindings::Instance::createLanguageInstanceForValue):

6987

* bindings/runtime_root.cpp:

6988

* bindings/runtime_root.h:

6989

* kjs/interpreter.cpp:

6990

(KJS::Interpreter::createLanguageInstanceForValue):

6991

* kjs/interpreter.h:

6992

6993

2006-08-22 Anders Carlsson <acarlsson@apple.com>

6994

6995

Reviewed by Darin.

6996

6997

Move the npruntime code over to using HashMap and the runtime_root code over to using

6998

HashMap and HashCountedSet.

6999

7000

* bindings/NP_jsobject.cpp:

7001

* bindings/c/c_utility.cpp:

7002

(KJS::Bindings::identifierFromNPIdentifier):

7003

* bindings/c/c_utility.h:

7004

* bindings/jni/jni_jsobject.cpp:

7005

(JavaJSObject::invoke):

7006

* bindings/npruntime.cpp:

7007

(getStringIdentifierMap):

7008

(getIntIdentifierMap):

7009

(_NPN_GetStringIdentifier):

7010

(_NPN_GetIntIdentifier):

7011

* bindings/runtime_root.cpp:

7012

(getReferencesByRootMap):

7013

(getReferencesSet):

7014

(KJS::Bindings::findReferenceSet):

7015

(KJS::Bindings::rootForImp):

7016

(KJS::Bindings::rootForInterpreter):

7017

(KJS::Bindings::addNativeReference):

7018

(KJS::Bindings::removeNativeReference):

7019

(RootObject::removeAllNativeReferences):

7020

* bindings/runtime_root.h:

7021

7022

2006-08-22 Anders Carlsson <acarlsson@apple.com>

7023

7024

Reviewed by Geoff.

7025

7026

Switch over the NPAPI and Java bindings to using HashMaps instead of dictionaries.

7027

7028

* JavaScriptCore.xcodeproj/project.pbxproj:

7029

* bindings/c/c_class.cpp:

7030

(KJS::Bindings::CClass::CClass):

7031

(KJS::Bindings::CClass::~CClass):

7032

(KJS::Bindings::CClass::classForIsA):

7033

(KJS::Bindings::CClass::methodsNamed):

7034

(KJS::Bindings::CClass::fieldNamed):

7035

* bindings/c/c_class.h:

7036

* bindings/jni/jni_class.cpp:

7037

(JavaClass::JavaClass):

7038

(JavaClass::~JavaClass):

7039

(JavaClass::methodsNamed):

7040

(JavaClass::fieldNamed):

7041

* bindings/jni/jni_class.h:

7042

* bindings/objc/objc_class.h:

7043

* bindings/objc/objc_class.mm:

7044

(KJS::Bindings::deleteMethod):

7045

(KJS::Bindings::deleteField):

7046

(KJS::Bindings::):

7047

(KJS::Bindings::ObjcClass::methodsNamed):

7048

(KJS::Bindings::ObjcClass::fieldNamed):

7049

* bindings/runtime.cpp:

7050

* bindings/runtime.h:

7051

* bindings/runtime_object.cpp:

7052

(RuntimeObjectImp::fieldGetter):

7053

(RuntimeObjectImp::getOwnPropertySlot):

7054

(RuntimeObjectImp::put):

7055

(RuntimeObjectImp::canPut):

7056

7057

2006-08-21 Vladimir Olexa <vladimir.olexa@gmail.com>

7058

7059

Reviewed by Darin.

7060

7061

http://bugs.webkit.org/show_bug.cgi?id=6252

7062

JavaScript 1.6 Array.lastIndexOf

7063

7064

Test: fast/js/array-lastIndexOf.html

7065

7066

* kjs/array_object.cpp:

7067

(ArrayProtoFunc::callAsFunction): Added a LastIndexOf case.

7068

* kjs/array_object.h:

7069

(KJS::ArrayProtoFunc::): Added LastIndexOf to enum.

7070

* tests/mozilla/expected.html: Two more tests now pass.

7071

7072

2006-08-20 Nikolas Zimmermann <zimmermann@kde.org>

7073

7074

Reviewed by Maciej. Landed by rwlbuis.

7075

7076

Fixes parts of: http://bugs.webkit.org/show_bug.cgi?id=10463

7077

WebKit should have Qt platform support

7078

7079

Removing obsolete QConstString/QString constructors in kjs code.

7080

7081

* kjs/identifier.h:

7082

* kjs/ustring.h:

7083

7084

2006-08-17 Nikolas Zimmermann <zimmermann@kde.org>

7085

7086

Reviewed by Maciej. Landed by rwlbuis.

7087

7088

Fixes: http://bugs.webkit.org/show_bug.cgi?id=10463

7089

WTF Changes needed for Qt platform code.

7090

7091

* wtf/Platform.h:

7092

* wtf/unicode/UnicodeDecomposition.h: Added.

7093

(WTF::Unicode::):

7094

* wtf/unicode/UnicodeDirection.h: Added.

7095

(WTF::Unicode::):

7096

* wtf/unicode/qt4/UnicodeQt4.cpp: Added.

7097

(WTF::Unicode::direction):

7098

(WTF::Unicode::category):

7099

(WTF::Unicode::decomposition):

7100

* wtf/unicode/qt4/UnicodeQt4.h:

7101

(WTF::Unicode::toLower):

7102

(WTF::Unicode::toUpper):

7103

(WTF::Unicode::isPrintableChar):

7104

(WTF::Unicode::isSpace):

7105

(WTF::Unicode::isPunct):

7106

(WTF::Unicode::isDigit):

7107

(WTF::Unicode::mirroredChar):

7108

(WTF::Unicode::compare):

7109

7110

2006-08-17 Nikolas Zimmermann <zimmermann@kde.org>

7111

7112

Reviewed by Eric. Landed by rwlbuis.

7113

7114

Fixes: http://bugs.webkit.org/show_bug.cgi?id=10464

7115

Offer a cmake build system for Qt platform.

7116

7117

* CMakeLists.txt: Added.

7118

* pcre/CMakeLists.txt: Added.

7119

7120

2006-08-17 Anders Carlsson <acarlsson@apple.com>

7121

7122

Reviewed by Maciej.

7123

7124

* bindings/npapi.h:

7125

Fix ifdef.

7126

7127

2006-08-15 Steve Falkenburg <sfalken@apple.com>

7128

7129

Reviewed by mjs.

7130

7131

Build fix.

7132

7133

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

7134

* wtf/Assertions.h:

7135

7136

2006-08-15 Mark Rowe <opendarwin.org@bdash.net.nz>

7137

7138

Reviewed by Tim H.

7139

7140

Build fix: DWARF and -gfull are incompatible with symbol separation.

7141

7142

* JavaScriptCore.xcodeproj/project.pbxproj:

7143

7144

2006-08-15 Mark Rowe <opendarwin.org@bdash.net.nz>

7145

7146

Reviewed by Tim H.

7147

7148

http://bugs.webkit.org/show_bug.cgi?id=10394

7149

Bug 10394: WebKit Release and Production configurations should enable dead code stripping

7150

7151

* JavaScriptCore.xcodeproj/project.pbxproj:

7152

7153

2006-08-15 Mark Rowe <opendarwin.org@bdash.net.nz>

7154

7155

Reviewed by Tim H.

7156

7157

http://bugs.webkit.org/show_bug.cgi?id=10384

7158

Bug 10384: Switch to DWARF for Release configuration

7159

7160

* JavaScriptCore.xcodeproj/project.pbxproj:

7161

7162

2006-08-13 Maks Orlovich <maksim@kde.org>

7163

7164

Reviewed (and tweaked a little) by Maciej.

7165

7166

- shrank the size of JSObject by 8 bytes and made the corresponding reduction to the cell size, resulting

7167

in a 1.2% speed improvement on JS iBench (and probably overall memory savings).

7168

7169

This was done by removing _scope and _internalValue data members

7170

from JSObject and moving them only to the subclasses that actually

7171

make use of them.

7172

7173

* kjs/object.cpp:

7174

(KJS::JSObject::mark): No need to mark scope or internal value here.

7175

* kjs/object.h:

7176

(KJS::JSObject::JSObject): Don't initialize them.

7177

* kjs/JSWrapperObject.cpp: Added. New base class for object types that

7178

wrap primitive values (Number, String, Boolean, Date).

7179

(KJS::JSWrapperObject::mark):

7180

* kjs/JSWrapperObject.h: Added.

7181

(KJS::JSWrapperObject::JSWrapperObject):

7182

(KJS::JSWrapperObject::internalValue):

7183

(KJS::JSWrapperObject::setInternalValue):

7184

* kjs/array_object.cpp:

7185

(ArrayPrototype::ArrayPrototype): Don't set useless internal value.

7186

* kjs/bool_object.cpp:

7187

(BooleanInstance::BooleanInstance): Inherit from JSWrapperObject.

7188

(BooleanProtoFunc::callAsFunction): Fixed to account for fact that not all

7189

JSObjects have an internal value.

7190

(BooleanObjectImp::construct): ditto.

7191

* kjs/bool_object.h:

7192

* kjs/collector.cpp: Lowered cell size to 48.

7193

(KJS::Collector::allocate): meaningless whitespace change

7194

* kjs/date_object.cpp:

7195

(KJS::DateInstance::DateInstance): Inherit from JSWrapperObject.

7196

(KJS::DateProtoFunc::callAsFunction): adjusted for move of internalValue

7197

(KJS::DateObjectImp::construct): ditto

7198

* kjs/date_object.h:

7199

* kjs/error_object.cpp:

7200

(ErrorPrototype::ErrorPrototype): don't set internal value

7201

* kjs/function.cpp: move _scope and related handling here

7202

(KJS::FunctionImp::mark): mark scope

7203

* kjs/function.h:

7204

(KJS::FunctionImp::scope): moved here from JSObject

7205

(KJS::FunctionImp::setScope): ditto

7206

* kjs/number_object.cpp:

7207

(NumberInstance::NumberInstance): inherit from JSWrapperObject

7208

(NumberProtoFunc::callAsFunction): adjusted

7209

(NumberObjectImp::construct): adjusted

7210

* kjs/number_object.h: shring RegExp-related objects a little

7211

* kjs/regexp_object.cpp:

7212

(RegExpPrototype::RegExpPrototype): Adjust for size tweaks

7213

(RegExpObjectImp::RegExpObjectImp): ditto

7214

* kjs/regexp_object.h:

7215

* kjs/string_object.cpp:

7216

(StringInstance::StringInstance): inherit from JSWrapperObject

7217

(StringProtoFunc::callAsFunction): adjusted

7218

* kjs/string_object.h:

7219

* JavaScriptCore.exp: Exported new methods as needed.

7220

* JavaScriptCore.xcodeproj/project.pbxproj: Added new files to build.

7221

7222

2006-08-04 Brady Eidson <beidson@apple.com>

7223

7224

Reviewed by Geoff's rubber stamp

7225

7226

Fix a build break on Intel hardware causes by adapting stricter

7227

compiler warnings (-Wshorten-64-to-32)

7228

7229

* API/testapi.c:

7230

(assertEqualsAsNumber): manually cast some doubles to floats

7231

(main): ditto

7232

7233

2006-08-04 Sam Weinig <sam.weinig@gmail.com>

7234

7235

Reviewed by Darin.

7236

7237

- patch for http://bugs.webkit.org/show_bug.cgi?id=10192

7238

Make WebCore (and friends) compile with -Wshorten-64-to-32

7239

7240

* Adds -Wshorten-64-to-32 flag to Xcode project.

7241

* Adds explicit casts where OK.

7242

7243

* API/JSNodeList.c:

7244

(JSNodeList_item):

7245

(JSNodeList_getProperty):

7246

* JavaScriptCore.xcodeproj/project.pbxproj:

7247

7248

2006-08-04 Adam Roben <aroben@apple.com>

7249

7250

Reviewed by Anders.

7251

7252

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: Convert

7253

spaces to tabs

7254

7255

2006-08-03 Sam Weinig <sam.weinig@gmail.com>

7256

7257

Reviewed by Darin.

7258

7259

- patch for http://bugs.webkit.org/show_bug.cgi?id=10176

7260

Make WebCore compile with -Wundef

7261

7262

* Adds -Wundef flag to Xcode project

7263

* Converts #ifs to #ifdef and #ifndefs where needed.

7264

* Added #define YYMAXDEPTH 10000 in kjs/grammar.y

7265

to fix a warning from within Bison.

7266

7267

* JavaScriptCore.xcodeproj/project.pbxproj:

7268

* bindings/jni/jni_jsobject.cpp:

7269

(JavaJSObject::getSlot):

7270

(JavaJSObject::setSlot):

7271

* bindings/npapi.h:

7272

* bindings/objc/objc_class.mm:

7273

(KJS::Bindings::ObjcClass::methodsNamed):

7274

(KJS::Bindings::ObjcClass::fieldNamed):

7275

* bindings/objc/objc_instance.mm:

7276

(ObjcInstance::invokeMethod):

7277

* bindings/objc/objc_runtime.mm:

7278

(ObjcMethod::getMethodSignature):

7279

(ObjcField::name):

7280

(ObjcField::type):

7281

* kjs/grammar.y:

7282

* kjs/identifier.h:

7283

7284

2006-08-03 Anders Carlsson <acarlsson@apple.com>

7285

7286

Reviewed by John Sullivan.

7287

7288

* wtf/HashSet.h:

7289

(WTF::::operator):

7290

Return *this in operator=

7291

7292

2006-08-03 Adam Roben <aroben@apple.com>

7293

7294

Reviewed by Anders.

7295

7296

- Fixed Windows build

7297

7298

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

7299

* wtf/MathExtras.h: Implement inline versions of these functions

7300

(nextafter):

7301

(nextafterf):

7302

7303

2006-08-02 Adam Roben <aroben@apple.com>

7304

7305

Reviewed by Darin.

7306

7307

- Fixed build

7308

7309

* kjs/date_object.cpp:

7310

(KJS::formatTime):

7311

7312

2006-07-29 Darin Adler <darin@apple.com>

7313

7314

- Removed tabs from these source files that still had them.

7315

We don't use them; that way source files look fine in editors

7316

that have tabs set to 8 spaces or to 4 spaces.

7317

- Removed allow-tabs Subversion property from the files too.

7318

7319

* bindings/NP_jsobject.cpp:

7320

* bindings/c/c_utility.cpp:

7321

* bindings/jni/jni_runtime.cpp:

7322

* bindings/jni/jni_utility.cpp:

7323

* bindings/objc/objc_utility.mm:

7324

* bindings/runtime.cpp:

7325

* bindings/runtime_method.cpp:

7326

* bindings/testbindings.cpp:

7327

* bindings/testbindings.mm:

7328

* kjs/date_object.cpp:

7329

* kjs/function.cpp:

7330

* kjs/list.cpp:

7331

* kjs/nodes.cpp:

7332

* kjs/nodes.h:

7333

* kjs/string_object.cpp:

7334

* kjs/ustring.cpp:

7335

7336

2006-07-29 Darin Adler <darin@apple.com>

7337

7338

* tests/mozilla/expected.html: Update test results now that regress-185165.js

7339

is succeeding. I suspect Anders fix for bug 4620655 is the reason.

7340

7341

2006-07-29 Sam Weinig <sam.weinig@gmail.com>

7342

7343

Reviewed by Darin.

7344

7345

- patch for http://bugs.webkit.org/show_bug.cgi?id=10080

7346

Adopt pedantic changes from the Unity project to improve

7347

cross-compiler compatibility

7348

7349

Changes include:

7350

* Removing trailing semicolon from namespace braces.

7351

* Removing trailing comma from last enum declaration.

7352

* Updating to match style guidelines.

7353

* Adding missing newline to the end of the file.

7354

* Turning on gcc warning for missing newline at the end of a source file

7355

(GCC_WARN_ABOUT_MISSING_NEWLINE in Xcode, -Wnewline in gcc).

7356

* Alphabetical sorting of Xcode source list files.

7357

* Replace use of non-portable variable-size array with Vector.

7358

* Use C-style comments instead of C++ comments in files that might

7359

be included by either C or C++ files.

7360

7361

* API/JSCallbackConstructor.cpp:

7362

(KJS::JSCallbackConstructor::construct):

7363

* API/JSCallbackFunction.cpp:

7364

(KJS::JSCallbackFunction::callAsFunction):

7365

* API/JSCallbackObject.cpp:

7366

(KJS::JSCallbackObject::construct):

7367

(KJS::JSCallbackObject::callAsFunction):

7368

* JavaScriptCore.xcodeproj/project.pbxproj:

7369

* JavaScriptCorePrefix.h:

7370

* bindings/jni/jni_class.cpp:

7371

(JavaClass::fieldNamed):

7372

* bindings/jni/jni_class.h:

7373

* bindings/jni/jni_instance.cpp:

7374

(JavaInstance::JavaInstance):

7375

(JavaInstance::valueOf):

7376

* bindings/jni/jni_objc.mm:

7377

(KJS::Bindings::dispatchJNICall):

7378

* bindings/jni/jni_runtime.cpp:

7379

(JavaParameter::JavaParameter):

7380

(JavaArray::JavaArray):

7381

* bindings/jni/jni_runtime.h:

7382

* bindings/jni/jni_utility.h:

7383

* bindings/objc/objc_instance.h:

7384

* bindings/runtime_array.h:

7385

* kjs/collector.h:

7386

* kjs/config.h:

7387

* kjs/ustring.cpp:

7388

* wtf/Platform.h:

7389

7390

2006-07-29 Mike Emmel <mike.emmel@gmail.com>

7391

7392

Reviewed by Darin.

7393

7394

- fixes for Linux build

7395

7396

* JavaScriptCoreSources.bkl: Added new files to build, kjs/PropertyNameArray.cpp

7397

and kjs/testkjs.cpp, and removed old files.

7398

7399

2006-07-24 Dan Waylonis <waylonis@google.com>

7400

7401

Reviewed and tweaked a bit by Darin.

7402

7403

- fix http://bugs.webkit.org/show_bug.cgi?id=9902

7404

jsNull and NSNull not properly converted between JS and ObjC

7405

7406

* bindings/objc/objc_utility.mm: (KJS::Bindings::convertObjcValueToValue):

7407

Added case for converting NSNull to jsNull.

7408

7409

2006-07-24 Rob Buis <buis@kde.org>

7410

7411

Reviewed by Darin.

7412

7413

http://bugs.webkit.org/show_bug.cgi?id=4258

7414

Date().toString() only includes GMT offset, not timezone string

7415

7416

Use the info in tm_zone to append timezone abbreviation

7417

to Date().toString().

7418

7419

* kjs/date_object.cpp:

7420

(KJS::formatTime):

7421

7422

2006-07-24 Rob Buis <buis@kde.org>

7423

7424

Reviewed by Darin.

7425

7426

http://bugs.webkit.org/show_bug.cgi?id=5257

7427

setYear() does not match FireFox/IE behavior

7428

7429

Make sure the right values end up in tm_year.

7430

7431

* kjs/date_object.cpp:

7432

(KJS::formatTime):

7433

7434

2006-07-23 Mark Rowe <opendarwin.org@bdash.net.nz>

7435

7436

Reviewed by Maciej.

7437

7438

Bug 9686: [Drosera] Need the ability to break into Drosera on Javascript exceptions

7439

http://bugs.webkit.org/show_bug.cgi?id=9686

7440

7441

JavaScriptCore portion of the fix.

7442

7443

* JavaScriptCore.exp: Update symbol for change in argument type.

7444

* kjs/debugger.cpp:

7445

(Debugger::detach): Clear map of recent exceptions.

7446

(Debugger::hasHandledException): Track the most recent exception

7447

thrown by an interpreter.

7448

(Debugger::exception): Change exception argument to a JSValue.

7449

* kjs/debugger.h:

7450

* kjs/nodes.cpp:

7451

(Node::debugExceptionIfNeeded): Notify the debugger of an exception

7452

if it hasn't seen it before.

7453

(ThrowNode::execute): Notify the debugger that an exception is being thrown.

7454

* kjs/nodes.h:

7455

7456

2006-07-23 Geoffrey Garen <ggaren@apple.com>

7457

7458

Patch by Eric Albert, reviewed by Darin and me.

7459

7460

- Fixed <rdar://problem/4645931> JavaScriptCore stack-scanning code

7461

crashes (Collector::markStackObjectsConservatively)

7462

7463

* bindings/jni/jni_jsobject.cpp: On 64bit systems, jint is a long, not an

7464

int.

7465

(JavaJSObject::getSlot):

7466

(JavaJSObject::setSlot):

7467

* kjs/collector.cpp:

7468

(KJS::Collector::markCurrentThreadConservatively): Use a pointer instead of

7469

an int as 'dummy,' because on LP64 systems, an int is not pointer-aligned,

7470

and we want to scan the stack for pointers.

7471

* JavaScriptCore.xcodeproj/project.pbxproj: After a tense cease-fire, the

7472

XCode war has started up again!

7473

7474

=== Safari-521.20 ===

7475

7476

2006-07-21 Geoffrey Garen <ggaren@apple.com>

7477

7478

Reviewed by Darin.

7479

7480

<rdar://problem/4507265> REGRESSION: overlays don't work on HousingMaps.com (Google Maps-based site)

7481

7482

- Added support for strings that masquerade as undefined. Currently used

7483

by WebCore to implement undetectable style.filter.

7484

7485

The name is a little long, but it's only used in one line of code, so I

7486

thought clarity should win over brevity.

7487

7488

* JavaScriptCore.exp:

7489

* JavaScriptCore.xcodeproj/project.pbxproj:

7490

* kjs/object.h:

7491

* kjs/string_object.h:

7492

(KJS::StringInstanceThatMasqueradesAsUndefined::StringInstanceThatMasqueradesAsUndefined):

7493

(KJS::StringInstanceThatMasqueradesAsUndefined::masqueradeAsUndefined):

7494

(KJS::StringInstanceThatMasqueradesAsUndefined::toBoolean):

7495

7496

=== Safari-521.19 ===

7497

7498

2006-07-20 Steve Falkenburg <sfalken@apple.com>

7499

7500

Fix the build

7501

7502

* kjs/function.cpp:

7503

(KJS::escapeStringForPrettyPrinting):

7504

7505

2006-07-19 Anders Carlsson <acarlsson@apple.com>

7506

7507

Reviewed by Darin.

7508

7509

<rdar://problem/4620655> REGRESSION(10.4.7-10.5): preview button for a blogger.com post doesn't work

7510

7511

* kjs/nodes2string.cpp:

7512

(StringNode::streamTo):

7513

Return the escaped string.

7514

7515

(RegExpNode::streamTo):

7516

Use the correct syntax.

7517

7518

* kjs/function.cpp:

7519

(KJS::escapeStringForPrettyPrinting):

7520

* kjs/function.h:

7521

Add escape function which escapes a string for pretty-printing so it can be parsed again.

7522

7523

* wtf/unicode/icu/UnicodeIcu.h:

7524

(WTF::Unicode::isPrintableChar):

7525

New function.

7526

7527

2006-07-18 Maciej Stachowiak <mjs@apple.com>

7528

7529

Reviewed by Adele Peterson.

7530

7531

<rdar://problem/4589530> REGRESSION: null character in JS string causes parse error (works in Tiger and in other browsers)

7532

7533

* kjs/lexer.cpp:

7534

(Lexer::shift):

7535

(Lexer::lex):

7536

(Lexer::record16):

7537

(Lexer::scanRegExp):

7538

* kjs/lexer.h:

7539

7540

2006-07-18 Tim Omernick <timo@apple.com>

7541

7542

Reviewed by Tim Hatcher.

7543

7544

Removed a misleading comment; we recently added support for the NPNVPluginElementNPObject

7545

variable.

7546

7547

* bindings/npapi.h:

7548

7549

=== Safari-521.18 ===

7550

7551

2006-07-18 Timothy Hatcher <timothy@apple.com>

7552

7553

Made the following headers public:

7554

7555

* JavaScriptCore.h

7556

* JSBase.h

7557

* JSContextRef.h

7558

* JSObjectRef.h

7559

* JSStringRef.h

7560

* JSValueRef.h

7561

7562

* JavaScriptCore.xcodeproj/project.pbxproj:

7563

7564

2006-07-17 Geoffrey Garen <ggaren@apple.com>

7565

7566

Reviewed by Maciej.

7567

7568

- Added automatic prototype creation for classes.

7569

7570

A class stores a weak reference to a prototype, which is cleared when

7571

the prototype is garbage collected, to avoid a reference cycle.

7572

7573

We now have an attributes field in JSClassDefinition, that currently is

7574

used only to override automatic prototype creation when you want to manage your

7575

own prototypes, but can be extended in the future for other nefarious purposes.

7576

7577

Similarly, we have JSObjectMake and JSObjectMakeWithPrototype, the latter

7578

allowing you to manage your own prototypes.

7579

7580

JSObjectMakeConstructor is more interesting now, able to make a constructor

7581

on your behalf if you just give it a class.

7582

7583

- Removed bogus old code from minidom.js.

7584

7585

- Tweaked the headerdocs.

7586

7587

- Added more GC testing, which caught some leaks, and tested more funny

7588

edge cases in lookup, which caught a lookup bug. Removed some testing

7589

we used to do with MyObject because it was redundant with the new, cool

7590

stuff.

7591

7592

While fixing the lookup bug I retracted this change:

7593

7594

"If a static setProperty callback returns 'false', to indicate that the

7595

property was not set, we no longer forward the set request up the class

7596

chain, because that's almost certainly not what the programmer expected."

7597

7598

Returning false when setting a static property is a little silly, but you can see

7599

it being useful when shadowing a base class's static properties, and, regardless

7600

of usefullness, this is the defined behavior of the setProperty callback.

7601

7602

- Plus a little ASCII art, for the kids.

7603

7604

2006-07-17 Timothy Hatcher <timothy@apple.com>

7605

7606

Reviewed by Maciej.

7607

7608

<rdar://problem/4634874> WebScriptObject and WebUndefined are no longer defined by WebKit

7609

7610

Moves WebScriptObject and WebUndefined up to WebCore.

7611

This change does create an upwards-dependancy on WebScriptObject existing

7612

in the loaded process, but this code path in JavaScriptCore does not get used

7613

unless it is through WebKit/WebCore. Moving all of the binding code out of

7614

JavaScriptCore might make sense in the future.

7615

7616

* JavaScriptCore.exp:

7617

* JavaScriptCore.xcodeproj/project.pbxproj:

7618

* bindings/objc/WebScriptObject.h: Replaced.

7619

* bindings/objc/WebScriptObject.mm: Removed.

7620

* bindings/objc/WebScriptObjectPrivate.h: Removed.

7621

* bindings/objc/objc_class.h:

7622

* bindings/objc/objc_instance.h:

7623

* bindings/objc/objc_instance.mm:

7624

(ObjcInstance::~ObjcInstance):

7625

* bindings/objc/objc_runtime.h:

7626

* bindings/objc/objc_runtime.mm:

7627

(convertValueToObjcObject):

7628

* bindings/objc/objc_utility.mm:

7629

(KJS::Bindings::convertValueToObjcValue):

7630

(KJS::Bindings::convertObjcValueToValue):

7631

(KJS::Bindings::createObjcInstanceForValue):

7632

7633

2006-07-17 Darin Adler <darin@apple.com>

7634

7635

* API/JSBase.h: Fix comment formatting where things used to be lined up but

7636

are now ragged. Got rid of spaces that attempted to line things up.

7637

* API/JSObjectRef.h: Ditto. Also add missing periods for a couple of comments.

7638

7639

2006-07-17 Geoffrey Garen <ggaren@apple.com>

7640

7641

Reviewed by Maciej.

7642

7643

- Removed the exception parameter from the initialize callback and, by extension,

7644

JSObjectMake. We have never had a need for exceptions when iniitializing,

7645

so the parameter seemed likely to "get in the way."

7646

7647

Also, an exception in JavaScript is thrown in response to input --

7648

"invalid URL", "index not a number", etc., so it's the job of the

7649

constructor function, not the initialize method, to throw.

7650

7651

If initialize *really* wants to throw, it can communicate the throw to

7652

the constructor through the constructed object's private data (e.g., set

7653

it to NULL, signaling to the consntructor that initialization failed.)

7654

7655

- Added JSObjectMakeWithData, which enables a constructor to set private

7656

data on an object *before* it has been initialized. That way, the initialize

7657

methods can properly operate on the data.

7658

7659

* API/JSNode.c: Moved ref into the initialize method, for better encapsulation,

7660

now that it's possible.

7661

* API/JSNodeList.c: ditto

7662

* API/minidom.c:

7663

(main): Do more aggressive garbage collection to test ref/deref and

7664

initialize/finalize.

7665

* API/minidom.js: store childNodes in a temporary so it doesn't get re-created

7666

like a thousand times. This makes debugging ref/deref easier

7667

7668

2006-07-17 Geoffrey Garen <ggaren@apple.com>

7669

7670

Reviewed by Maciej.

7671

7672

- Changed the initialize callback to run from least derived class (parent

7673

class) to most derived class. This enables C++ style initialization,

7674

and derived class overriding of member data.

7675

7676

- Added excpetion propopgation to JSObjectMake, to support initialize

7677

exceptions, and generally round out our policy of making function

7678

signatures as long as possible.

7679

7680

* API/JSCallbackObject.h: Use ExecState instead of ContextRef, cuz we're

7681

in C++ land now.

7682

7683

2006-07-17 Geoffrey Garen <ggaren@apple.com>

7684

7685

Reviewed by Maciej.

7686

7687

- Changed JSObjectMakeConstructor to JSObjectMakeConstructorWithCallback,

7688

to match JSObjectMakeFunctionWithCallback.

7689

7690

- Added prototype parameter, so the generated constructor

7691

automatically works with hasInstance / instanceof

7692

7693

- Moved hasInstance implementation from InternalFunctionImp to JSObject

7694

so that subclasses can inherit it without inheriting function-related baggage.

7695

More refactoring here would be good, but this seems like a good short-term

7696

solution.

7697

7698

(KJS::JSCallbackFunction::implementsHasInstance): override and return false,

7699

because callback functions aren't constructors.

7700

7701

2006-07-17 Maciej Stachowiak <mjs@apple.com>

7702

7703

Reviewed by Geoff.

7704

7705

- add a JSContextRef parameter to all JSValueRef, JSObjectRef, and JSContextRef operations;

7706

except JSObject{Get,Set}PrivateData which can be assumed to be simple pure accessors.

7707

7708

Also renamed the parameter "context" to "ctx" because it makes the code read better with this pervasive

7709

but usually uninteresting parameter.

7710

7711

* API/JSBase.cpp:

7712

(JSEvaluateScript):

7713

(JSCheckScriptSyntax):

7714

(JSGarbageCollect):

7715

* API/JSBase.h:

7716

* API/JSCallbackObject.cpp:

7717

(KJS::JSCallbackObject::JSCallbackObject):

7718

(KJS::JSCallbackObject::init):

7719

(KJS::JSCallbackObject::getOwnPropertySlot):

7720

(KJS::JSCallbackObject::put):

7721

(KJS::JSCallbackObject::deleteProperty):

7722

(KJS::JSCallbackObject::toNumber):

7723

(KJS::JSCallbackObject::toString):

7724

* API/JSContextRef.cpp:

7725

(JSGlobalContextCreate):

7726

(JSGlobalContextRetain):

7727

(JSGlobalContextRelease):

7728

(JSContextGetGlobalObject):

7729

* API/JSContextRef.h:

7730

* API/JSNode.c:

7731

(JSNodePrototype_appendChild):

7732

(JSNodePrototype_removeChild):

7733

(JSNodePrototype_replaceChild):

7734

(JSNode_getNodeType):

7735

(JSNode_getFirstChild):

7736

(JSNode_prototype):

7737

* API/JSNodeList.c:

7738

(JSNodeListPrototype_item):

7739

(JSNodeList_length):

7740

(JSNodeList_getProperty):

7741

(JSNodeList_prototype):

7742

* API/JSObjectRef.cpp:

7743

(JSObjectMake):

7744

(JSObjectMakeFunctionWithCallback):

7745

(JSObjectMakeConstructor):

7746

(JSObjectMakeFunction):

7747

(JSObjectGetPrototype):

7748

(JSObjectSetPrototype):

7749

(JSObjectHasProperty):

7750

(JSObjectGetProperty):

7751

(JSObjectSetProperty):

7752

(JSObjectGetPropertyAtIndex):

7753

(JSObjectSetPropertyAtIndex):

7754

(JSObjectDeleteProperty):

7755

(JSObjectIsFunction):

7756

(JSObjectCallAsFunction):

7757

(JSObjectIsConstructor):

7758

(JSObjectCallAsConstructor):

7759

(JSObjectCopyPropertyNames):

7760

* API/JSObjectRef.h:

7761

* API/JSStringRef.cpp:

7762

* API/JSValueRef.cpp:

7763

(JSValueGetType):

7764

(JSValueIsUndefined):

7765

(JSValueIsNull):

7766

(JSValueIsBoolean):

7767

(JSValueIsNumber):

7768

(JSValueIsString):

7769

(JSValueIsObject):

7770

(JSValueIsObjectOfClass):

7771

(JSValueIsEqual):

7772

(JSValueIsStrictEqual):

7773

(JSValueIsInstanceOfConstructor):

7774

(JSValueMakeUndefined):

7775

(JSValueMakeNull):

7776

(JSValueMakeBoolean):

7777

(JSValueMakeNumber):

7778

(JSValueMakeString):

7779

(JSValueToBoolean):

7780

(JSValueToNumber):

7781

(JSValueToStringCopy):

7782

(JSValueToObject):

7783

(JSValueProtect):

7784

(JSValueUnprotect):

7785

* API/JSValueRef.h:

7786

* API/minidom.c:

7787

(print):

7788

* API/testapi.c:

7789

(MyObject_getProperty):

7790

(MyObject_deleteProperty):

7791

(MyObject_callAsFunction):

7792

(MyObject_callAsConstructor):

7793

(MyObject_convertToType):

7794

(print_callAsFunction):

7795

(main):

7796

7797

2006-07-16 Geoffrey Garen <ggaren@apple.com>

7798

7799

Approved by Maciej, RS by Beth.

7800

7801

JSObjectMakeFunction -> JSObjectMakeFunctionWithCallback

7802

JSObjectMakeFunctionWithBody -> JSObjectMakeFunction

7803

7804

because the latter is more common, and more fundamental, than the former.

7805

7806

* API/APICast.h:

7807

(toJS):

7808

* API/JSBase.h:

7809

* API/JSCallbackObject.cpp:

7810

(KJS::JSCallbackObject::getOwnPropertySlot):

7811

(KJS::JSCallbackObject::put):

7812

(KJS::JSCallbackObject::deleteProperty):

7813

(KJS::JSCallbackObject::getPropertyNames):

7814

(KJS::JSCallbackObject::staticValueGetter):

7815

(KJS::JSCallbackObject::staticFunctionGetter):

7816

* API/JSClassRef.cpp:

7817

(OpaqueJSClass::OpaqueJSClass):

7818

(OpaqueJSClass::~OpaqueJSClass):

7819

* API/JSClassRef.h:

7820

* API/JSObjectRef.cpp:

7821

(JSClassCreate):

7822

(JSObjectMakeFunctionWithCallback):

7823

(JSObjectMakeFunction):

7824

(OpaqueJSPropertyNameArray::OpaqueJSPropertyNameArray):

7825

(JSObjectCopyPropertyNames):

7826

* API/JSObjectRef.h:

7827

* API/minidom.c:

7828

(main):

7829

* API/testapi.c:

7830

(main):

7831

* ChangeLog:

7832

* JavaScriptCore.exp:

7833

7834

2006-07-16 Geoffrey Garen <ggaren@apple.com>

7835

7836

Laughed at by Beth.

7837

7838

Replace __JS with OpaqueJS because the former, while used by CF, is

7839

a prefix that's triply-reserved by the compiler. (_* is reserved in global

7840

names, _[A-Z] is reserved in all names, and __ is reserved in all names

7841

in C++.)

7842

7843

Opaque is an alternative used by other Mac OS X framewokrs.

7844

7845

* API/APICast.h:

7846

(toJS):

7847

* API/JSBase.h:

7848

* API/JSCallbackObject.cpp:

7849

(KJS::JSCallbackObject::getOwnPropertySlot):

7850

(KJS::JSCallbackObject::put):

7851

(KJS::JSCallbackObject::deleteProperty):

7852

(KJS::JSCallbackObject::getPropertyNames):

7853

(KJS::JSCallbackObject::staticValueGetter):

7854

(KJS::JSCallbackObject::staticFunctionGetter):

7855

* API/JSClassRef.cpp:

7856

(OpaqueJSClass::OpaqueJSClass):

7857

(OpaqueJSClass::~OpaqueJSClass):

7858

* API/JSClassRef.h:

7859

* API/JSObjectRef.cpp:

7860

(JSClassCreate):

7861

(OpaqueJSPropertyNameArray::OpaqueJSPropertyNameArray):

7862

(JSObjectCopyPropertyNames):

7863

7864

2006-07-16 Darin Adler <darin@apple.com>

7865

7866

- try to fix Windows build

7867

7868

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

7869

Added some recently added files, removed some recently removed.

7870

7871

2006-07-16 Geoffrey Garen <ggaren@apple.com>

7872

7873

Reviewed by Maciej.

7874

7875

- Change getProperty* to return undefined, rather than NULL, for missing

7876

properties, since that's what the spec says. Also added exception out

7877

parameters to the *Index functions, because they can call through to the

7878

regular functions, which can throw for custom objects.

7879

7880

* API/JSObjectRef.cpp:

7881

(JSObjectGetProperty):

7882

(JSObjectGetPropertyAtIndex):

7883

(JSObjectSetPropertyAtIndex):

7884

* API/JSObjectRef.h:

7885

* API/testapi.c:

7886

(main):

7887

7888

2006-07-16 Geoffrey Garen <ggaren@apple.com>

7889

7890

Reviewed by Maciej.

7891

7892

- Properly document and handle NULL callbacks for static properties. We

7893

throw an exception in any case other than a ReadOnly property with a NULL

7894

setProperty callback, because a NULL callback almost certainly indicates

7895

a programming error. Also throw an exception if hasProperty returns true

7896

for a property that getProperty can't get.

7897

7898

- If a static setProperty callback returns 'false', to indicate that the

7899

property was not set, we no longer forward the set request up the class

7900

chain, because that's almost certainly not what the programmer expected.

7901

7902

* API/JSCallbackObject.cpp:

7903

(KJS::JSCallbackObject::getOwnPropertySlot):

7904

(KJS::JSCallbackObject::put):

7905

(KJS::JSCallbackObject::staticValueGetter):

7906

(KJS::JSCallbackObject::staticFunctionGetter):

7907

(KJS::JSCallbackObject::callbackGetter):

7908

* API/JSObjectRef.h:

7909

* API/minidom.js:

7910

* API/testapi.c:

7911

(MyObject_hasProperty):

7912

* API/testapi.js:

7913

7914

2006-07-16 Geoffrey Garen <ggaren@apple.com>

7915

7916

Reviewed by Maciej.

7917

7918

- Added names to functions.

7919

7920

- Removed GetPrivate/SetPrivate from callbackFunctions and callbackConstructors.

7921

The private data idiom is that a JS object stores its native implementation

7922

as private data. For functions and constructors, the native implementation is nothing

7923

more than the callback they already store, so supporting private data, too,

7924

confuses the idiom. If you *really* want, you can still create a custom

7925

function with private data.

7926

7927

* API/JSCallbackConstructor.cpp:

7928

* API/JSCallbackConstructor.h:

7929

* API/JSCallbackFunction.cpp:

7930

(KJS::JSCallbackFunction::JSCallbackFunction):

7931

* API/JSCallbackFunction.h:

7932

* API/JSCallbackObject.cpp:

7933

(KJS::JSCallbackObject::staticFunctionGetter):

7934

* API/JSObjectRef.cpp:

7935

(JSObjectMakeFunction):

7936

(JSObjectMakeFunctionWithBody):

7937

(JSObjectGetPrivate):

7938

(JSObjectSetPrivate):

7939

* API/JSObjectRef.h:

7940

* API/minidom.c:

7941

(main):

7942

* API/testapi.c:

7943

(main):

7944

7945

2006-07-15 Maciej Stachowiak <mjs@apple.com>

7946

7947

Reviewed by Darin.

7948

7949

- switch property lists to be vector+set of Identifiers instead of list of References

7950

7951

This has the following benefits:

7952

7953

- no duplicates in property lists

7954

- simplifies API calls

7955

- probably more efficient, since linked list is gone

7956

- entirely removed Reference, ReferenceList and ProtectedReference types from the API

7957

7958

* kjs/PropertyNameArray.cpp: Added.

7959

(KJS::PropertyNameArray::add): Check set, if not already there, add to

7960

vector.

7961

* kjs/PropertyNameArray.h: Added.

7962

(KJS::PropertyNameArray::PropertyNameArray): Newly added type, combines

7963

a set and a vector to make a unique but ordered list of identifiers.

7964

(KJS::PropertyNameArray::begin): ditto

7965

(KJS::PropertyNameArray::end): ditto

7966

(KJS::PropertyNameArray::size): ditto

7967

(KJS::PropertyNameArray::operator[]): ditto

7968

* kjs/array_instance.h:

7969

* kjs/array_object.cpp:

7970

(ArrayInstance::getPropertyNames): renamed from getPropertyList, updated

7971

for PropertyNameArray

7972

(ArrayInstance::setLength): updated for PropertyNameArray

7973

(ArrayInstance::pushUndefinedObjectsToEnd): ditto

7974

* kjs/nodes.cpp:

7975

(ForInNode::execute): updated for PropertyNameArray

7976

* kjs/nodes.h:

7977

* kjs/object.cpp:

7978

(KJS::JSObject::getPropertyNames): renamed from getPropertyList, updated

7979

for PropertyNameArray

7980

* kjs/object.h:

7981

* kjs/property_map.cpp:

7982

(KJS::PropertyMap::getEnumerablePropertyNames): updated for PropertyNameArray

7983

(KJS::PropertyMap::getSparseArrayPropertyNames): ditto

7984

* kjs/property_map.h:

7985

* kjs/protected_reference.h: Removed.

7986

* kjs/reference.cpp: Removed.

7987

* kjs/reference.h: Removed.

7988

* kjs/reference_list.cpp: Removed.

7989

* kjs/reference_list.h: Removed.

7990

* kjs/scope_chain.cpp:

7991

(KJS::ScopeChain::print): Use PropertyNamesArray instead of ReferenceList.

7992

* kjs/string_object.cpp:

7993

(StringInstance::getPropertyNames): Updated for new approach.

7994

* kjs/string_object.h:

7995

* kjs/ustring.h:

7996

* API/APICast.h:

7997

(toJS): Added overload for PropertyNameAccumulatorRef / PropertyNameArray*

7998

(toRef): ditto

7999

* API/JSBase.h:

8000

* API/JSCallbackObject.cpp:

8001

(KJS::JSCallbackObject::getPropertyNames): Fixed for new API.

8002

* API/JSCallbackObject.h:

8003

* API/JSObjectRef.cpp:

8004

(__JSPropertyNameArray::__JSPropertyNameArray): Type used for a publicly vended

8005

JSPropertyNameArrayRef.

8006

(JSObjectCopyPropertyNames): New API call - renamed / refactored from

8007

JSObjectCreatePropertyList

8008

(JSPropertyNameArrayRetain): new retain call for JSPropertyNameArray.

8009

(JSPropertyNameArrayRelease): new release call for - " -.

8010

(JSPropertyNameArrayGetCount): Instead of having to use a stateful enumerator you

8011

can now get the count and items in any order.

8012

(JSPropertyNameArrayGetNameAtIndex): See above.

8013

(JSPropertyNameAccumulatorAddName): What you add properties to is now an opaque

8014

accumulator object.

8015

* API/JSObjectRef.h: Prototyped new functions, removed old ones

8016

* JavaScriptCore.exp: Updated exported symbols.

8017

* JavaScriptCore.xcodeproj/project.pbxproj: Added new files, removed old.

8018

* API/testapi.c:

8019

(MyObject_getPropertyNames): Renamed / fixed callback to fit new paradigm.

8020

(main): Updated for new API.

8021

8022

2006-07-15 Darin Adler <darin@apple.com>

8023

8024

- oops, missed a few more arrays that had to be const

8025

8026

* API/JSNode.c:

8027

(JSNodePrototype_appendChild): Added const.

8028

(JSNodePrototype_removeChild): Ditto.

8029

(JSNodePrototype_replaceChild): Ditto.

8030

(JSNode_construct): Ditto.

8031

* API/JSNodeList.c:

8032

(JSNodeListPrototype_item): Ditto.

8033

* API/JSObjectRef.cpp:

8034

(JSObjectMakeFunctionWithBody): Ditto.

8035

(JSObjectCallAsFunction): Ditto.

8036

(JSObjectCallAsConstructor): Ditto.

8037

* API/minidom.c:

8038

(print): Ditto.

8039

* API/testapi.c:

8040

(MyObject_callAsFunction): Ditto.

8041

(MyObject_callAsConstructor): Ditto.

8042

(print_callAsFunction): Ditto.

8043

(myConstructor_callAsConstructor): Ditto.

8044

8045

2006-07-15 Darin Adler <darin@apple.com>

8046

8047

Reviewed by Maciej.

8048

8049

* API/JSNode.h: Made an array parameter const.

8050

* API/JSObjectRef.h: Made array parameters const. Fixed a comment.

8051

8052

2006-07-15 Geoffrey Garen <ggaren@apple.com>

8053

8054

Reviewed by Maciej.

8055

8056

- JSObjectMakeFunctionWithBody includes a function name and named parameters now.

8057

8058

* API/JSObjectRef.cpp:

8059

(JSObjectMakeFunctionWithBody):

8060

* API/JSObjectRef.h:

8061

* API/testapi.c:

8062

(assertEqualsAsUTF8String): More informative failure reporting.

8063

(main): Test more function cases.

8064

8065

2006-07-15 Geoffrey Garen <ggaren@apple.com>

8066

8067

Reviewed by Maciej.

8068

8069

- Moved the arguments passed to JSClassCreate into a single structure,

8070

called JSClassDefinition. This will enable easier structure

8071

migration/versioning in the future, if necessary.

8072

8073

- Added support for class names.

8074

8075

- kJSClassDefinitionNull replaces kJSObjectCallbacksNone.

8076

8077

- JSClass is becoming a fairly complex struct, so I migrated all of its

8078

implementation other than reference counting to the sruct.

8079

8080

- Also moved JSClass* functions in the API to JSObjectRef.cpp, since they're

8081

declared in JSObjectRef.h

8082

8083

- Also added some more informative explanation to the class structure doc.

8084

8085

2006-07-15 Darin Adler <darin@apple.com>

8086

8087

Reviewed by Geoff.

8088

8089

- fix http://bugs.webkit.org/show_bug.cgi?id=8395

8090

<rdar://problem/4613467>

8091

REGRESSION: RegEx seems broken for hex escaped non breaking space

8092

8093

Test: fast/js/regexp-extended-characters-more.html

8094

8095

* pcre/pcre_exec.c:

8096

(match): Got rid of utf16Length local variable to guarantee there's no

8097

extra stack usage in recursive calls. Fixed two places in the PCRE_UTF16

8098

code that were using the length variable, which is the UTF-8 length of

8099

a character in the pattern, to move in the UTF-16 subject string. Instead

8100

they hardcode lengths of 1 and 2 since the code already handles BMP

8101

characters and surrogate pairs separately. Also fixed some DPRINTF so

8102

I could compile with DEBUG on.

8103

(pcre_exec): Changed a place that was checking for multibyte characters

8104

in the subject string to use ISMIDCHAR. Instead it was using hardcoded

8105

logic that was right for UTF-8 but wrong for UTF-16.

8106

8107

* pcre/pcre_compile.c: (pcre_compile2): Fixed a DPRINTF so I could compile

8108

with DEBUG on.

8109

8110

2006-07-14 Geoffrey Garen <ggaren@apple.com>

8111

8112

RS by Maciej.

8113

8114

Global replace in the API of argc/argv with argumentCount/arguments.

8115

8116

2006-07-14 Geoffrey Garen <ggaren@apple.com>

8117

8118

Reviewed by Maciej.

8119

8120

- Finalized exception handling in the API.

8121

8122

setProperty can throw because it throws for built-in arrays. getProperty

8123

and deleteProperty can throw because setProperty can throw and we want

8124

to be consistent, and also because they seem like "actions." callAsFunction,

8125

callAsConstructor, and hasInstance can throw, because they caan throw for

8126

all built-ins.

8127

8128

toBoolean can't throw because it's defined that way in the spec.

8129

8130

- Documented that toBoolean and toObject can't be overridden by custom

8131

objects because they're defined that way in the spec.

8132

8133

=== Safari-521.17 ===

8134

8135

2006-07-14 Geoffrey Garen <ggaren@apple.com>

8136

8137

Reviewed by Maciej.

8138

8139

- Implemented ref-counting of JSContexts by splitting into two datatypes:

8140

JSGlobalContext, which you can create/retain/release, and JSContext, which

8141

you can't.

8142

8143

Internally, you retain a JSGlobalContext/ExecState by retaining its

8144

interpreter, which, in the case of a global ExecState, owns it.

8145

8146

- Also made ~Interpreter() protected to catch places where Interpreter

8147

is manually deleted. (Can't make it private because some crazy fool

8148

decided it would be a good idea to subclass Interpreter in other frameworks.

8149

I pity da fool.)

8150

8151

* API/APICast.h:

8152

(toJS): Added cast for new JSGlobalContext

8153

* API/JSStringRef.h: Changed vague "you must" language to more specific

8154

(but, ultimately, equally vague) "behavior is undefined if you don't"

8155

language.

8156

(KJS::Interpreter::Interpreter): Factored more common initialization into

8157

init()

8158

* kjs/interpreter.h:

8159

(KJS::Interpreter::ref): new

8160

(KJS::Interpreter::deref): new

8161

(KJS::Interpreter::refCount): new

8162

* kjs/testkjs.cpp:

8163

(doIt): Ref-count the interpreter.

8164

8165

2006-07-14 Maciej Stachowiak <mjs@apple.com>

8166

8167

Reviewed by Geoff.

8168

8169

- removed bool return value from JSObjectSetProperty, since it is inefficient and

8170

also doesn't work quite right

8171

- added JSObjectGetPropertyAtIndex and JSObjectSetPropertyAtIndex

8172

8173

* API/JSObjectRef.cpp:

8174

(JSObjectSetProperty): Removed return value and canPut stuff.

8175

(JSObjectGetPropertyAtIndex): Added.

8176

(JSObjectSetPropertyAtIndex): Added.

8177

* API/JSObjectRef.h: Prototyped and documented new functions.

8178

8179

2006-07-14 Geoffrey Garen <ggaren@apple.com>

8180

8181

Reviewed by Beth.

8182

8183

Moved JSCheckScriptSyntax, JSEvaluateScript, and JSGarbageCollect into

8184

JSBase.h/.cpp. They don't belong in the value-specific or context-specific

8185

files because they're not part of the value or context implementations.

8186

8187

* API/JSBase.h:

8188

* API/JSContextRef.cpp:

8189

(JSContextGetGlobalObject):

8190

* API/JSContextRef.h:

8191

* API/JSValueRef.cpp:

8192

(JSValueUnprotect):

8193

* API/JSValueRef.h:

8194

* JavaScriptCore.xcodeproj/project.pbxproj:

8195

8196

2006-07-13 Timothy Hatcher <timothy@apple.com>

8197

8198

Reviewed by Maciej.

8199

8200

Moved JavaScriptCore to be a public framework.

8201

8202

* JavaScriptCore.xcodeproj/project.pbxproj:

8203

8204

2006-07-13 Mark Rowe <opendarwin.org@bdash.net.nz>

8205

8206

Reviewed by Geoffrey.

8207

8208

http://bugs.webkit.org/show_bug.cgi?id=9742

8209

Bug 9742: REGRESSION: WebKit hangs when loading <http://www.vtbook.com>

8210

8211

* kjs/value.h:

8212

(KJS::JSValue::getUInt32): Only types tagged as numeric can be converted to UInt32.

8213

8214

2006-07-13 Geoffrey Garen <ggaren@apple.com>

8215

8216

Pleasing to Maciej.

8217

8218

- Renamed JSEvaluate -> JSEvaluateScript, JSCheckSyntax -> JSCheckScriptSyntax

8219

- Added exception out parameters to JSValueTo* and JSValueIsEqual because

8220

they can throw

8221

- Removed JSObjectGetDescription because it's useless and vague, and

8222

JSValueToString/JSValueIsObjectOfClass do a better job, anyway

8223

- Clarified comments about "IsFunction/Constructor" to indicate that they

8224

are true of all functions/constructors, not just those created by JSObjectMake*

8225

8226

2006-07-12 Geoffrey Garen <ggaren@apple.com>

8227

8228

RS by Beth.

8229

8230

Finished previously approved JSInternalString -> JSString conversion

8231

by renaming the files.

8232

8233

* API/JSCallbackObject.cpp:

8234

* API/JSInternalStringRef.cpp: Removed.

8235

* API/JSInternalStringRef.h: Removed.

8236

* API/JSStringRef.cpp: Added.

8237

* API/JSStringRef.h: Added.

8238

* API/JavaScriptCore.h:

8239

* JavaScriptCore.xcodeproj/project.pbxproj:

8240

8241

2006-07-12 Geoffrey Garen <ggaren@apple.com>

8242

8243

Reviewed by Maciej.

8244

8245

- Removed context and exception parameters from JSObjectGetPropertyEnumerator,

8246

removing the spurious use of ExecState inside JavaScriptCore that made

8247

us think this was necessary in the first place.

8248

8249

(StringInstance::getPropertyList): Use getString instead of toString because

8250

we know we're dealing with a string -- we put it there in the first place.

8251

While we're at it, store the string's size instead of retrieving it each time

8252

through the loop, to avoid the unnecessary killing of puppies.

8253

* kjs/string_object.h:

8254

8255

2006-07-12 Maciej Stachowiak <mjs@apple.com>

8256

8257

Reviewed by Geoff.

8258

8259

- add handling of hasInstance callback for API objects

8260

8261

* API/JSCallbackObject.cpp:

8262

(KJS::JSCallbackObject::implementsHasInstance): Check if callback is present.

8263

(KJS::JSCallbackObject::hasInstance): Invoke appropriate callback.

8264

* API/JSCallbackObject.h:

8265

* API/JSClassRef.cpp:

8266

* API/JSObjectRef.h:

8267

* API/testapi.c:

8268

(MyObject_hasInstance): Test case; should match what construct would do.

8269

* API/testapi.js:

8270

8271

2006-07-11 Geoffrey Garen <ggaren@apple.com>

8272

8273

Reviewed by Maciej.

8274

8275

- Implemented a vast number of renames and comment clarifications

8276

suggested during API review.

8277

8278

JSInternalString -> JSString

8279

JS*Make -> JSValueMake*, JSObjectMake*

8280

JSTypeCode -> JSType

8281

JSValueIsInstanceOf -> JSValueIsInstanceOfConstructor (reads strangely well in client code)

8282

JSGC*Protect -> JSValue*Protect

8283

JS*Callback -> JSObject*Callback

8284

JSGetPropertyListCallback -> JSObjectAddPropertiesToListCallback

8285

JSPropertyEnumeratorGetNext -> JSPropertyEnumeratorGetNextName

8286

JSString* ->

8287

JSStringCreateWithUTF8CString, JSStringGetUTF8CString,

8288

JSStringGetMaximumUTF8CStringSize JSStringIsEqualToUTF8CString,

8289

JSStringCreateWithCFString, JSStringCopyCFString, JSStringCreateWithCharacters.

8290

8291

- Changed functions taking a JSValue out arg and returning a bool indicating

8292

whether it was set to simply return a JSValue or NULL.

8293

8294

- Removed JSStringGetCharacters because it's more documentation than code,

8295

and it's just a glorified memcpy built on existing API functionality.

8296

8297

- Moved standard library includes into the headers that actually require them.

8298

8299

- Standardized use of the phrase "Create Rule."

8300

8301

- Removed JSLock from make functions that don't allocate.

8302

8303

- Added exception handling to JSValueToBoolean, since we now allow

8304

callback objects to throw exceptions upon converting to boolean.

8305

8306

- Renamed JSGCCollect to JSGarbageCollect.

8307

8308

2006-07-10 Geoffrey Garen <ggaren@apple.com>

8309

8310

Reviewed by Darin.

8311

8312

- Changed public header includes to the <JavaScriptCore/ style.

8313

- Changed instances of 'buffer' to 'string' since we decided on

8314

JSInternalString instead of JSStringBuffer.

8315

8316

* API/JSContextRef.h:

8317

* API/JSInternalStringRef.cpp:

8318

(JSStringMake):

8319

(JSInternalStringRetain):

8320

(JSInternalStringRelease):

8321

(JSValueCopyStringValue):

8322

(JSInternalStringGetLength):

8323

(JSInternalStringGetCharactersPtr):

8324

(JSInternalStringGetCharacters):

8325

(JSInternalStringGetMaxLengthUTF8):

8326

(JSInternalStringGetCharactersUTF8):

8327

(CFStringCreateWithJSInternalString):

8328

* API/JSInternalStringRef.h:

8329

* API/JSNode.c:

8330

(JSNodePrototype_appendChild):

8331

(JSNode_getNodeType):

8332

* API/JSObjectRef.cpp:

8333

(JSObjectCallAsConstructor):

8334

* API/JSValueRef.h:

8335

* API/JavaScriptCore.h:

8336

* API/minidom.c:

8337

(main):

8338

(print):

8339

* API/testapi.c:

8340

(MyObject_getPropertyList):

8341

(myConstructor_callAsConstructor):

8342

(main): I noticed that we were prematurely releasing some string buffers,

8343

so I moved their release calls to the end of main(). I got rid of 'Buf' in *Buf

8344

(sometimes changing to 'IString', when necessary to differentiate a variable)

8345

to match the buffer->string change.

8346

8347

=== Safari-521.16 ===

8348

8349

2006-07-10 Darin Adler <darin@apple.com>

8350

8351

* kjs/value.cpp: (KJS::JSValue::toInt32Inline): Added inline keyword one more place.

8352

Just in case.

8353

8354

2006-07-10 Darin Adler <darin@apple.com>

8355

8356

- fix the release build

8357

8358

* kjs/value.h:

8359

* kjs/value.cpp:

8360

(KJS::JSValue::toInt32Inline): Move the code here to an inline.

8361

(KJS::JSValue::toInt32): Call the inline from both overloaded toInt32 functions.

8362

8363

2006-07-10 David Kilzer <ddkilzer@kilzer.net>

8364

8365

Reviewed by Darin.

8366

8367

- fix http://bugs.webkit.org/show_bug.cgi?id=9179

8368

Implement select.options.add() method

8369

8370

* JavaScriptCore.exp: Added overloaded KJS::JSValue::toInt32() method.

8371

* JavaScriptCore.xcodeproj/project.pbxproj: Altered attributes metadata for

8372

kjs/value.h to make it available as a forwarded header.

8373

* kjs/lookup.h:

8374

(KJS::lookupPut): Extracted a lookupPut() method from the existing lookupPut() method.

8375

The new method returns a boolean value if no entry is found in the lookup table.

8376

* kjs/value.cpp:

8377

(KJS::JSValue::toInt32): Overloaded toInt32() method with boolean "Ok" argument.

8378

* kjs/value.h: Ditto.

8379

8380

2006-07-10 Geoffrey Garen <ggaren@apple.com>

8381

8382

No review necessary. Removed bogus file I accidentally checked in before.

8383

8384

* API/JSInternalSringRef.h: Removed.

8385

8386

2006-07-10 Geoffrey Garen <ggaren@apple.com>

8387

8388

Reviewed by Darin.

8389

8390

Added exception out parameter to API object callbacks, removed semi-bogus

8391

JSContext(.*)Exception functions.

8392

8393

To make these calls syntactically simple, I added an exceptionSlot()

8394

method to the ExecState class, which provides a JSValue** slot in which to

8395

store a JSValue* exception.

8396

8397

* API/APICast.h:

8398

(toRef):

8399

* API/JSCallbackConstructor.cpp:

8400

(KJS::JSCallbackConstructor::construct):

8401

* API/JSCallbackFunction.cpp:

8402

(KJS::JSCallbackFunction::callAsFunction):

8403

* API/JSCallbackObject.cpp:

8404

(KJS::JSCallbackObject::init):

8405

(KJS::JSCallbackObject::getOwnPropertySlot):

8406

(KJS::JSCallbackObject::put):

8407

(KJS::JSCallbackObject::deleteProperty):

8408

(KJS::JSCallbackObject::construct):

8409

(KJS::JSCallbackObject::callAsFunction):

8410

(KJS::JSCallbackObject::getPropertyList):

8411

(KJS::JSCallbackObject::toBoolean):

8412

(KJS::JSCallbackObject::toNumber):

8413

(KJS::JSCallbackObject::toString):

8414

(KJS::JSCallbackObject::staticValueGetter):

8415

(KJS::JSCallbackObject::callbackGetter):

8416

* API/JSContextRef.cpp:

8417

(JSCheckSyntax):

8418

* API/JSContextRef.h:

8419

* API/JSNode.c:

8420

(JSNodePrototype_appendChild):

8421

(JSNodePrototype_removeChild):

8422

(JSNodePrototype_replaceChild):

8423

(JSNode_getNodeType):

8424

(JSNode_getChildNodes):

8425

(JSNode_getFirstChild):

8426

(JSNode_construct):

8427

* API/JSNode.h:

8428

* API/JSNodeList.c:

8429

(JSNodeListPrototype_item):

8430

(JSNodeList_length):

8431

(JSNodeList_getProperty):

8432

* API/JSObjectRef.h:

8433

* API/minidom.c:

8434

(print):

8435

* API/testapi.c:

8436

(MyObject_initialize):

8437

(MyObject_hasProperty):

8438

(MyObject_getProperty):

8439

(MyObject_setProperty):

8440

(MyObject_deleteProperty):

8441

(MyObject_getPropertyList):

8442

(MyObject_callAsFunction):

8443

(MyObject_callAsConstructor):

8444

(MyObject_convertToType):

8445

(print_callAsFunction):

8446

(myConstructor_callAsConstructor):

8447

(main):

8448

* JavaScriptCore.exp:

8449

* kjs/ExecState.h:

8450

(KJS::ExecState::exceptionHandle):

8451

8452

2006-07-10 Geoffrey Garen <ggaren@apple.com>

8453

8454

Reviewed by Darin.

8455

8456

Improved type safety by implementing opaque JSValue/JSObject typing through

8457

abuse of 'const', not void*. Also fixed an alarming number of bugs

8458

exposed by this new type safety.

8459

8460

I made one design change in JavaScriptCore, which is that the JSObject

8461

constructor should take a JSValue* as its prototype argument, not a JSObject*,

8462

since we allow the prototype to be any JSValue*, including jsNull(), for

8463

example.

8464

8465

* API/APICast.h:

8466

(toJS):

8467

* API/JSBase.h:

8468

* API/JSCallbackConstructor.cpp:

8469

(KJS::JSCallbackConstructor::construct):

8470

* API/JSCallbackFunction.cpp:

8471

(KJS::JSCallbackFunction::callAsFunction):

8472

* API/JSCallbackObject.cpp:

8473

(KJS::JSCallbackObject::JSCallbackObject):

8474

(KJS::JSCallbackObject::getOwnPropertySlot):

8475

(KJS::JSCallbackObject::put):

8476

(KJS::JSCallbackObject::construct):

8477

(KJS::JSCallbackObject::callAsFunction):

8478

(KJS::JSCallbackObject::staticFunctionGetter):

8479

* API/JSCallbackObject.h:

8480

* API/JSContextRef.cpp:

8481

(JSEvaluate):

8482

* API/JSNode.c:

8483

(JSNodePrototype_appendChild):

8484

(JSNodePrototype_removeChild):

8485

(JSNodePrototype_replaceChild):

8486

* API/JSObjectRef.cpp:

8487

(JSObjectMake):

8488

(JSFunctionMakeWithBody):

8489

(JSObjectGetProperty):

8490

(JSObjectCallAsFunction):

8491

(JSObjectCallAsConstructor):

8492

* API/JSObjectRef.h:

8493

* API/testapi.c:

8494

(main):

8495

* ChangeLog:

8496

* kjs/object.h:

8497

(KJS::JSObject::JSObject):

8498

8499

2006-07-10 Geoffrey Garen <ggaren@apple.com>

8500

8501

Approved by Maciej, Darin.

8502

8503

Renamed JSStringBufferRef to JSInternalStringRef. "Internal string" means the

8504

JavaScript engine's internal string representation, which is the most

8505

low-level and efficient representation to use when interfacing with JavaScript.

8506

8507

* API/APICast.h:

8508

(toJS):

8509

(toRef):

8510

* API/JSBase.h:

8511

* API/JSCallbackObject.cpp:

8512

(KJS::JSCallbackObject::getOwnPropertySlot):

8513

(KJS::JSCallbackObject::put):

8514

(KJS::JSCallbackObject::deleteProperty):

8515

(KJS::JSCallbackObject::staticValueGetter):

8516

(KJS::JSCallbackObject::callbackGetter):

8517

* API/JSContextRef.cpp:

8518

(JSEvaluate):

8519

(JSCheckSyntax):

8520

* API/JSContextRef.h:

8521

* API/JSInternalStringRef.cpp: Added.

8522

(JSStringMake):

8523

(JSInternalStringCreate):

8524

(JSInternalStringCreateUTF8):

8525

(JSInternalStringRetain):

8526

(JSInternalStringRelease):

8527

(JSValueCopyStringValue):

8528

(JSInternalStringGetLength):

8529

(JSInternalStringGetCharactersPtr):

8530

(JSInternalStringGetCharacters):

8531

(JSInternalStringGetMaxLengthUTF8):

8532

(JSInternalStringGetCharactersUTF8):

8533

(JSInternalStringIsEqual):

8534

(JSInternalStringIsEqualUTF8):

8535

(JSInternalStringCreateCF):

8536

(CFStringCreateWithJSInternalString):

8537

* API/JSInternalStringRef.h: Added.

8538

* API/JSNode.c:

8539

(JSNodePrototype_appendChild):

8540

(JSNode_getNodeType):

8541

(JSNode_getChildNodes):

8542

(JSNode_getFirstChild):

8543

* API/JSNodeList.c:

8544

(JSNodeList_length):

8545

(JSNodeList_getProperty):

8546

* API/JSObjectRef.cpp:

8547

(JSFunctionMakeWithBody):

8548

(JSObjectGetDescription):

8549

(JSObjectHasProperty):

8550

(JSObjectGetProperty):

8551

(JSObjectSetProperty):

8552

(JSObjectDeleteProperty):

8553

(JSPropertyEnumeratorGetNext):

8554

(JSPropertyListAdd):

8555

* API/JSObjectRef.h:

8556

* API/JSStringBufferRef.cpp: Removed.

8557

* API/JSStringBufferRef.h: Removed.

8558

* API/JSValueRef.h:

8559

* API/JavaScriptCore.h:

8560

* API/minidom.c:

8561

(main):

8562

(print):

8563

* API/testapi.c:

8564

(assertEqualsAsUTF8String):

8565

(assertEqualsAsCharactersPtr):

8566

(assertEqualsAsCharacters):

8567

(MyObject_hasProperty):

8568

(MyObject_getProperty):

8569

(MyObject_setProperty):

8570

(MyObject_deleteProperty):

8571

(MyObject_getPropertyList):

8572

(print_callAsFunction):

8573

(myConstructor_callAsConstructor):

8574

(main):

8575

* JavaScriptCore.exp:

8576

* JavaScriptCore.xcodeproj/project.pbxproj:

8577

8578

2006-07-08 Tim Omernick <timo@apple.com>

8579

8580

Reviewed by Maciej.

8581

8582

Added an OpenGL drawing model to the Netscape Plug-in API.

8583

8584

* bindings/npapi.h:

8585

8586

2006-07-08 Timothy Hatcher <timothy@apple.com>

8587

8588

Reviewed by Maciej.

8589

8590

Moved KJS_GetCreatedJavaVMs to jni_utility.cpp.

8591

Switched KJS_GetCreatedJavaVMs over to use dlopen and dlsym

8592

now that NSAddImage, NSLookupSymbolInImage and NSAddressOfSymbol

8593

are deprecated in Leopard.

8594

8595

* JavaScriptCore.exp:

8596

* JavaScriptCore.xcodeproj/project.pbxproj:

8597

* bindings/jni/jni_utility.cpp:

8598

(KJS::Bindings::KJS_GetCreatedJavaVMs):

8599

* bindings/softlinking.c: Removed.

8600

* bindings/softlinking.h: Removed.

8601

8602

2006-07-08 Geoffrey Garen <ggaren@apple.com>

8603

8604

Reviewed by Anders.

8605

8606

- Make JSObjectGetProperty return a JSValue or NULL, like JSEvaluate does.

8607

8608

* API/JSObjectRef.cpp:

8609

(JSObjectGetProperty):

8610

* API/JSObjectRef.h:

8611

* API/testapi.c:

8612

(main):

8613

8614

2006-07-08 Geoffrey Garen <ggaren@apple.com>

8615

8616

Style change -- no review necessary.

8617

8618

Use 0 instead of NULL in API .cpp files, to match our style guidelines.

8619

8620

* API/JSContextRef.cpp:

8621

(JSEvaluate):

8622

* API/JSObjectRef.cpp:

8623

(JSFunctionMakeWithBody):

8624

(JSObjectCallAsFunction):

8625

(JSObjectCallAsConstructor):

8626

* API/JSValueRef.cpp:

8627

(JSValueToObject):

8628

8629

2006-07-08 Geoffrey Garen <ggaren@apple.com>

8630

8631

Reviewed by TimO.

8632

8633

- Added ability to pass NULL for thisObject when calling JSObjectCallAsFunction,

8634

to match JSEvaluate.

8635

8636

* API/JSObjectRef.cpp:

8637

(JSObjectCallAsFunction):

8638

* API/JSObjectRef.h:

8639

* API/testapi.c:

8640

(main):

8641

8642

=== Safari-521.15 ===

8643

8644

2006-07-07 Geoffrey Garen <ggaren@apple.com>

8645

8646

Reviewed by Maciej.

8647

8648

- Standardized which functions take a JSContext as an argument. The rule is:

8649

if you might execute JavaScript, you take a JSContext, otherwise you don't.

8650

8651

The FIXME in JSObjectRef.h requires refactoring some parts of Interpreter,

8652

but not API changes, so I'm putting it off until later.

8653

8654

* API/JSCallbackObject.cpp:

8655

(KJS::JSCallbackObject::JSCallbackObject):

8656

(KJS::JSCallbackObject::init):

8657

* API/JSCallbackObject.h:

8658

* API/JSContextRef.cpp:

8659

(JSContextCreate):

8660

* API/JSContextRef.h:

8661

* API/JSObjectRef.cpp:

8662

(JSObjectMake):

8663

(JSPropertyEnumeratorGetNext):

8664

* API/JSObjectRef.h:

8665

* API/testapi.c:

8666

(MyObject_initialize):

8667

(main):

8668

* JavaScriptCore.exp:

8669

* kjs/array_object.cpp:

8670

(ArrayInstance::setLength):

8671

(ArrayInstance::pushUndefinedObjectsToEnd):

8672

* kjs/nodes.cpp:

8673

(ForInNode::execute):

8674

* kjs/reference.cpp:

8675

(KJS::Reference::getPropertyName):

8676

(KJS::Reference::getValue):

8677

* kjs/reference.h:

8678

* kjs/scope_chain.cpp:

8679

(KJS::ScopeChain::print):

8680

8681

2006-07-06 Geoffrey Garen <ggaren@apple.com>

8682

8683

Reviewed by Maciej.

8684

8685

More API action.

8686

8687

- Headerdoc finished

8688

8689

Semantic Changes:

8690

- Added a JSContextRef argument to many functions, because you need a

8691

JSContextRef for doing virtually anything. I expect to add this argument

8692

to even more functions in a future patch.

8693

8694

- Removed the globalObjectPrototype argument to JSContextCreate because

8695

you can't create an object until you have a context, so it's impossible

8696

to pass a prototype object to JSContextCreate. That's OK because (1) there's

8697

no reason to give the global object a prototype and (2) if you really want

8698

to, you can just use a separate call to JSObjectSetPrototype.

8699

8700

- Removed the JSClassRef argument to JSClassCreate because it was unnecessary,

8701

and you need to be able to make the global object's class before you've

8702

created a JSContext.

8703

8704

- Added an optional exception parameter to JSFunctionMakeWithBody because anything

8705

less would be uncivilized.

8706

8707

- Made the return value parameter to JSObjectGetProperty optional to match

8708

all other return value parameters in the API.

8709

8710

- Made JSObjectSetPrivate/JSObjectGetPrivate work on JSCallbackFunctions

8711

and JSCallbackConstructors. You could use an abstract base class or strategic

8712

placement of m_privateData in the class structure to implement this, but

8713

the former seemed like overkill, and the latter seemed too dangerous.

8714

8715

- Fixed a bug where JSPropertyEnumeratorGetNext would skip the first property.

8716

8717

Cosmetic Changes:

8718

- Reversed the logic of the JSChar #ifdef to avoid confusing headerdoc

8719

8720

- Removed function names from @function declarations because headeroc

8721

can parse them automatically, and I wanted to rule out manual mismatch.

8722

8723

- Changed Error::create to take a const UString& instead of a UString*

8724

because it was looking at me funny.

8725

8726

- Renamed JSStringBufferCreateWithCFString to JSStringBufferCreateCF

8727

because the latter is more concise and it matches JSStringBufferCreateUTF8.

8728

8729

* API/JSCallbackObject.cpp:

8730

(KJS::JSCallbackObject::getOwnPropertySlot):

8731

(KJS::JSCallbackObject::put):

8732

(KJS::JSCallbackObject::deleteProperty):

8733

(KJS::JSCallbackObject::getPropertyList):

8734

(KJS::JSCallbackObject::toBoolean):

8735

(KJS::JSCallbackObject::toNumber):

8736

(KJS::JSCallbackObject::toString):

8737

* API/JSClassRef.cpp:

8738

(JSClassCreate):

8739

* API/JSContextRef.cpp:

8740

(JSContextCreate):

8741

(JSContextSetException):

8742

* API/JSContextRef.h:

8743

* API/JSNode.c:

8744

(JSNodePrototype_class):

8745

(JSNode_class):

8746

* API/JSNodeList.c:

8747

(JSNodeListPrototype_class):

8748

(JSNodeList_class):

8749

* API/JSObjectRef.cpp:

8750

(JSObjectGetProperty):

8751

(JSObjectGetPrivate):

8752

(JSObjectSetPrivate):

8753

(JSObjectCallAsFunction):

8754

(JSObjectCallAsConstructor):

8755

(JSPropertyEnumeratorGetNext):

8756

* API/JSObjectRef.h:

8757

* API/JSStringBufferRef.cpp:

8758

(JSStringBufferCreateCF):

8759

* API/JSStringBufferRef.h:

8760

* API/JSValueRef.cpp:

8761

(JSValueIsInstanceOf):

8762

* API/JSValueRef.h:

8763

* API/minidom.c:

8764

(main):

8765

* API/minidom.js:

8766

* API/testapi.c:

8767

(MyObject_hasProperty):

8768

(MyObject_setProperty):

8769

(MyObject_deleteProperty):

8770

(MyObject_getPropertyList):

8771

(MyObject_convertToType):

8772

(MyObject_class):

8773

(main):

8774

* JavaScriptCore.exp:

8775

8776

2006-07-07 Geoffrey Garen <ggaren@apple.com>

8777

8778

Reviewed by John.

8779

8780

- Fixed a few crashes resulting from NULL parameters to JSClassCreate.

8781

8782

* API/JSClassRef.cpp:

8783

(JSClassCreate):

8784

(JSClassRelease):

8785

* API/testapi.c: Added test for NULL parameters.

8786

(main):

8787

8788

2006-07-07 Geoffrey Garen <ggaren@apple.com>

8789

8790

Reviewed by John, mocked by Darin.

8791

8792

- Changed JSEvaluate to take a JSObjectRef instead of a JSValueRef as

8793

"this," since "this" must be an object.

8794

8795

* API/JSContextRef.cpp:

8796

(JSEvaluate):

8797

* API/JSContextRef.h:

8798

8799

2006-07-07 Geoffrey Garen <ggaren@apple.com>

8800

8801

Reviewed by John.

8802

8803

- More headerdoc

8804

8805

* API/JSBase.h:

8806

* JavaScriptCore.xcodeproj/project.pbxproj:

8807

8808

2006-07-05 Geoffrey Garen <ggaren@apple.com>

8809

8810

RS by Beth.

8811

8812

Renamed JSCharBufferRef, which was universally unpopular, to JSStringBufferRef,

8813

which, hopefully, will be less unpopular.

8814

8815

* API/APICast.h:

8816

(toJS):

8817

(toRef):

8818

* API/JSBase.h:

8819

* API/JSCallbackObject.cpp:

8820

(KJS::JSCallbackObject::getOwnPropertySlot):

8821

(KJS::JSCallbackObject::put):

8822

(KJS::JSCallbackObject::deleteProperty):

8823

(KJS::JSCallbackObject::staticValueGetter):

8824

(KJS::JSCallbackObject::callbackGetter):

8825

* API/JSCharBufferRef.cpp: Removed.

8826

* API/JSCharBufferRef.h: Removed.

8827

* API/JSContextRef.cpp:

8828

(JSEvaluate):

8829

(JSCheckSyntax):

8830

* API/JSContextRef.h:

8831

* API/JSNode.c:

8832

(JSNodePrototype_appendChild):

8833

(JSNode_getNodeType):

8834

(JSNode_getChildNodes):

8835

(JSNode_getFirstChild):

8836

* API/JSNodeList.c:

8837

(JSNodeList_length):

8838

(JSNodeList_getProperty):

8839

* API/JSObjectRef.cpp:

8840

(JSFunctionMakeWithBody):

8841

(JSObjectGetDescription):

8842

(JSObjectHasProperty):

8843

(JSObjectGetProperty):

8844

(JSObjectSetProperty):

8845

(JSObjectDeleteProperty):

8846

(JSPropertyEnumeratorGetNext):

8847

(JSPropertyListAdd):

8848

* API/JSObjectRef.h:

8849

* API/JSStringBufferRef.cpp: Added.

8850

(JSStringMake):

8851

(JSStringBufferCreate):

8852

(JSStringBufferCreateUTF8):

8853

(JSStringBufferRetain):

8854

(JSStringBufferRelease):

8855

(JSValueCopyStringValue):

8856

(JSStringBufferGetLength):

8857

(JSStringBufferGetCharactersPtr):

8858

(JSStringBufferGetCharacters):

8859

(JSStringBufferGetMaxLengthUTF8):

8860

(JSStringBufferGetCharactersUTF8):

8861

(JSStringBufferIsEqual):

8862

(JSStringBufferIsEqualUTF8):

8863

(JSStringBufferCreateWithCFString):

8864

(CFStringCreateWithJSStringBuffer):

8865

* API/JSStringBufferRef.h: Added.

8866

* API/JSValueRef.h:

8867

* API/JavaScriptCore.h:

8868

* API/minidom.c:

8869

(main):

8870

(print):

8871

* API/testapi.c:

8872

(assertEqualsAsUTF8String):

8873

(assertEqualsAsCharactersPtr):

8874

(assertEqualsAsCharacters):

8875

(MyObject_hasProperty):

8876

(MyObject_getProperty):

8877

(MyObject_setProperty):

8878

(MyObject_deleteProperty):

8879

(MyObject_getPropertyList):

8880

(print_callAsFunction):

8881

(myConstructor_callAsConstructor):

8882

(main):

8883

* JavaScriptCore.exp:

8884

* JavaScriptCore.xcodeproj/project.pbxproj:

8885

8886

2006-07-05 Geoffrey Garen <ggaren@apple.com>

8887

8888

RS by Beth.

8889

8890

Moved some code around for more logical file separation.

8891

8892

* API/JSBase.h:

8893

* API/JSContextRef.h:

8894

* API/JSObjectRef.cpp:

8895

* API/JSValueRef.cpp:

8896

(JSValueToObject):

8897

* API/JSValueRef.h:

8898

8899

2006-07-03 Geoffrey Garen <ggaren@apple.com>

8900

8901

Reviewed by Maciej.

8902

8903

Implemented JSFunctionMakeWithBody, which parses a script as a function body

8904

in the global scope, and returns the resulting anonymous function.

8905

8906

I also removed private data from JSCallbackFunction. It never worked,

8907

since JSCallbackFunction doesn't inherit from JSCallbackObject.

8908

8909

* API/JSCallbackConstructor.cpp: Removed.

8910

* API/JSCallbackConstructor.h: Removed.

8911

* API/JSCallbackFunction.cpp:

8912

(KJS::JSCallbackFunction::JSCallbackFunction):

8913

(KJS::JSCallbackFunction::implementsConstruct):

8914

(KJS::JSCallbackFunction::construct):

8915

(KJS::JSCallbackFunction::implementsCall):

8916

(KJS::JSCallbackFunction::callAsFunction):

8917

* API/JSCallbackFunction.h:

8918

* API/JSCallbackObject.cpp:

8919

(KJS::JSCallbackObject::staticFunctionGetter):

8920

* API/JSObjectRef.cpp:

8921

(JSFunctionMake):

8922

(JSFunctionMakeWithCallbacks):

8923

* API/JSObjectRef.h:

8924

* API/JSValueRef.h:

8925

* API/minidom.c:

8926

(main):

8927

* API/testapi.c:

8928

(main):

8929

* JavaScriptCore.exp: Programmatically added all symbols exported by

8930

API object files, and sorted results

8931

* JavaScriptCore.xcodeproj/project.pbxproj:

8932

8933

2006-07-03 Geoffrey Garen <ggaren@apple.com>

8934

8935

Reviewed by Maciej.

8936

8937

- Return syntax error in JSCheckSyntax through a JSValueRef* exception

8938

argument

8939

8940

* API/JSBase.h:

8941

* API/JSContextRef.cpp:

8942

(JSCheckSyntax):

8943

* API/testapi.c:

8944

(main):

8945

* JavaScriptCore.exp:

8946

* kjs/interpreter.cpp:

8947

(KJS::Interpreter::checkSyntax):

8948

* kjs/interpreter.h:

8949

8950

2006-07-04 Darin Adler <darin@apple.com>

8951

8952

- fixed build

8953

8954

* wtf/MathExtras.h: Oops. Added missing #endif.

8955

8956

2006-07-04 Bjoern Graf <bjoern.graf@gmail.com>

8957

8958

Reviewed by Maciej.

8959

Tweaked a bit by Darin.

8960

8961

- http://bugs.webkit.org/show_bug.cgi?id=9678

8962

work around MSVCRT's fmod function returning NaN for fmod(x, infinity) instead of x

8963

8964

* wtf/MathExtras.h: Added include of <float.h>.

8965

(isinf): Fix to return false for NAN.

8966

(wtf_fmod): Added. An inline that works around the bug.

8967

8968

* kjs/nodes.cpp:

8969

* kjs/number_object.cpp:

8970

* kjs/operations.cpp:

8971

* kjs/value.cpp:

8972

Added includes of MathExtras.h to all files using fmod.

8973

8974

* JavaScriptCore.xcodeproj/project.pbxproj: Let Xcode 2.3 have its way with

8975

the project.

8976

8977

2006-07-01 Geoffrey Garen <ggaren@apple.com>

8978

8979

Reviewed by Darin.

8980

8981

- Refined value conversions in the API:

8982

- failed toNumber returns NaN

8983

- failed toObject returns NULL

8984

- failed toString returns empty string

8985

8986

- Refined excpetion handling in the API:

8987

- failed value conversions do not throw exceptions

8988

- uncaught exceptions in JSEvaluate, JSObjectCallAsFunction, and

8989

JSObjectCallAsConstructor are returned through a JSValueRef* exception

8990

argument

8991

- removed JSContextHasException, because JSContextGetException does

8992

the same job

8993

8994

* API/JSBase.h:

8995

* API/JSCharBufferRef.cpp:

8996

(JSValueCopyStringValue):

8997

* API/JSContextRef.cpp:

8998

(JSEvaluate):

8999

* API/JSContextRef.h:

9000

* API/JSNodeList.c: Added test code demonstrating how you would use

9001

toNumber, and why you probably don't need toUInt32, etc.

9002

(JSNodeListPrototype_item):

9003

(JSNodeList_getProperty):

9004

* API/JSObjectRef.cpp:

9005

(JSValueToObject):

9006

(JSObjectCallAsFunction):

9007

(JSObjectCallAsConstructor):

9008

* API/JSObjectRef.h:

9009

* API/JSValueRef.cpp:

9010

(JSValueToNumber):

9011

* API/JSValueRef.h:

9012

* API/minidom.c:

9013

(main):

9014

* API/testapi.c:

9015

(main): Added tests for new rules, and call to JSGCProtect to fix Intel

9016

crash

9017

* JavaScriptCore.exp:

9018

9019

2006-07-03 Darin Adler <darin@apple.com>

9020

9021

- Rolled out HashMap implementation of NPRuntime, at least temporarily.

9022

9023

Fixes hang in the bindings section of layout tests seen on the

9024

buildbot.

9025

9026

This code was using HashMap<const char*, PrivateIdentifier*>.

9027

But that hashes based on pointer identity, not string value.

9028

The default hash for any pointer type is to hash based on the pointer.

9029

And WTF doesn't currently have a string hash for char*.

9030

We'll need to fix that before re-landing this patch.

9031

9032

(Formatting was also incorrect -- extra spaces in parentheses.)

9033

9034

* bindings/npruntime.cpp: Rolled out last change.

9035

9036

2006-07-02 Justin Haygood <jhaygood@spsu.edu>

9037

9038

Reviewed, tweaked, landed by ggaren.

9039

9040

- Port NPRuntime from CFDictionary to HashMap.

9041

9042

* bindings/npruntime.cpp:

9043

(getStringIdentifierDictionary):

9044

(getIntIdentifierDictionary):

9045

(_NPN_GetStringIdentifier):

9046

(_NPN_GetIntIdentifier):

9047

* bindings/npruntime.h:

9048

9049

2006-07-01 Geoffrey Garen <ggaren@apple.com>

9050

9051

Reviewed by Adele.

9052

9053

- Fixed <rdar://problem/4611197> REGRESSION: Liveconnect with Java test

9054

fails at http://www-sor.inria.fr/~dedieu/notes/liveconnect/simple_example.html

9055

9056

* JavaScriptCore.exp: Export symbols used by liveconnect

9057

9058

2006-06-29 Geoffrey Garen <ggaren@apple.com>

9059

9060

Reviewed by Maciej.

9061

9062

- Phase 2 in the JS API.

9063

9064

- Added support for specifying static tables of values -- this should

9065

obviate the need for using complicated callbacks for most lookups.

9066

9067

- API objects are now created with classes (JSClassRef) -- in order to support

9068

static values, and in order to prevent API objects from storing their

9069

data inline, and thus falling into the oversized (read: slow and prone to

9070

giving Maciej the frowny face) heap.

9071

9072

- Added two specialized JSObject subclasses -- JSCallbackFunction and JSCallbackConstructor --

9073

to allow JSFunctionMake and JSConstructorMake to continue to work with

9074

the new class model. Another solution to this problem would be to create

9075

a custom class object for each function and constructor you make. This

9076

solution is more code but also more efficient.

9077

9078

- Substantially beefed up the minidom example to demonstrate and test a

9079

lot of these techniques. Its output is still pretty haphazard, though.

9080

9081

- Gave the <kjs/ preface to some includes -- I'm told this matters to

9082

building on some versions of Linux.

9083

9084

- Implemented JSValueIsInstanceOf and JSValueIsObjectOfClass

9085

9086

- Removed GetDescription callback. Something in the class datastructure

9087

should take care of this.

9088

9089

* API/JSBase.h:

9090

* API/JSCallbackConstructor.cpp: Added.

9091

(KJS::):

9092

(KJS::JSCallbackConstructor::JSCallbackConstructor):

9093

(KJS::JSCallbackConstructor::implementsConstruct):

9094

(KJS::JSCallbackConstructor::construct):

9095

(KJS::JSCallbackConstructor::setPrivate):

9096

(KJS::JSCallbackConstructor::getPrivate):

9097

* API/JSCallbackConstructor.h: Added.

9098

(KJS::JSCallbackConstructor::classInfo):

9099

* API/JSCallbackFunction.cpp: Added.

9100

(KJS::):

9101

(KJS::JSCallbackFunction::JSCallbackFunction):

9102

(KJS::JSCallbackFunction::implementsCall):

9103

(KJS::JSCallbackFunction::callAsFunction):

9104

(KJS::JSCallbackFunction::setPrivate):

9105

(KJS::JSCallbackFunction::getPrivate):

9106

* API/JSCallbackFunction.h: Added.

9107

(KJS::JSCallbackFunction::classInfo):

9108

* API/JSCallbackObject.cpp:

9109

(KJS::):

9110

(KJS::JSCallbackObject::JSCallbackObject):

9111

(KJS::JSCallbackObject::init):

9112

(KJS::JSCallbackObject::~JSCallbackObject):

9113

(KJS::JSCallbackObject::className):

9114

(KJS::JSCallbackObject::getOwnPropertySlot):

9115

(KJS::JSCallbackObject::put):

9116

(KJS::JSCallbackObject::deleteProperty):

9117

(KJS::JSCallbackObject::implementsConstruct):

9118

(KJS::JSCallbackObject::construct):

9119

(KJS::JSCallbackObject::implementsCall):

9120

(KJS::JSCallbackObject::callAsFunction):

9121

(KJS::JSCallbackObject::getPropertyList):

9122

(KJS::JSCallbackObject::toBoolean):

9123

(KJS::JSCallbackObject::toNumber):

9124

(KJS::JSCallbackObject::toString):

9125

(KJS::JSCallbackObject::inherits):

9126

(KJS::JSCallbackObject::staticValueGetter):

9127

(KJS::JSCallbackObject::staticFunctionGetter):

9128

(KJS::JSCallbackObject::callbackGetter):

9129

* API/JSCallbackObject.h:

9130

* API/JSCharBufferRef.cpp:

9131

* API/JSClassRef.cpp: Added.

9132

(JSClassCreate):

9133

(JSClassRetain):

9134

(JSClassRelease):

9135

* API/JSClassRef.h: Added.

9136

(StaticValueEntry::StaticValueEntry):

9137

(StaticFunctionEntry::StaticFunctionEntry):

9138

(__JSClass::__JSClass):

9139

* API/JSContextRef.cpp:

9140

(JSContextCreate):

9141

(JSEvaluate):

9142

* API/JSContextRef.h:

9143

* API/JSNode.c: Added.

9144

(JSNodePrototype_appendChild):

9145

(JSNodePrototype_removeChild):

9146

(JSNodePrototype_replaceChild):

9147

(JSNodePrototype_class):

9148

(JSNode_getNodeType):

9149

(JSNode_getChildNodes):

9150

(JSNode_getFirstChild):

9151

(JSNode_finalize):

9152

(JSNode_class):

9153

(JSNode_prototype):

9154

(JSNode_new):

9155

(JSNode_construct):

9156

* API/JSNode.h: Added.

9157

* API/JSNodeList.c: Added.

9158

(JSNodeListPrototype_item):

9159

(JSNodeListPrototype_class):

9160

(JSNodeList_length):

9161

(JSNodeList_getProperty):

9162

(JSNodeList_finalize):

9163

(JSNodeList_class):

9164

(JSNodeList_prototype):

9165

(JSNodeList_new):

9166

* API/JSNodeList.h: Added.

9167

* API/JSObjectRef.cpp:

9168

(JSObjectMake):

9169

(JSFunctionMake):

9170

(JSConstructorMake):

9171

(__JSPropertyEnumerator::__JSPropertyEnumerator):

9172

(JSObjectCreatePropertyEnumerator):

9173

(JSPropertyEnumeratorGetNext):

9174

(JSPropertyEnumeratorRetain):

9175

(JSPropertyEnumeratorRelease):

9176

* API/JSObjectRef.h:

9177

(__JSObjectCallbacks::):

9178

* API/JSValueRef.cpp:

9179

(JSValueIsObjectOfClass):

9180

(JSValueIsInstanceOf):

9181

* API/JSValueRef.h:

9182

* API/Node.c: Added.

9183

(Node_new):

9184

(Node_appendChild):

9185

(Node_removeChild):

9186

(Node_replaceChild):

9187

(Node_ref):

9188

(Node_deref):

9189

* API/Node.h: Added.

9190

* API/NodeList.c: Added.

9191

(NodeList_new):

9192

(NodeList_length):

9193

(NodeList_item):

9194

(NodeList_ref):

9195

(NodeList_deref):

9196

* API/NodeList.h: Added.

9197

* API/minidom.c:

9198

(main):

9199

(print):

9200

(createStringWithContentsOfFile):

9201

* API/minidom.js:

9202

* API/testapi.c:

9203

(assertEqualsAsCharacters):

9204

(MyObject_getProperty):

9205

(MyObject_class):

9206

(myConstructor_callAsConstructor):

9207

(main):

9208

* API/testapi.js:

9209

* JavaScriptCore.xcodeproj/project.pbxproj:

9210

9211

2006-06-26 Kevin Ollivier <kevino@theolliviers.com>

9212

9213

Reviewed and tweaked by Darin.

9214

9215

- Compile fixes for wx port / gcc 4.0.2

9216

9217

* kjs/array_object.cpp:

9218

Added missing headers.

9219

9220

* kjs/ExecState.h:

9221

gcc needs class prototypes before defining those classes as friend classes

9222

9223

2006-06-30 Mike Emmel <mike.emmel@gmail.com>

9224

9225

Reviewed by Darin.

9226

9227

Compilation fixes for Linux/Gdk.

9228

9229

* JavaScriptCore/kjs/interpreter.cpp: added include of signal.h

9230

* JavaScriptCore/kjs/ExecState.h: added missing class declaration

9231

* JavaScriptCore/kjs/ExecState.cpp: case wrong on include of context.h

9232

* JavaScriptCore/JavaScriptCoreSources.bkl: added Context.cpp and ExecState.cpp

9233

9234

=== Safari-521.14 ===

9235

9236

2006-06-29 Maciej Stachowiak <mjs@apple.com>

9237

9238

Reviewed by Geoff.

9239

9240

- add headerdoc comments to some of the new JS API headers

9241

9242

* API/JSBase.h:

9243

* API/JSValueRef.h:

9244

9245

2006-06-28 Timothy Hatcher <timothy@apple.com>

9246

9247

Prefer the Stabs debugging symbols format until DWARF bugs are fixed.

9248

9249

* JavaScriptCore.xcodeproj/project.pbxproj:

9250

9251

2006-06-27 Timothy Hatcher <timothy@apple.com>

9252

9253

Reviewed by Tim O.

9254

9255

<rdar://problem/4448350> Deprecated ObjC language API used in JavaScriptCore, WebCore, WebKit and WebBrowser

9256

9257

Switch to the new ObjC 2 API, ifdefed the old code around OBJC_API_VERSION so it still works on Tiger.

9258

Removed the use of the old stringWithCString, switched to the new Tiger version that accepts an encoding.

9259

Lots of code style cleanup.

9260

9261

* JavaScriptCore.xcodeproj/project.pbxproj:

9262

* bindings/objc/objc_class.h:

9263

* bindings/objc/objc_class.mm:

9264

(KJS::Bindings::ObjcClass::~ObjcClass):

9265

(KJS::Bindings::_createClassesByIsAIfNecessary):

9266

(KJS::Bindings::ObjcClass::classForIsA):

9267

(KJS::Bindings::ObjcClass::name):

9268

(KJS::Bindings::ObjcClass::methodsNamed):

9269

(KJS::Bindings::ObjcClass::fieldNamed):

9270

(KJS::Bindings::ObjcClass::fallbackObject):

9271

* bindings/objc/objc_header.h:

9272

* bindings/objc/objc_instance.h:

9273

* bindings/objc/objc_instance.mm:

9274

(ObjcInstance::ObjcInstance):

9275

(ObjcInstance::~ObjcInstance):

9276

(ObjcInstance::operator=):

9277

(ObjcInstance::begin):

9278

(ObjcInstance::end):

9279

(ObjcInstance::getClass):

9280

(ObjcInstance::invokeMethod):

9281

(ObjcInstance::invokeDefaultMethod):

9282

(ObjcInstance::setValueOfField):

9283

(ObjcInstance::supportsSetValueOfUndefinedField):

9284

(ObjcInstance::setValueOfUndefinedField):

9285

(ObjcInstance::getValueOfField):

9286

(ObjcInstance::getValueOfUndefinedField):

9287

(ObjcInstance::defaultValue):

9288

(ObjcInstance::stringValue):

9289

(ObjcInstance::numberValue):

9290

(ObjcInstance::booleanValue):

9291

(ObjcInstance::valueOf):

9292

* bindings/objc/objc_runtime.h:

9293

* bindings/objc/objc_runtime.mm:

9294

(ObjcMethod::ObjcMethod):

9295

(ObjcMethod::name):

9296

(ObjcMethod::getMethodSignature):

9297

(ObjcMethod::setJavaScriptName):

9298

(ObjcField::name):

9299

(ObjcField::type):

9300

(ObjcField::valueFromInstance):

9301

(convertValueToObjcObject):

9302

(ObjcField::setValueToInstance):

9303

(ObjcArray::operator=):

9304

(ObjcArray::setValueAt):

9305

(ObjcArray::valueAt):

9306

(ObjcFallbackObjectImp::ObjcFallbackObjectImp):

9307

(ObjcFallbackObjectImp::callAsFunction):

9308

(ObjcFallbackObjectImp::defaultValue):

9309

9310

2006-06-28 Anders Carlsson <acarlsson@apple.com>

9311

9312

Reviewed by Geoff.

9313

9314

http://bugs.webkit.org/show_bug.cgi?id=8636

9315

REGRESSION: JavaScript access to Java applet causes hang (_webViewURL not implemented)

9316

9317

* bindings/jni/jni_objc.mm:

9318

(KJS::Bindings::dispatchJNICall):

9319

Just pass nil as the calling URL. This will cause the Java plugin to use the URL of the page

9320

containing the applet (which is what we used to do).

9321

9322

2006-06-27 Timothy Hatcher <timothy@apple.com>

9323

9324

Reviewed by Darin.

9325

9326

<rdar://problem/4406785> Add an export file to TOT JavaScriptCore like the Safari-2-0-branch

9327

9328

* JavaScriptCore.exp: Added.

9329

* JavaScriptCore.xcodeproj/project.pbxproj:

9330

9331

2006-06-25 Geoffrey Garen <ggaren@apple.com>

9332

9333

Reviewed by Adele.

9334

9335

- Added JSConstructorMake to match JSFunctionMake, along with test code.

9336

9337

[ I checked in the ChangeLog before without the actual files. ]

9338

9339

* API/JSObjectRef.cpp:

9340

(JSConstructorMake):

9341

* API/JSObjectRef.h:

9342

* API/testapi.c:

9343

(myConstructor_callAsConstructor):

9344

(main):

9345

* API/testapi.js:

9346

* ChangeLog:

9347

* JavaScriptCore.xcodeproj/project.pbxproj: Moved testapi.c to the testapi

9348

target -- this was an oversight in my earlier check-in.

9349

9350

2006-06-25 Timothy Hatcher <timothy@apple.com>

9351

9352

Reviewed by Darin.

9353

9354

Bug 9574: Drosera should show inline scripts within the original HTML

9355

http://bugs.webkit.org/show_bug.cgi?id=9574

9356

9357

Pass the starting line number and error message to the debugger.

9358

9359

* kjs/debugger.cpp:

9360

(Debugger::sourceParsed):

9361

* kjs/debugger.h:

9362

* kjs/function.cpp:

9363

(KJS::GlobalFuncImp::callAsFunction):

9364

* kjs/function_object.cpp:

9365

(FunctionObjectImp::construct):

9366

* kjs/interpreter.cpp:

9367

(KJS::Interpreter::evaluate):

9368

9369

2006-06-24 Alexey Proskuryakov <ap@nypop.com>

9370

9371

Rubber-stamped by Eric.

9372

9373

Add a -h (do not follow symlinks) option to ln in derived sources build script (without it,

9374

a symlink was created inside the source directory on second build).

9375

9376

* JavaScriptCore.xcodeproj/project.pbxproj:

9377

9378

2006-06-24 David Kilzer <ddkilzer@kilzer.net>

9379

9380

Reviewed by Timothy.

9381

9382

* Info.plist: Fixed copyright to include 2003-2006.

9383

9384

2006-06-24 Alexey Proskuryakov <ap@nypop.com>

9385

9386

Reviewed by Darin.

9387

9388

- http://bugs.webkit.org/show_bug.cgi?id=9418

9389

WebKit will not build when Space exists in path

9390

9391

* JavaScriptCore.xcodeproj/project.pbxproj: Enclose search paths in quotes; create symlinks to

9392

avoid passing paths with spaces to make.

9393

9394

2006-06-23 Timothy Hatcher <timothy@apple.com>

9395

9396

Reviewed by Darin.

9397

9398

Adding more operator[] overloads for long and short types.

9399

9400

* wtf/Vector.h:

9401

(WTF::Vector::operator[]):

9402

9403

=== JavaScriptCore-521.13 ===

9404

9405

2006-06-22 Alexey Proskuryakov <ap@nypop.com>

9406

9407

Build fix.

9408

9409

- http://bugs.webkit.org/show_bug.cgi?id=9539

9410

Another case error preventing build

9411

9412

* API/JSObjectRef.cpp: Changed "identifier.h" to "Identifier.h"

9413

9414

2006-06-22 David Kilzer <ddkilzer@kilzer.net>

9415

9416

Build fix.

9417

9418

http://bugs.webkit.org/show_bug.cgi?id=9539

9419

Another case error preventing build

9420

9421

* API/APICast.h: Changed "UString.h" to "ustring.h".

9422

9423

2006-06-21 Geoffrey Garen <ggaren@apple.com>

9424

9425

Fixed release build, fixed accidental infinite recursion due to

9426

last minute global replace gone awry.

9427

9428

* API/APICast.h:

9429

(toRef):

9430

* API/testapi.c:

9431

(assertEqualsAsBoolean):

9432

(assertEqualsAsNumber):

9433

(assertEqualsAsUTF8String):

9434

(assertEqualsAsCharactersPtr):

9435

* JavaScriptCore.xcodeproj/project.pbxproj:

9436

9437

2006-06-21 Geoffrey Garen <ggaren@apple.com>

9438

9439

Reviewed by Anders.

9440

9441

- First cut at C API to JavaScript. Includes a unit test, 'testapi.c',

9442

and the outline of a test app, 'minidom.c'.

9443

9444

Includes one change to JSC internals: Rename propList to getPropertyList and have it

9445

take its target property list by reference so that subclasses can

9446

add properties to the list before calling through to their superclasses.

9447

9448

Also, I just ran prepare-ChangeLog in about 10 seconds, and I would like

9449

to give a shout-out to that.

9450

9451

* API/APICast.h: Added.

9452

(toJS):

9453

(toRef):

9454

* API/JSBase.h: Added.

9455

* API/JSCallbackObject.cpp: Added.

9456

(KJS::):

9457

(KJS::JSCallbackObject::JSCallbackObject):

9458

(KJS::JSCallbackObject::~JSCallbackObject):

9459

(KJS::JSCallbackObject::className):

9460

(KJS::JSCallbackObject::getOwnPropertySlot):

9461

(KJS::JSCallbackObject::put):

9462

(KJS::JSCallbackObject::deleteProperty):

9463

(KJS::JSCallbackObject::implementsConstruct):

9464

(KJS::JSCallbackObject::construct):

9465

(KJS::JSCallbackObject::implementsCall):

9466

(KJS::JSCallbackObject::callAsFunction):

9467

(KJS::JSCallbackObject::getPropertyList):

9468

(KJS::JSCallbackObject::toBoolean):

9469

(KJS::JSCallbackObject::toNumber):

9470

(KJS::JSCallbackObject::toString):

9471

(KJS::JSCallbackObject::setPrivate):

9472

(KJS::JSCallbackObject::getPrivate):

9473

(KJS::JSCallbackObject::cachedValueGetter):

9474

(KJS::JSCallbackObject::callbackGetter):

9475

* API/JSCallbackObject.h: Added.

9476

(KJS::JSCallbackObject::classInfo):

9477

* API/JSCharBufferRef.cpp: Added.

9478

(JSStringMake):

9479

(JSCharBufferCreate):

9480

(JSCharBufferCreateUTF8):

9481

(JSCharBufferRetain):

9482

(JSCharBufferRelease):

9483

(JSValueCopyStringValue):

9484

(JSCharBufferGetLength):

9485

(JSCharBufferGetCharactersPtr):

9486

(JSCharBufferGetCharacters):

9487

(JSCharBufferGetMaxLengthUTF8):

9488

(JSCharBufferGetCharactersUTF8):

9489

(JSCharBufferIsEqual):

9490

(JSCharBufferIsEqualUTF8):

9491

(JSCharBufferCreateWithCFString):

9492

(CFStringCreateWithJSCharBuffer):

9493

* API/JSCharBufferRef.h: Added.

9494

* API/JSContextRef.cpp: Added.

9495

(JSContextCreate):

9496

(JSContextDestroy):

9497

(JSContextGetGlobalObject):

9498

(JSEvaluate):

9499

(JSCheckSyntax):

9500

(JSContextHasException):

9501

(JSContextGetException):

9502

(JSContextClearException):

9503

(JSContextSetException):

9504

* API/JSContextRef.h: Added.

9505

* API/JSObjectRef.cpp: Added.

9506

(JSValueToObject):

9507

(JSObjectMake):

9508

(JSFunctionMake):

9509

(JSObjectGetDescription):

9510

(JSObjectGetPrototype):

9511

(JSObjectSetPrototype):

9512

(JSObjectHasProperty):

9513

(JSObjectGetProperty):

9514

(JSObjectSetProperty):

9515

(JSObjectDeleteProperty):

9516

(JSObjectGetPrivate):

9517

(JSObjectSetPrivate):

9518

(JSObjectIsFunction):

9519

(JSObjectCallAsFunction):

9520

(JSObjectIsConstructor):

9521

(JSObjectCallAsConstructor):

9522

(__JSPropertyListEnumerator::__JSPropertyListEnumerator):

9523

(JSObjectCreatePropertyEnumerator):

9524

(JSPropertyEnumeratorGetNext):

9525

(JSPropertyEnumeratorRetain):

9526

(JSPropertyEnumeratorRelease):

9527

(JSPropertyListAdd):

9528

* API/JSObjectRef.h: Added.

9529

* API/JSValueRef.cpp: Added.

9530

(JSValueGetType):

9531

(JSValueIsUndefined):

9532

(JSValueIsNull):

9533

(JSValueIsBoolean):

9534

(JSValueIsNumber):

9535

(JSValueIsString):

9536

(JSValueIsObject):

9537

(JSValueIsEqual):

9538

(JSValueIsStrictEqual):

9539

(JSUndefinedMake):

9540

(JSNullMake):

9541

(JSBooleanMake):

9542

(JSNumberMake):

9543

(JSValueToBoolean):

9544

(JSValueToNumber):

9545

(JSGCProtect):

9546

(JSGCUnprotect):

9547

(JSGCCollect):

9548

* API/JSValueRef.h: Added.

9549

* API/JavaScriptCore.h: Added.

9550

* API/minidom.c: Added.

9551

(main):

9552

* API/minidom.html: Added.

9553

* API/minidom.js: Added.

9554

* API/testapi.c: Added.

9555

(assertEqualsAsBoolean):

9556

(assertEqualsAsNumber):

9557

(assertEqualsAsUTF8String):

9558

(assertEqualsAsCharactersPtr):

9559

(assertEqualsAsCharacters):

9560

(MyObject_initialize):

9561

(MyObject_copyDescription):

9562

(MyObject_hasProperty):

9563

(MyObject_getProperty):

9564

(MyObject_setProperty):

9565

(MyObject_deleteProperty):

9566

(MyObject_getPropertyList):

9567

(MyObject_callAsFunction):

9568

(MyObject_callAsConstructor):

9569

(MyObject_convertToType):

9570

(MyObject_finalize):

9571

(print_callAsFunction):

9572

(main):

9573

(createStringWithContentsOfFile):

9574

* API/testapi.js: Added.

9575

* ChangeLog:

9576

* JavaScriptCore.xcodeproj/project.pbxproj:

9577

* bindings/npruntime_impl.h:

9578

* kjs/array_instance.h:

9579

* kjs/array_object.cpp:

9580

(ArrayInstance::getPropertyList):

9581

* kjs/interpreter.cpp:

9582

(KJS::Interpreter::evaluate):

9583

* kjs/nodes.cpp:

9584

(ForInNode::execute):

9585

* kjs/object.cpp:

9586

(KJS::JSObject::put):

9587

(KJS::JSObject::canPut):

9588

(KJS::JSObject::deleteProperty):

9589

(KJS::JSObject::propertyIsEnumerable):

9590

(KJS::JSObject::getPropertyAttributes):

9591

(KJS::JSObject::getPropertyList):

9592

* kjs/object.h:

9593

* kjs/property_map.cpp:

9594

(KJS::PropertyMap::get):

9595

* kjs/property_map.h:

9596

* kjs/scope_chain.cpp:

9597

(KJS::ScopeChain::print):

9598

* kjs/string_object.cpp:

9599

(StringInstance::getPropertyList):

9600

* kjs/string_object.h:

9601

* kjs/ustring.h:

9602

(KJS::UString::Rep::ref):

9603

9604

2006-06-20 Timothy Hatcher <timothy@apple.com>

9605

9606

Reviewed by Geoff.

9607

9608

Make sure we clear the exception before returning so

9609

that future calls will not fail because of an earlier

9610

exception state. Assert on entry that the WebScriptObject

9611

is working with an ExecState that dose not have an exception.

9612

Document that evaluateWebScript and callWebScriptMethod return

9613

WebUndefined when an exception is thrown.

9614

9615

* bindings/objc/WebScriptObject.h:

9616

* bindings/objc/WebScriptObject.mm:

9617

(-[WebScriptObject callWebScriptMethod:withArguments:]):

9618

(-[WebScriptObject evaluateWebScript:]):

9619

(-[WebScriptObject setValue:forKey:]):

9620

(-[WebScriptObject valueForKey:]):

9621

(-[WebScriptObject removeWebScriptKey:]):

9622

(-[WebScriptObject webScriptValueAtIndex:]):

9623

(-[WebScriptObject setWebScriptValueAtIndex:value:]):

9624

9625

2006-06-19 Anders Carlsson <acarlsson@apple.com>

9626

9627

Reviewed by John.

9628

9629

* kjs/interpreter.cpp:

9630

(KJS::TimeoutChecker::pauseTimeoutCheck):

9631

(KJS::TimeoutChecker::resumeTimeoutCheck):

9632

Fix argument order in setitimer calls.

9633

9634

2006-06-18 Anders Carlsson <acarlsson@apple.com>

9635

9636

Reviewed by Geoff.

9637

9638

* kjs/interpreter.cpp:

9639

(KJS::TimeoutChecker::pauseTimeoutCheck):

9640

Do nothing if the timeout check hasn't been started.

9641

9642

(KJS::TimeoutChecker::resumeTimeoutCheck):

9643

Do nothing if the timeout check hasn't been started.

9644

Use the right signal handler when unblocking.

9645

9646

(KJS::Interpreter::handleTimeout):

9647

pause/resume the timeout check around the call to

9648

shouldInterruptScript().

9649

9650

2006-06-16 Ben Goodger <beng@google.com>

9651

9652

Reviewed by Maciej

9653

9654

http://bugs.webkit.org/show_bug.cgi?id=9491

9655

Windows build breaks in interpreter.cpp

9656

9657

* kjs/interpreter.cpp

9658

(KJS::TimeoutChecker::pauseTimeoutCheck):

9659

(KJS::TimeoutChecker::resumeTimeoutCheck):

9660

Make sure to only assert equality with s_executingInterpreter when it

9661

is being used (i.e. when HAVE(SYS_TIME_H) == true)

9662

9663

2006-06-17 David Kilzer <ddkilzer@kilzer.net>

9664

9665

Reviewed by darin.

9666

9667

http://bugs.webkit.org/show_bug.cgi?id=9477

9668

REGRESSION: fast/dom/replaceChild.html crashes on WebKit ToT in debug build

9669

9670

* kjs/function.cpp:

9671

(KJS::FunctionImp::callAsFunction): Refetch the debugger after executing the function

9672

in case the WebFrame it was running in has since been destroyed.

9673

9674

2006-06-17 David Kilzer <ddkilzer@kilzer.net>

9675

9676

Reviewed by ggaren.

9677

9678

http://bugs.webkit.org/show_bug.cgi?id=9476

9679

REGRESSION: Reproducible crash after closing window after viewing

9680

css2.1/t0803-c5501-imrgn-t-00-b-ag.html

9681

9682

* kjs/debugger.cpp:

9683

(Debugger::detach): Call setDebugger(0) for all interpreters removed from

9684

the 'attached to a debugger' list.

9685

9686

2006-06-17 Anders Carlsson <acarlsson@apple.com>

9687

9688

Reviewed by Maciej and Geoff.

9689

9690

http://bugs.webkit.org/show_bug.cgi?id=7080

9691

Provide some way to stop a JavaScript infinite loop

9692

9693

* kjs/completion.h:

9694

(KJS::):

9695

Add Interrupted completion type.

9696

9697

* kjs/function.cpp:

9698

(KJS::FunctionImp::callAsFunction):

9699

(KJS::GlobalFuncImp::callAsFunction):

9700

Only set the exception on the new ExecState if the current one has had one.

9701

9702

* kjs/interpreter.cpp:

9703

(KJS::TimeoutChecker::startTimeoutCheck):

9704

(KJS::TimeoutChecker::stopTimeoutCheck):

9705

(KJS::TimeoutChecker::alarmHandler):

9706

(KJS::TimeoutChecker::pauseTimeoutCheck):

9707

(KJS::TimeoutChecker::resumeTimeoutCheck):

9708

New TimeoutChecker class which handles setting Interpreter::m_timedOut flag after a given

9709

period of time. This currently only works on Unix platforms where setitimer and signals are used.

9710

9711

(KJS::Interpreter::Interpreter):

9712

Initialize new member variables.

9713

9714

(KJS::Interpreter::~Interpreter):

9715

Destroy the timeout checker.

9716

9717

(KJS::Interpreter::startTimeoutCheck):

9718

(KJS::Interpreter::stopTimeoutCheck):

9719

(KJS::Interpreter::pauseTimeoutCheck):

9720

(KJS::Interpreter::resumeTimeoutCheck):

9721

Call the timeout checker.

9722

9723

(KJS::Interpreter::handleTimeout):

9724

Called on timeout. Resets the m_timedOut flag and calls shouldInterruptScript.

9725

9726

* kjs/interpreter.h:

9727

(KJS::Interpreter::setTimeoutTime):

9728

New function for setting the timeout time.

9729

9730

(KJS::Interpreter::shouldInterruptScript):

9731

New function. The idea is that this should be overridden by subclasses in order to for example

9732

pop up a dialog asking the user if the script should be interrupted.

9733

9734

(KJS::Interpreter::checkTimeout):

9735

New function which checks the m_timedOut flag and calls handleTimeout if it's set.

9736

9737

* kjs/nodes.cpp:

9738

(DoWhileNode::execute):

9739

(WhileNode::execute):

9740

(ForNode::execute):

9741

Call Interpreter::checkTimeout after each iteration of the loop.

9742

9743

2006-06-15 Timothy Hatcher <timothy@apple.com>

9744

9745

Reviewed by Geoff and Darin.

9746

9747

Prefer the DWARF debugging symbols format for use in Xcode 2.3.

9748

9749

* JavaScriptCore.xcodeproj/project.pbxproj:

9750

9751

2006-06-14 Geoffrey Garen <ggaren@apple.com>

9752

9753

Reviewed by Beth.

9754

9755

- fixed http://bugs.webkit.org/show_bug.cgi?id=9438

9756

Someone broke ToT: cannot build

9757

9758

* JavaScriptCore.xcodeproj/project.pbxproj:

9759

* bindings/runtime_root.h: Changed "Interpreter.h" to "interpreter.h"

9760

9761

2006-06-12 Geoffrey Garen <ggaren@apple.com>

9762

9763

build fix

9764

9765

* bindings/objc/WebScriptObject.mm:

9766

(+[WebScriptObject throwException:]): Restore assignment I accidentally

9767

deleted in previous commit

9768

9769

2006-06-12 Geoffrey Garen <ggaren@apple.com>

9770

9771

Reviewed by TimO, Maciej.

9772

9773

- Merged InterpreterImp code into Interpreter, which implements

9774

all interpreter functionality now. This is part of my continuing quest

9775

to create an external notion of JS "execution context" that is unified and simple --

9776

something to replace the mix of Context, ContextImp, ExecState, Interpreter,

9777

InterpreterImp, and JSRun.

9778

9779

All tests pass. Leaks test has not regressed from its baseline ~207 leaks

9780

with ~3460 leaked nodes.

9781

9782

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

9783

* JavaScriptCore.xcodeproj/project.pbxproj:

9784

* bindings/NP_jsobject.cpp:

9785

* bindings/objc/WebScriptObject.mm:

9786

(+[WebScriptObject throwException:]):

9787

* bindings/runtime_root.cpp:

9788

* bindings/runtime_root.h:

9789

* kjs/Context.cpp:

9790

(KJS::Context::Context):

9791

* kjs/ExecState.cpp: Added.

9792

(KJS::ExecState::lexicalInterpreter):

9793

* kjs/ExecState.h: Added.

9794

(KJS::ExecState::dynamicInterpreter):

9795

* kjs/SavedBuiltins.h: Added.

9796

* kjs/bool_object.cpp:

9797

(BooleanPrototype::BooleanPrototype):

9798

* kjs/collector.cpp:

9799

(KJS::Collector::collect):

9800

(KJS::Collector::numInterpreters):

9801

* kjs/context.h:

9802

* kjs/debugger.cpp:

9803

(Debugger::attach):

9804

(Debugger::detach):

9805

* kjs/function.cpp:

9806

(KJS::FunctionImp::callAsFunction):

9807

(KJS::GlobalFuncImp::callAsFunction):

9808

* kjs/function_object.cpp:

9809

(FunctionObjectImp::construct):

9810

* kjs/internal.cpp:

9811

* kjs/internal.h:

9812

* kjs/interpreter.cpp:

9813

(KJS::interpreterMap):

9814

(KJS::Interpreter::Interpreter):

9815

(KJS::Interpreter::init):

9816

(KJS::Interpreter::~Interpreter):

9817

(KJS::Interpreter::globalObject):

9818

(KJS::Interpreter::initGlobalObject):

9819

(KJS::Interpreter::globalExec):

9820

(KJS::Interpreter::checkSyntax):

9821

(KJS::Interpreter::evaluate):

9822

(KJS::Interpreter::builtinObject):

9823

(KJS::Interpreter::builtinFunction):

9824

(KJS::Interpreter::builtinArray):

9825

(KJS::Interpreter::builtinBoolean):

9826

(KJS::Interpreter::builtinString):

9827

(KJS::Interpreter::builtinNumber):

9828

(KJS::Interpreter::builtinDate):

9829

(KJS::Interpreter::builtinRegExp):

9830

(KJS::Interpreter::builtinError):

9831

(KJS::Interpreter::builtinObjectPrototype):

9832

(KJS::Interpreter::builtinFunctionPrototype):

9833

(KJS::Interpreter::builtinArrayPrototype):

9834

(KJS::Interpreter::builtinBooleanPrototype):

9835

(KJS::Interpreter::builtinStringPrototype):

9836

(KJS::Interpreter::builtinNumberPrototype):

9837

(KJS::Interpreter::builtinDatePrototype):

9838

(KJS::Interpreter::builtinRegExpPrototype):

9839

(KJS::Interpreter::builtinErrorPrototype):

9840

(KJS::Interpreter::builtinEvalError):

9841

(KJS::Interpreter::builtinRangeError):

9842

(KJS::Interpreter::builtinReferenceError):

9843

(KJS::Interpreter::builtinSyntaxError):

9844

(KJS::Interpreter::builtinTypeError):

9845

(KJS::Interpreter::builtinURIError):

9846

(KJS::Interpreter::builtinEvalErrorPrototype):

9847

(KJS::Interpreter::builtinRangeErrorPrototype):

9848

(KJS::Interpreter::builtinReferenceErrorPrototype):

9849

(KJS::Interpreter::builtinSyntaxErrorPrototype):

9850

(KJS::Interpreter::builtinTypeErrorPrototype):

9851

(KJS::Interpreter::builtinURIErrorPrototype):

9852

(KJS::Interpreter::mark):

9853

(KJS::Interpreter::interpreterWithGlobalObject):

9854

(KJS::Interpreter::saveBuiltins):

9855

(KJS::Interpreter::restoreBuiltins):

9856

* kjs/interpreter.h:

9857

(KJS::Interpreter::setCompatMode):

9858

(KJS::Interpreter::compatMode):

9859

(KJS::Interpreter::firstInterpreter):

9860

(KJS::Interpreter::nextInterpreter):

9861

(KJS::Interpreter::prevInterpreter):

9862

(KJS::Interpreter::debugger):

9863

(KJS::Interpreter::setDebugger):

9864

(KJS::Interpreter::setContext):

9865

(KJS::Interpreter::context):

9866

* kjs/nodes.cpp:

9867

(StatementNode::hitStatement):

9868

(RegExpNode::evaluate):

9869

* kjs/protect.h:

9870

9871

2006-06-12 Geoffrey Garen <ggaren@apple.com>

9872

9873

Reviewed by Maciej.

9874

9875

- Have *.lut.h files #include lookup.h to eliminate surprising header

9876

include order dependency.

9877

9878

* DerivedSources.make:

9879

* kjs/array_object.cpp:

9880

* kjs/date_object.cpp:

9881

* kjs/date_object.h:

9882

(KJS::DateProtoFunc::):

9883

* kjs/lexer.cpp:

9884

* kjs/math_object.cpp:

9885

* kjs/number_object.cpp:

9886

* kjs/regexp_object.cpp:

9887

* kjs/string_object.cpp:

9888

9889

2006-06-10 Geoffrey Garen <ggaren@apple.com>

9890

9891

- http://bugs.webkit.org/show_bug.cgi?id=8515

9892

Linux porting compile bug

9893

9894

Fix by Mike Emmel, Reviewed by Darin.

9895

9896

* JavaScriptCoreSources.bkl:

9897

* jscore.bkl:

9898

* wtf/Platform.h:

9899

9900

2006-06-09 Geoffrey Garen <ggaren@apple.com>

9901

9902

Build fix -- I think :).

9903

9904

* JavaScriptCore.xcodeproj/project.pbxproj:

9905

* kjs/context.h:

9906

9907

2006-06-09 Geoffrey Garen <ggaren@apple.com>

9908

9909

Reviewed by Eric (yay!).

9910

9911

- Removed Context wrapper for ContextImp, renamed ContextImp to Context,

9912

split Context into its own file -- Context.cpp -- renamed _var to m_var,

9913

change ' *' to '* '.

9914

9915

* JavaScriptCore.xcodeproj/project.pbxproj:

9916

* kjs/Context.cpp: Added.

9917

(KJS::Context::Context):

9918

(KJS::Context::~Context):

9919

(KJS::Context::mark):

9920

* kjs/context.h:

9921

(KJS::Context::scopeChain):

9922

(KJS::Context::variableObject):

9923

(KJS::Context::setVariableObject):

9924

(KJS::Context::thisValue):

9925

(KJS::Context::callingContext):

9926

(KJS::Context::activationObject):

9927

(KJS::Context::currentBody):

9928

(KJS::Context::function):

9929

(KJS::Context::arguments):

9930

(KJS::Context::pushScope):

9931

(KJS::Context::seenLabels):

9932

* kjs/function.cpp:

9933

(KJS::FunctionImp::callAsFunction):

9934

(KJS::FunctionImp::processParameters):

9935

(KJS::FunctionImp::argumentsGetter):

9936

(KJS::GlobalFuncImp::callAsFunction):

9937

* kjs/internal.cpp:

9938

(KJS::InterpreterImp::evaluate):

9939

* kjs/internal.h:

9940

(KJS::InterpreterImp::setContext):

9941

(KJS::InterpreterImp::context):

9942

* kjs/interpreter.cpp:

9943

* kjs/interpreter.h:

9944

(KJS::ExecState::context):

9945

(KJS::ExecState::ExecState):

9946

* kjs/nodes.cpp:

9947

(currentSourceId):

9948

(currentSourceURL):

9949

(ThisNode::evaluate):

9950

(ResolveNode::evaluate):

9951

(FunctionCallResolveNode::evaluate):

9952

(PostfixResolveNode::evaluate):

9953

(DeleteResolveNode::evaluate):

9954

(TypeOfResolveNode::evaluate):

9955

(PrefixResolveNode::evaluate):

9956

(AssignResolveNode::evaluate):

9957

(VarDeclNode::evaluate):

9958

(VarDeclNode::processVarDecls):

9959

(DoWhileNode::execute):

9960

(WhileNode::execute):

9961

(ForNode::execute):

9962

(ForInNode::execute):

9963

(ContinueNode::execute):

9964

(BreakNode::execute):

9965

(ReturnNode::execute):

9966

(WithNode::execute):

9967

(SwitchNode::execute):

9968

(LabelNode::execute):

9969

(TryNode::execute):

9970

(FuncDeclNode::processFuncDecl):

9971

(FuncExprNode::evaluate):

9972

9973

2006-06-07 Geoffrey Garen <ggaren@apple.com>

9974

9975

Removed API directory I prematurely/accidentally added.

9976

9977

* API: Removed.

9978

9979

2006-06-05 Mitz Pettel <opendarwin.org@mitzpettel.com>

9980

9981

Reviewed and landed by Geoff.

9982

9983

- fix a regression in ecma_3/String/regress-104375.js

9984

9985

* kjs/string_object.cpp:

9986

(substituteBackreferences): If a 2-digit back reference is out of range,

9987

parse it as a 1-digit reference (followed by the other digit). This matches

9988

Firefox's behavior.

9989

9990

2006-06-05 Geoffrey Garen <ggaren@apple.com>

9991

9992

Reviewed By Maciej.

9993

Darin already reviewed this change on the branch. See <rdar://problem/4317701>.

9994

9995

- Fixed <rdar://problem/4291345> PCRE overflow in Safari JavaScriptCore

9996

9997

No test case because there's no behavior change.

9998

9999

* pcre/pcre_compile.c:

10000

(read_repeat_counts): Check for integer overflow / out of bounds

10001

10002

2006-06-05 Geoffrey Garen <ggaren@apple.com>

10003

10004

Reviewed by aliu.

10005

10006

- Changed CString length from int to size_t. We sould probably do this

10007

for UString, too. (Darin, if you're reading this: Maciej said so.)

10008

10009

* kjs/function.cpp:

10010

(KJS::encode):

10011

* kjs/ustring.cpp:

10012

(KJS::CString::CString):

10013

(KJS::operator==):

10014

* kjs/ustring.h:

10015

(KJS::CString::size):

10016

10017

2006-06-04 Geoffrey Garen <ggaren@apple.com>

10018

10019

Reviewed by Maciej.

10020

10021

- http://bugs.webkit.org/show_bug.cgi?id=9304

10022

Minor cleanup in JavaScriptCore

10023

10024

* kjs/value.h: Removed redundant declarations

10025

10026

2006-06-04 Darin Adler <darin@apple.com>

10027

10028

Reviewed by Anders.

10029

10030

- changed deleteAllValues so it can work on "const" collections

10031

Deleting the values affects the values, not the pointers in the

10032

collection, so it's legitimate to do it to a const collection,

10033

and a case of that actually came up in the XPath code.

10034

10035

* wtf/HashMap.h:

10036

(WTF::deleteAllPairSeconds): Use const iterators.

10037

(WTF::deleteAllValues): Take const HashMap reference as a parameter.

10038

* wtf/HashSet.h:

10039

(WTF::deleteAllValues): Take const HashSet reference as a parameter,

10040

and use const iterators.

10041

* wtf/Vector.h:

10042

(WTF::deleteAllValues): Take const Vector reference as a parameter.

10043

10044

- added more functions that are present in <math.h> on some platforms,

10045

but not on others; moved here from various files in WebCore

10046

10047

* wtf/MathExtras.h:

10048

(isinf): Added.

10049

(isnan): Added.

10050

(lround): Added.

10051

(lroundf): Tweaked.

10052

(round): Added.

10053

(roundf): Tweaked.

10054

(signbit): Added.

10055

10056

2006-06-02 Mitz Pettel <opendarwin.org@mitzpettel.com>

10057

10058

Reviewed by ggaren.

10059

10060

- http://bugs.webkit.org/show_bug.cgi?id=9234

10061

Implement $&, $' and $` replacement codes in String.prototype.replace

10062

10063

Test: fast/js/string-replace-3.html

10064

10065

* kjs/string_object.cpp:

10066

(substituteBackreferences): Added support for $& (matched substring),

10067

$` (everything preceding matched substring), $' (everything following

10068

matched substring) and 2-digit back references, and cleaned up a little.

10069

10070

2006-06-02 Adele Peterson <adele@apple.com>

10071

10072

Reviewed by Darin.

10073

10074

Set incremental linking to no. This seems to fix a build problem I was seeing

10075

where dftables couldn't find a dll.

10076

10077

* JavaScriptCore.vcproj/dftables/dftables.vcproj:

10078

10079

2006-05-26 Steve Falkenburg <sfalken@apple.com>

10080

10081

Build fixes/tweaks

10082

10083

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

10084

10085

=== JavaScriptCore-521.11 ===

10086

10087

2006-05-24 Geoffrey Garen <ggaren@apple.com>

10088

10089

Reviewed by mjs.

10090

10091

- JSC half of fix for <rdar://problem/4557926> TOT REGRESSSION: Crash

10092

occurs when attempting to view image in slideshow mode at

10093

http://d.smugmug.com/gallery/581716 ( KJS::IfNode::execute

10094

(KJS::ExecState*) + 312)

10095

10096

On alternate threads, DOMObjects remain in the

10097

ScriptInterpreter's cache because they're not collected. So, they

10098

need an opportunity to mark their children.

10099

10100

I'm not particularly happy with this solution because it fails to

10101

resolve many outstanding issues with the DOM object cache. Since none

10102

of those issues is a crasher or a serious compatibility concern,

10103

and since the behavior of other browsers is not much to go on in this

10104

case, I've filed <rdar://problem/4561439> about that, and I'm moving on

10105

with my life.

10106

10107

* JavaScriptCore.xcodeproj/project.pbxproj:

10108

* kjs/collector.cpp:

10109

(KJS::Collector::collect):

10110

* kjs/internal.cpp:

10111

(KJS::InterpreterImp::mark):

10112

* kjs/internal.h:

10113

* kjs/interpreter.cpp:

10114

(KJS::Interpreter::mark):

10115

* kjs/interpreter.h:

10116

10117

=== JavaScriptCore-521.10 ===

10118

10119

2006-05-22 Timothy Hatcher <timothy@apple.com>

10120

10121

Reviewed by Eric, Kevin and Geoff.

10122

10123

Merge open source build fixes. <rdar://problem/4555500>

10124

10125

* kjs/collector.cpp: look at the rsp register in x86_64

10126

(KJS::Collector::markOtherThreadConservatively):

10127

* wtf/Platform.h: add x86_64 to the platform list

10128

10129

2006-05-19 Anders Carlsson <acarlsson@apple.com>

10130

10131

Reviewed by Geoff.

10132

10133

http://bugs.webkit.org/show_bug.cgi?id=8993

10134

Support function declaration in case statements

10135

10136

* kjs/grammar.y: Get rid of StatementList and use SourceElements instead.

10137

10138

* kjs/nodes.cpp:

10139

(CaseClauseNode::evalStatements):

10140

(CaseClauseNode::processVarDecls):

10141

(CaseClauseNode::processFuncDecl):

10142

(ClauseListNode::processFuncDecl):

10143

(CaseBlockNode::processFuncDecl):

10144

(SwitchNode::processFuncDecl):

10145

* kjs/nodes.h:

10146

(KJS::CaseClauseNode::CaseClauseNode):

10147

(KJS::ClauseListNode::ClauseListNode):

10148

(KJS::ClauseListNode::getClause):

10149

(KJS::ClauseListNode::getNext):

10150

(KJS::ClauseListNode::releaseNext):

10151

(KJS::SwitchNode::SwitchNode):

10152

Add processFuncDecl for the relevant nodes.

10153

10154

* kjs/nodes2string.cpp:

10155

(CaseClauseNode::streamTo):

10156

next got renamed to source.

10157

10158

2006-05-17 George Staikos <staikos@kde.org>

10159

10160

Reviewed by Maciej, Alexey, and Eric.

10161

10162

* pcre/pcre_compile.c:

10163

* pcre/pcre_get.c:

10164

* pcre/pcre_exec.c:

10165

* wtf/UnusedParam.h:

10166

Use /**/ in .c files to compile with non-C99 and non-GCC compilers.

10167

10168

* kjs/testkjs.cpp:

10169

Change include to <wtf/HashTraits.h> from "HashTraits.h" to avoid -I

10170

10171

* wtf/unicode/qt4/UnicodeQt4.h:

10172

Use correct parentheses and correct mask for utf-32 support.

10173

10174

2006-05-17 Alexey Proskuryakov <ap@nypop.com>

10175

10176

Reviewed by Darin.

10177

10178

- fix http://bugs.webkit.org/show_bug.cgi?id=8870

10179

Crash typing in Yahoo auto-complete widget.

10180

10181

Test: fast/js/regexp-stack-overflow.html

10182

10183

* pcre/pcre-config.h: Define NO_RECURSE.

10184

10185

2006-05-16 George Staikos <staikos@kde.org>

10186

10187

Reviewed by Maciej.

10188

10189

Fix some warnings and strict compilation errors.

10190

10191

* kjs/nodes.cpp:

10192

* kjs/value.cpp:

10193

10194

2006-05-15 Alexey Proskuryakov <ap@nypop.com>

10195

10196

* make-generated-sources.sh: Changed to be executable and removed

10197

text in the file generated by "svn diff".

10198

10199

2006-05-15 Geoffrey Garen <ggaren@apple.com>

10200

10201

Reviewed by Maciej.

10202

10203

- Fixed <rdar://problem/4534904> please do not treat "debugger" as

10204

a reserved word while parsing JavaScript (and other ECMA reserved

10205

words)

10206

10207

AKA

10208

10209

http://bugs.webkit.org/show_bug.cgi?id=6179

10210

We treat "char" as a reserved word in JavaScript and firefox/IE do

10211

not

10212

10213

(1) I unreserved most of the spec's "future reserved words" because

10214

they're not reserved in IE or FF. (Most, but not all, because IE

10215

somewhat randomly *does* reserve a few of them.)

10216

(2) I made 'debugger' a legitimate statement that acts like an empty

10217

statement because FF and IE support it.

10218

10219

* kjs/grammar.y:

10220

* kjs/keywords.table:

10221

10222

2006-05-15 Tim Omernick <timo@apple.com>

10223

10224

Reviewed by John Sullivan.

10225

10226

Part of <rdar://problem/4466508> Add 64-bit support to the Netscape Plugin API

10227

10228

Added to the Netscape Plugin API the concept of "plugin drawing models". The drawing model

10229

determines the kind of graphics context created by the browser for the plugin, as well as

10230

the Mac types of various Netscape Plugin API data structures.

10231

10232

There is a drawing model to represent the old QuickDraw-based API. It is used by default

10233

if QuickDraw is available on the system, unless the plugin specifies another drawing model.

10234

10235

The big change is the addition of the CoreGraphics drawing model. A plugin may request this

10236

drawing model to obtain access to a CGContextRef for drawing, instead of a QuickDraw CGrafPtr.

10237

10238

* bindings/npapi.h:

10239

Define NP_NO_QUICKDRAW when compiling 64-bit; there is no 64-bit QuickDraw.

10240

Added NPNVpluginDrawingModel, NPNVsupportsQuickDrawBool, and NPNVsupportsCoreGraphicsBool

10241

variables.

10242

Added NPDrawingModel enumeration. Currently the only drawing models are QuickDraw and

10243

CoreGraphics.

10244

NPRegion's type now depends on the drawing model specified by the plugin.

10245

NP_Port is now only defined when QuickDraw is available.

10246

Added NP_CGContext, which is the type of the NPWindow's "window" member in CoreGraphics mode.

10247

10248

2006-05-13 Kevin M. Ollivier <kevino@theolliviers.com>

10249

10250

Reviewed by Darin, landed by ap.

10251

10252

- http://bugs.webkit.org/show_bug.cgi?id=8528

10253

Bakefiles (and generated Makefiles) for wx and gdk ports

10254

10255

* make-generated-sources.sh:

10256

Added script to configure environment to run DerivedSources.make

10257

10258

* JavaScriptCoreSources.bkl:

10259

Added JavaScriptCore sources list for Bakefile.

10260

10261

* jscore.bkl:

10262

Bakefile used to generate JavaScriptCore project files

10263

(currently only used by wx and gdk ports)

10264

10265

2006-05-09 Steve Falkenburg <sfalken@apple.com>

10266

10267

Fix Windows build.

10268

Minor fixes to WTF headers.

10269

10270

Reviewed by kevin.

10271

10272

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: Fix include dirs, paths to files.

10273

* JavaScriptCore.vcproj/testkjs/testkjs.vcproj: Fix include dirs.

10274

* wtf/Assertions.h: include Platform.h to get definition for COMPILER()

10275

* wtf/Vector.h: include FastMalloc.h for definition of fastMalloc, fastFree

10276

10277

2006-05-09 Maciej Stachowiak <mjs@apple.com>

10278

10279

Rubber stamped by Anders.

10280

10281

- renamed kxmlcore to wtf

10282

10283

kxmlcore --> wtf

10284

KXMLCore --> WTF

10285

KXC --> WTF

10286

10287

* JavaScriptCore.xcodeproj/project.pbxproj:

10288

* bindings/c/c_instance.cpp:

10289

* bindings/objc/WebScriptObject.mm:

10290

* kjs/JSImmediate.h:

10291

* kjs/Parser.cpp:

10292

* kjs/Parser.h:

10293

* kjs/array_object.cpp:

10294

* kjs/collector.cpp:

10295

(KJS::Collector::registerThread):

10296

* kjs/collector.h:

10297

* kjs/config.h:

10298

* kjs/function.cpp:

10299

(KJS::isStrWhiteSpace):

10300

* kjs/function.h:

10301

* kjs/identifier.cpp:

10302

* kjs/internal.cpp:

10303

* kjs/internal.h:

10304

* kjs/lexer.cpp:

10305

(Lexer::shift):

10306

(Lexer::isWhiteSpace):

10307

(Lexer::isIdentStart):

10308

(Lexer::isIdentPart):

10309

* kjs/lookup.cpp:

10310

* kjs/nodes.cpp:

10311

* kjs/nodes.h:

10312

* kjs/number_object.cpp:

10313

* kjs/object.h:

10314

* kjs/property_map.cpp:

10315

* kjs/property_map.h:

10316

* kjs/string_object.cpp:

10317

(StringProtoFunc::callAsFunction):

10318

* kjs/testkjs.cpp:

10319

(testIsInteger):

10320

* kjs/ustring.cpp:

10321

* kjs/ustring.h:

10322

* kxmlcore: Removed.

10323

* kxmlcore/AlwaysInline.h: Removed.

10324

* kxmlcore/Assertions.cpp: Removed.

10325

* kxmlcore/Assertions.h: Removed.

10326

* kxmlcore/FastMalloc.cpp: Removed.

10327

* kxmlcore/FastMalloc.h: Removed.

10328

* kxmlcore/FastMallocInternal.h: Removed.

10329

* kxmlcore/Forward.h: Removed.

10330

* kxmlcore/HashCountedSet.h: Removed.

10331

* kxmlcore/HashFunctions.h: Removed.

10332

* kxmlcore/HashMap.h: Removed.

10333

* kxmlcore/HashSet.h: Removed.

10334

* kxmlcore/HashTable.cpp: Removed.

10335

* kxmlcore/HashTable.h: Removed.

10336

* kxmlcore/HashTraits.h: Removed.

10337

* kxmlcore/ListRefPtr.h: Removed.

10338

* kxmlcore/Noncopyable.h: Removed.

10339

* kxmlcore/OwnArrayPtr.h: Removed.

10340

* kxmlcore/OwnPtr.h: Removed.

10341

* kxmlcore/PassRefPtr.h: Removed.

10342

* kxmlcore/Platform.h: Removed.

10343

* kxmlcore/RefPtr.h: Removed.

10344

* kxmlcore/TCPageMap.h: Removed.

10345

* kxmlcore/TCSpinLock.h: Removed.

10346

* kxmlcore/TCSystemAlloc.cpp: Removed.

10347

* kxmlcore/TCSystemAlloc.h: Removed.

10348

* kxmlcore/UnusedParam.h: Removed.

10349

* kxmlcore/Vector.h: Removed.

10350

* kxmlcore/VectorTraits.h: Removed.

10351

* kxmlcore/unicode: Removed.

10352

* kxmlcore/unicode/Unicode.h: Removed.

10353

* kxmlcore/unicode/UnicodeCategory.h: Removed.

10354

* kxmlcore/unicode/icu: Removed.

10355

* kxmlcore/unicode/icu/UnicodeIcu.h: Removed.

10356

* kxmlcore/unicode/posix: Removed.

10357

* kxmlcore/unicode/qt3: Removed.

10358

* kxmlcore/unicode/qt4: Removed.

10359

* kxmlcore/unicode/qt4/UnicodeQt4.h: Removed.

10360

* pcre/pcre_get.c:

10361

* wtf: Added.

10362

* wtf/Assertions.cpp:

10363

* wtf/Assertions.h:

10364

* wtf/FastMalloc.cpp:

10365

(WTF::TCMalloc_ThreadCache::Scavenge):

10366

(WTF::do_malloc):

10367

(WTF::do_free):

10368

(WTF::TCMallocGuard::TCMallocGuard):

10369

(WTF::malloc):

10370

(WTF::free):

10371

(WTF::calloc):

10372

(WTF::cfree):

10373

(WTF::realloc):

10374

* wtf/FastMalloc.h:

10375

* wtf/FastMallocInternal.h:

10376

* wtf/Forward.h:

10377

* wtf/HashCountedSet.h:

10378

* wtf/HashFunctions.h:

10379

* wtf/HashMap.h:

10380

* wtf/HashSet.h:

10381

* wtf/HashTable.cpp:

10382

* wtf/HashTable.h:

10383

* wtf/HashTraits.h:

10384

* wtf/ListRefPtr.h:

10385

* wtf/Noncopyable.h:

10386

* wtf/OwnArrayPtr.h:

10387

* wtf/OwnPtr.h:

10388

* wtf/PassRefPtr.h:

10389

* wtf/RefPtr.h:

10390

* wtf/TCSystemAlloc.cpp:

10391

(TCMalloc_SystemAlloc):

10392

* wtf/Vector.h:

10393

* wtf/VectorTraits.h:

10394

* wtf/unicode/UnicodeCategory.h:

10395

* wtf/unicode/icu/UnicodeIcu.h:

10396

10397

2006-05-08 Timothy Hatcher <timothy@apple.com>

10398

10399

Reviewed by Tim O.

10400

10401

* bindings/npapi.h: do not define #pragma options align=mac68k if we are 64-bit

10402

10403

2006-05-07 Darin Adler <darin@apple.com>

10404

10405

Reviewed and landed by Maciej.

10406

10407

- fix http://bugs.webkit.org/show_bug.cgi?id=8765

10408

Random crashes on TOT since the form state change

10409

10410

I haven't figured out how to construct a test for this, but this does seem to fix the

10411

problem; Mitz mentioned that a double-destroy was occurring in these functions.

10412

10413

* kxmlcore/HashMap.h: (KXMLCore::HashMap::remove): Use RefCounter::deref instead of calling

10414

~ValueType, because ~ValueType often results in a double-destroy, since the HashTable also

10415

destroys the element based on the storage type. The RefCounter template correctly does work

10416

only in cases where ValueType and ValueStorageType differ and this class is what's used

10417

elsewhere for the same purpose; I somehow missed this case when optimizing HashMap.

10418

* kxmlcore/HashSet.h: (KXMLCore::HashSet::remove): Ditto.

10419

10420

2006-05-05 Darin Adler <darin@apple.com>

10421

10422

- http://bugs.webkit.org/show_bug.cgi?id=8722

10423

IE compatibility fix in date parsing

10424

10425

* kjs/date_object.cpp: (KJS::parseDate): Merged change that George Staikos provided

10426

from KDE 3.4.3 branch that allows day values of 0 and values that are > 1000.

10427

10428

2006-05-04 Anders Carlsson <andersca@mac.com>

10429

10430

Reviewed by Maciej.

10431

10432

http://bugs.webkit.org/show_bug.cgi?id=8734

10433

Would like a Vector::append that takes another Vector

10434

10435

* kxmlcore/Vector.h:

10436

(KXMLCore::::append):

10437

New function that takes another array.

10438

10439

2006-05-02 Steve Falkenburg <sfalken@apple.com>

10440

10441

Reviewed by eric.

10442

10443

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: set NDEBUG for release build

10444

* kxmlcore/FastMalloc.cpp: Prevent USE_SYSTEM_MALLOC from being defined twice

10445

10446

2006-05-02 Anders Carlsson <andersca@mac.com>

10447

10448

Reviewed by Maciej.

10449

10450

* kxmlcore/HashMap.h:

10451

(KXMLCore::::operator):

10452

Return *this

10453

10454

2006-05-01 Tim Omernick <timo@apple.com>

10455

10456

Reviewed by Tim Hatcher.

10457

10458

<rdar://problem/4476875> Support printing for embedded Netscape plugins

10459

10460

* bindings/npapi.h:

10461

Fixed struct alignment problem in our npapi.h. Structs must be 68k-aligned on both pre-Mac OS X

10462

and Mac OS X systems, as this is what plugins expect.

10463

10464

2006-05-01 Timothy Hatcher <timothy@apple.com>

10465

10466

Reviewed by Maciej.

10467

10468

<rdar://problem/4308243> 8F36 Regression: crash in malloc_consolidate if you use a .PAC file

10469

10470

The original fix missed the oversized cell case. Added a test for "currentThreadIsMainThread ||

10471

imp->m_destructorIsThreadSafe" where we collect oversized cells.

10472

10473

We don't have a way to test PAC files yet, so there's no test attached.

10474

10475

* kjs/collector.cpp:

10476

(KJS::Collector::collect): test the thread when we collect oversized cells

10477

10478

2006-05-01 Tim Omernick <timo@apple.com>

10479

10480

Reviewed by Adele.

10481

10482

<rdar://problem/4526114> REGRESSION (two days ago): LOG() just prints @ for NSObject substitutions

10483

10484

* kxmlcore/Assertions.cpp:

10485

Changed sense of strstr("%@") check. I already made the same fix to the WebBrowser assertions.

10486

10487

2006-04-28 Steve Falkenburg <sfalken@apple.com>

10488

10489

Reviewed by kdecker

10490

10491

Actually apply the change that was reviewed insted of checking it in with an #if 0 (oops).

10492

10493

* kjs/testkjs.cpp:

10494

(main): Suppress C runtime alerts

10495

10496

2006-04-28 Steve Falkenburg <sfalken@apple.com>

10497

10498

Reviewed by kdecker

10499

10500

Suppress error reporting dialog that blocks Javascript tests from completing.

10501

10502

Real error is due to an overflow in the date/time handling functions that needs

10503

to be addressed, but this will prevent the hang running the Javascript tests

10504

on the build bot (along with the related changes).

10505

10506

* kjs/testkjs.cpp:

10507

(main): Suppress C runtime alerts

10508

10509

2006-04-27 Geoffrey Garen <ggaren@apple.com>

10510

10511

Reviewed by Maciej

10512

10513

- Minor fixups I discovered while working on the autogenerator.

10514

10515

* kjs/lookup.cpp:

10516

(findEntry): ASSERT that size is not 0, because otherwise we'll % by 0,

10517

compute a garbage address, and possibly crash.

10518

* kjs/lookup.h:

10519

(cacheGlobalObject): Don't enumerate cached objects -- ideally, they

10520

would be hidden entirely.

10521

10522

2006-04-21 Kevin M. Ollivier <kevino@theolliviers.com>

10523

10524

Reviewed by Darin.

10525

10526

- http://bugs.webkit.org/show_bug.cgi?id=8507

10527

Compilation fixes for building on gcc 4.0.2, and without precomp headers

10528

10529

* kjs/operations.h:

10530

* kxmlcore/Assertions.cpp:

10531

* kxmlcore/FastMalloc.cpp:

10532

Added necessary headers to resolve compilation issues when not using

10533

precompiled headers.

10534

10535

* kjs/value.h: Declare the JSCell class before friend declaration

10536

to resolve compilation issues with gcc 4.0.2.

10537

10538

* kxmlcore/Platform.h: Set Unicode support to use ICU on platforms

10539

other than KDE (previously only defined for Win and Mac OS)

10540

10541

2006-04-18 Eric Seidel <eseidel@apple.com>

10542

10543

Reviewed by ggaren.

10544

10545

Fix "new Function()" to correctly use lexical scoping.

10546

Add ScopeChain::print() function for debugging.

10547

<rdar://problem/4067864> REGRESSION (125-407): JavaScript failure on PeopleSoft REN Server

10548

10549

* kjs/function_object.cpp:

10550

(FunctionObjectImp::construct):

10551

* kjs/scope_chain.cpp:

10552

(KJS::ScopeChain::print):

10553

* kjs/scope_chain.h:

10554

10555

2006-04-14 James G. Speth <speth@end.com>

10556

10557

Reviewed by Timothy.

10558

10559

Bug 8389: support for Cocoa bindings - binding an NSTreeController to the WebView's DOM

10560

http://bugs.webkit.org/show_bug.cgi?id=8389

10561

10562

Adds a category to WebScriptObject with array accessors for KVC/KVO.

10563

10564

If super valueForKey: fails it will call valueForUndefinedKey:, which is

10565

important because it causes the right behavior to happen with bindings using

10566

the "Raises for Not Applicable Keys" flag and the "Not Applicable Placeholder"

10567

10568

* bindings/objc/WebScriptObject.mm:

10569

(-[WebScriptObject valueForKey:]):

10570

(-[WebScriptObject count]):

10571

(-[WebScriptObject objectAtIndex:]):

10572

(-[WebUndefined description]): return "undefined"

10573

10574

2006-04-13 Geoffrey Garen <ggaren@apple.com>

10575

10576

Reviewed by Darin.

10577

10578

* kjs/internal.cpp:

10579

(KJS::InterpreterImp::initGlobalObject): Add the built-in object

10580

prototype to the end of the global object's prototype chain instead of

10581

just blowing away its existing prototype. We need to do this because

10582

the window object has a meaningful prototype now.

10583

10584

2006-04-13 Maciej Stachowiak <mjs@apple.com>

10585

10586

Reviewed by Geoff.

10587

10588

- fix testkjs to not show false-positive KJS::Node leaks in debug builds

10589

10590

* kjs/testkjs.cpp:

10591

(doIt):

10592

(kjsmain):

10593

10594

2006-04-11 Geoffrey Garen <ggaren@apple.com>

10595

10596

Reviewed by Maciej.

10597

10598

Minor code cleanup -- passes all the JS tests.

10599

10600

* kjs/object_object.cpp:

10601

(ObjectObjectImp::construct):

10602

(ObjectObjectImp::callAsFunction):

10603

10604

2006-04-11 Darin Adler <darin@apple.com>

10605

10606

- another attempt to fix Windows build -- Vector in Forward.h was not working

10607

10608

* kxmlcore/Forward.h: Remove Vector.

10609

* kxmlcore/Vector.h: Add back default arguments, remove include of

10610

Forward.h.

10611

10612

2006-04-11 Darin Adler <darin@apple.com>

10613

10614

- try to fix Windows build -- HashForward.h was not working

10615

10616

* kxmlcore/HashForward.h: Removed.

10617

10618

* JavaScriptCore.xcodeproj/project.pbxproj: Remove HashForward.h.

10619

* kjs/collector.h: Remove use of HashForward.h.

10620

* kxmlcore/HashCountedSet.h: Remove include of HashForward.h, restore

10621

default arguments.

10622

* kxmlcore/HashMap.h: Ditto.

10623

* kxmlcore/HashSet.h: Ditto.

10624

10625

2006-04-11 David Harrison <harrison@apple.com>

10626

10627

Reviewed by Darin.

10628

10629

- fixed clean build, broken by Darin's check-in

10630

10631

* kjs/date_object.cpp: Add needed include of lookup.h.

10632

* kjs/regexp_object.cpp: Move include of .lut.h file below other includes.

10633

10634

2006-04-10 Darin Adler <darin@apple.com>

10635

10636

Rubber-stamped by John Sullivan.

10637

10638

- switched from a shell script to a makefile for generated files

10639

- removed lots of unneeded includes

10640

- added new Forward.h and HashForward.h headers that allow compiling with

10641

fewer unneeded templates

10642

10643

* DerivedSources.make: Added.

10644

* generate-derived-sources: Removed.

10645

* JavaScriptCore.xcodeproj/project.pbxproj: Added new files, changed to use

10646

DerivedSources.make.

10647

10648

* kxmlcore/Forward.h: Added.

10649

* kxmlcore/HashForward.h: Added.

10650

10651

* kxmlcore/HashCountedSet.h: Include HashForward for default args.

10652

* kxmlcore/HashMap.h: Ditto.

10653

* kxmlcore/HashSet.h: Ditto.

10654

10655

* kjs/object.h:

10656

* kjs/object.cpp:

10657

Moved KJS_MAX_STACK into the .cpp file.

10658

10659

* bindings/NP_jsobject.cpp:

10660

* bindings/c/c_instance.h:

10661

* bindings/jni/jni_class.h:

10662

* bindings/jni/jni_runtime.h:

10663

* bindings/jni/jni_utility.h:

10664

* bindings/objc/WebScriptObject.mm:

10665

* bindings/objc/WebScriptObjectPrivate.h:

10666

* bindings/objc/objc_class.h:

10667

* bindings/objc/objc_class.mm:

10668

* bindings/objc/objc_instance.h:

10669

* bindings/objc/objc_instance.mm:

10670

* bindings/objc/objc_runtime.mm:

10671

* bindings/objc/objc_utility.mm:

10672

* bindings/runtime.h:

10673

* bindings/runtime_array.cpp:

10674

* bindings/runtime_array.h:

10675

* bindings/runtime_method.cpp:

10676

* bindings/runtime_method.h:

10677

* bindings/runtime_object.cpp:

10678

* bindings/runtime_root.h:

10679

* kjs/JSImmediate.cpp:

10680

* kjs/Parser.h:

10681

* kjs/array_object.cpp:

10682

* kjs/array_object.h:

10683

* kjs/bool_object.cpp:

10684

* kjs/bool_object.h:

10685

* kjs/collector.h:

10686

* kjs/context.h:

10687

* kjs/debugger.cpp:

10688

* kjs/error_object.h:

10689

* kjs/function_object.h:

10690

* kjs/internal.h:

10691

* kjs/lexer.cpp:

10692

* kjs/math_object.cpp:

10693

* kjs/math_object.h:

10694

* kjs/nodes.cpp:

10695

* kjs/nodes.h:

10696

* kjs/number_object.cpp:

10697

* kjs/number_object.h:

10698

* kjs/object_object.cpp:

10699

* kjs/operations.cpp:

10700

* kjs/protected_reference.h:

10701

* kjs/reference.h:

10702

* kjs/reference_list.h:

10703

* kjs/regexp_object.h:

10704

* kjs/string_object.cpp:

10705

* kjs/string_object.h:

10706

* kjs/testkjs.cpp:

10707

* kjs/value.cpp:

10708

* kjs/value.h:

10709

* kxmlcore/HashTable.h:

10710

* kxmlcore/ListRefPtr.h:

10711

* kxmlcore/TCPageMap.h:

10712

* kxmlcore/Vector.h:

10713

Removed unneeded header includes.

10714

10715

2006-04-09 Geoffrey Garen <ggaren@apple.com>

10716

10717

Reviewed by eric.

10718

10719

- Fixed http://bugs.webkit.org/show_bug.cgi?id=8284

10720

prevent unnecessary entries in the "nodes with extra refs" hash table

10721

10722

This patch switches manually RefPtr exchange with use of

10723

RefPtr::release to ensure that a node's ref count never tops 1

10724

(in the normal case).

10725

10726

* kjs/nodes.cpp:

10727

(BlockNode::BlockNode):

10728

(CaseBlockNode::CaseBlockNode):

10729

* kjs/nodes.h:

10730

(KJS::ArrayNode::ArrayNode):

10731

(KJS::ObjectLiteralNode::ObjectLiteralNode):

10732

(KJS::ArgumentsNode::ArgumentsNode):

10733

(KJS::VarStatementNode::VarStatementNode):

10734

(KJS::ForNode::ForNode):

10735

(KJS::CaseClauseNode::CaseClauseNode):

10736

(KJS::FuncExprNode::FuncExprNode):

10737

(KJS::FuncDeclNode::FuncDeclNode):

10738

10739

2006-04-08 Alexey Proskuryakov <ap@nypop.com>

10740

10741

Reviewed by Darin.

10742

10743

One more attempt - use reinterpret_cast, rather than static_cast.

10744

10745

2006-04-08 Alexey Proskuryakov <ap@nypop.com>

10746

10747

Reviewed by Darin.

10748

10749

An attempt to fix Win32 build - ICU uses wchar_t on Windows, so we need a type cast.

10750

10751

* kxmlcore/unicode/icu/UnicodeIcu.h:

10752

(KXMLCore::Unicode::toLower):

10753

(KXMLCore::Unicode::toUpper):

10754

10755

2006-04-08 Alexey Proskuryakov <ap@nypop.com>

10756

10757

Reviewed by Darin.

10758

10759

- fix http://bugs.webkit.org/show_bug.cgi?id=8264

10760

toLowerCase and toUpperCase don't honor special mappings

10761

10762

Test: fast/js/string-capitalization.html

10763

10764

* JavaScriptCore.xcodeproj/project.pbxproj: Added KXMLCore::Unicode headers to the project.

10765

* icu/unicode/putil.h: Added (copied from WebCore).

10766

* icu/unicode/uiter.h: Ditto.

10767

* icu/unicode/ustring.h: Ditto.

10768

* kjs/string_object.cpp:

10769

(StringProtoFunc::callAsFunction): Use the new KXMLCore::Unicode::toUpper() and toLower().

10770

* kjs/ustring.cpp: Removed unused (and evil) UChar::toLower() and toUpper().

10771

* kjs/ustring.h: Ditto.

10772

10773

* kxmlcore/unicode/Unicode.h: Corrected capitalization of the word Unicode.

10774

* kxmlcore/unicode/UnicodeCategory.h: Renamed include guard macro to match file name.

10775

10776

* kxmlcore/unicode/icu/UnicodeIcu.h:

10777

(KXMLCore::Unicode::toLower): Work on strings, not individual characters. Use ICU root locale.

10778

(KXMLCore::Unicode::toUpper): Ditto.

10779

(KXMLCore::Unicode::isFormatChar): Use int32_t, which can hold a complete code point.

10780

(KXMLCore::Unicode::isSeparatorSpace): Ditto.

10781

(KXMLCore::Unicode::category): Ditto.

10782

* kxmlcore/unicode/qt4/UnicodeQt4.h:

10783

(KXMLCore::Unicode::toLower): Work on strings, not individual characters.

10784

(KXMLCore::Unicode::toUpper): Ditto.

10785

(KXMLCore::Unicode::isFormatChar): Use int32_t, which can hold a complete code point.

10786

(KXMLCore::Unicode::isSeparatorSpace): Ditto.

10787

(KXMLCore::Unicode::category): Ditto.

10788

10789

* tests/mozilla/ecma/String/15.5.4.12-1.js: Corrected expected results.

10790

* tests/mozilla/ecma/String/15.5.4.12-5.js: Corrected expected results.

10791

10792

2006-04-05 Darin Adler <darin@apple.com>

10793

10794

- attempt to fix Windows build

10795

10796

* kxmlcore/HashMap.h: (KXMLCore::HashMap::remove): Use (*it). instead of it->.

10797

* kxmlcore/HashSet.h: (KXMLCore::HashSet::remove): Ditto.

10798

10799

2006-04-05 Darin Adler <darin@apple.com>

10800

10801

- attempt to fix Windows build

10802

10803

* os-win32/stdint.h: Add int8_t, uint8_t, int64_t.

10804

10805

2006-04-05 Darin Adler <darin@apple.com>

10806

10807

Reviewed by Maciej.

10808

10809

- fix memory leak introduced by the previous change

10810

10811

* kxmlcore/HashTable.h: Specialize NeedsRef so that it correctly returns true when

10812

the value in question is a pair where one of the pair needs a ref and the other

10813

of the pair does not.

10814

10815

2006-04-05 Darin Adler <darin@apple.com>

10816

10817

Reviewed by Maciej.

10818

10819

- JavaScriptCore part of fix for http://bugs.webkit.org/show_bug.cgi?id=8049

10820

StringImpl hash traits deleted value creates an init routine for WebCore

10821

<rdar://problem/4442248> REGRESSION: WebCore has init routines (8049)

10822

10823

Change HashMap and HashSet implementation so they fold various types together.

10824

This allows us to implement maps and sets that use RefPtr<WebCore::StringImpl>

10825

and WebCore::String in terms of the underlying raw pointer type, and hence use

10826

-1 for the deleted value.

10827

10828

* kxmlcore/HashTraits.h: Added a new type to HashTraits, StorageTraits, which is a

10829

type to be used when storing a value that has the same layout as the type itself.

10830

This is used only for non-key cases. In the case of keys, the hash function must also

10831

be considered. Moved emptyValue out of GenericHashTraitsBase into GenericHashTraits.

10832

Added a new bool to HashTraits, needsRef, which indicates whether the type needs

10833

explicit reference counting. If the type itself has needsRef true, but the storage

10834

type has needsRef false, then the HashSet or HashMap has to handle the reference

10835

counting explicitly. Added hash trait specializations for all signed integer values

10836

that give -1 as the deleted value. Gave all integers StorageTraits of the canonical

10837

integer type of the same size so int and long will share code. Gave all pointers and

10838

RefPtrs StorageTraits of the appropriately sized integer type. Removed redundant

10839

TraitType and emptyValue definitions in the pointer specialization for HashTraits.

10840

Added PairBaseHashTraits, which doesn't try to set up needsDestruction and deletedValue.

10841

Useful for types where we don't want to force the existence of deletedValue, such as

10842

the type of a pair in a HashMap which is not the actual storage type. Removed an

10843

unneeded parameter from the DeletedValueAssigner template. Added HashKeyStorageTraits

10844

template, which determines what type can be used to store a given hash key type with

10845

a given hash function, and specialized it for pointers and RefPtr so that pointer

10846

hash tables share an underlying HashTable that uses IntHash.

10847

10848

* kxmlcore/HashTable.h: Added HashTableConstIteratorAdapter, HashTableIteratorAdapter,

10849

NeedsRef, RefCountManagerBase, RefCountManager, HashTableRefCountManagerBase, and

10850

HashTableRefCountManager. All are used by both HashSet and HashMap to handle hash

10851

tables where the type stored is not the same as the real value type.

10852

10853

* kxmlcore/HashFunctions.h: Added a new struct named IntTypes that finds an

10854

integer type given a sizeof value. Renamed pointerHash to intHash and made it

10855

use overloading and take integer parameters. Added an IntHash struct which is

10856

a hash function that works for integers. Changed PtrHash to call IntHash with

10857

an appropriately sized integer. Made IntHash the default hash function for

10858

many integer types. Made PtrHash the default hash function for RefPtr as well

10859

as for raw pointers.

10860

10861

* kxmlcore/HashSet.h: Changed implementation to use a separate "storage type"

10862

derived from the new traits. The HashTable will use the storage type and all

10863

necessary translation and ref/deref is done at the HashSet level. Also reorganized

10864

the file so that the HashSet is at the top and has no inline implementation inside

10865

it so it's easy to read the interface to HashSet.

10866

10867

* kxmlcore/HashMap.h: Changed implementation to use a separate "storage type"

10868

derived from the new traits. The HashTable will use the storage type and all

10869

necessary translation and ref/deref is done at the HashMap level. Also reorganized

10870

the file so that the HashMap is at the top and has no inline implementation inside

10871

it so it's easy to read the interface to HashMap.

10872

10873

* kxmlcore/HashMapPtrSpec.h: Removed. Superceded by optimizations in HashMap itself.

10874

10875

* JavaScriptCore.xcodeproj/project.pbxproj: Remove HashMapPtrSpec.h, resort files,

10876

and also remove some unnecessary build settings from the aggregate target that

10877

generates derived sources.

10878

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: Ditto.

10879

10880

2006-04-04 Timothy Hatcher <timothy@apple.com>

10881

10882

Reviewed by Darin.

10883

10884

The Debug and Release frameworks are now built with install paths relative to the build products directory.

10885

This removes the need for other projects to build with -framework WebCore and -framework JavaScriptCore.

10886

10887

* JavaScriptCore.xcodeproj/project.pbxproj:

10888

10889

2006-04-04 Eric Seidel <eseidel@apple.com>

10890

10891

Reviewed by ggaren.

10892

10893

Fix win32 build.

10894

Disable ASSERT redefinition warnings for now.

10895

10896

* JavaScriptCore.vcproj/testkjs/testkjs.vcproj:

10897

* kxmlcore/Assertions.h:

10898

10899

2006-04-04 Bjrn Graf <bjoern.graf@gmail.com>

10900

10901

Reviewed by ggaren & darin. Landed by eseidel.

10902

10903

Integrate CURL version of gettimeofday

10904

http://bugs.webkit.org/show_bug.cgi?id=7399

10905

Disable crash report dialogs for testkjs.exe in Release mode

10906

http://bugs.webkit.org/show_bug.cgi?id=8113

10907

10908

* kjs/testkjs.cpp:

10909

(StopWatch::start):

10910

(StopWatch::stop):

10911

(StopWatch::getElapsedMS):

10912

(main):

10913

(kjsmain):

10914

10915

2006-04-04 Eric Seidel <eseidel@apple.com>

10916

10917

Reviewed by mjs.

10918

10919

* kjs/number_object.cpp:

10920

(NumberProtoFunc::callAsFunction): remove trunc() to fix win32.

10921

10922

2006-03-12 Maciej Stachowiak <mjs@apple.com>

10923

10924

Reviewed by Darin.

10925

10926

- fixed "toPrecision sometimes messes up the last digit on intel Macs"

10927

http://bugs.webkit.org/show_bug.cgi?id=7748

10928

10929

* kjs/number_object.cpp:

10930

(intPow10): Compute integer powers of 10 using exponentiation by squaring.

10931

(NumberProtoFunc::callAsFunction): Use intPow10(n) in place of all pow(10.0, n),

10932

plus a bit of refactoring.

10933

10934

2006-04-03 Darin Adler <darin@apple.com>

10935

10936

- tweak config.h and Platform.h to try to get buildbot working

10937

(making some small changes at the same time)

10938

10939

* kjs/config.h: Removed now-unneeded HAVE_ICU.

10940

* kxmlcore/Platform.h: Tweak how platform gets set up. Move all the

10941

USE stuff to the end.

10942

10943

2006-04-03 George Staikos <staikos@opensource.apple.com>

10944

10945

Reviewed by Maciej.

10946

10947

Fix Win32 build breakage from previous commit, remove unused forward.

10948

10949

2006-04-03 George Staikos <staikos@opensource.apple.com>

10950

10951

Reviewed by Maciej.

10952

10953

Implement a unicode abstraction layer to make JavaScriptCore much more

10954

easily ported to other platforms without having to take in libicu. Also

10955

makes the unicode related code easier to understand.

10956

10957

2006-04-03 Timothy Hatcher <timothy@apple.com>

10958

10959

Reviewed by Adele.

10960

10961

Fixes <rdar://problem/4498338> JavaScriptCore fails to compile for ppc64

10962

Other 64 bit build fixes.

10963

10964

* kjs/collector.cpp:

10965

(KJS::Collector::markOtherThreadConservatively): test for __DARWIN_UNIX03 and use __r1

10966

* kjs/dtoa.cpp:

10967

(Bigint::): cast PRIVATE_mem to unsigned to prevent warning

10968

* bindings/jni/jni_utility.cpp:

10969

(KJS::Bindings::getJavaVM): cast jniError to long to prevent format warning

10970

(KJS::Bindings::getJNIEnv): cast jniError to long to prevent format warning

10971

* bindings/runtime_root.cpp:

10972

(KJS::Bindings::addNativeReference): cast CFDictionaryGetValue to unsigned long to prevent warning

10973

(KJS::Bindings::removeNativeReference): cast CFDictionaryGetValue to unsigned long to prevent warning

10974

10975

2006-03-31 Darin Adler <darin@apple.com>

10976

10977

Reviewed by Geoff.

10978

10979

- <rdar://problem/4395622> API: WebScriptObject.h incorrectly reports that -isSelectorExcludedFromWebScript returns NO by default

10980

10981

* bindings/objc/WebScriptObject.h: Fixed comment.

10982

10983

2006-03-31 Eric Seidel <eseidel@apple.com>

10984

10985

Reviewed by mjs.

10986

10987

A bit more code cleanup.

10988

10989

* bindings/c/c_utility.cpp:

10990

(KJS::Bindings::convertValueToNPVariant):

10991

* bindings/objc/objc_runtime.mm:

10992

(convertValueToObjcObject):

10993

* bindings/objc/objc_utility.mm:

10994

(KJS::Bindings::convertValueToObjcValue):

10995

* kjs/function.cpp:

10996

(KJS::GlobalFuncImp::callAsFunction):

10997

* kjs/interpreter.cpp:

10998

(KJS::ExecState::lexicalInterpreter):

10999

* kjs/interpreter.h:

11000

* kjs/operations.cpp:

11001

(KJS::equal):

11002

11003

2006-03-30 Eric Seidel <eseidel@apple.com>

11004

11005

Reviewed by anders.

11006

11007

Small code-style update.

11008

11009

* kjs/operations.cpp:

11010

(KJS::isNaN):

11011

(KJS::isInf):

11012

(KJS::isPosInf):

11013

(KJS::isNegInf):

11014

(KJS::equal):

11015

(KJS::strictEqual):

11016

(KJS::relation):

11017

(KJS::maxInt):

11018

(KJS::minInt):

11019

(KJS::add):

11020

(KJS::mult):

11021

11022

2006-03-31 Anders Carlsson <andersca@mac.com>

11023

11024

Reviewed by Maciej.

11025

11026

Make sure the GetterSetterImp objects are marked as well.

11027

11028

* kjs/internal.cpp:

11029

(KJS::GetterSetterImp::mark):

11030

Call JSCell::mark().

11031

11032

2006-03-30 Eric Seidel <eseidel@apple.com>

11033

11034

Reviewed by ggaren.

11035

11036

* kjs/nodes.h: Some various small style fixes.

11037

11038

2006-03-30 Eric Seidel <eseidel@apple.com>

11039

11040

Reviewed by ggaren.

11041

11042

Clean-up style issues in node.h, remove redundant initializations.

11043

11044

* kjs/nodes.h:

11045

(KJS::StatementNode::evaluate):

11046

(KJS::ArrayNode::ArrayNode):

11047

(KJS::ObjectLiteralNode::ObjectLiteralNode):

11048

(KJS::ArgumentsNode::ArgumentsNode):

11049

(KJS::NewExprNode::NewExprNode):

11050

(KJS::CaseClauseNode::CaseClauseNode):

11051

(KJS::FuncDeclNode::FuncDeclNode):

11052

11053

2006-03-30 Tim Omernick <timo@apple.com>

11054

11055

Reviewed by Geoff.

11056

11057

<rdar://problem/4212626> REGRESSION: LIVECONNECT: JavaScript type for Java Strings is function,

11058

not object

11059

11060

* bindings/runtime.h:

11061

(KJS::Bindings::Instance::implementsCall):

11062

New method. Returns false by default. Concrete subclasses can override this return true when

11063

the bound object may be called as a function.

11064

(KJS::Bindings::Instance::invokeDefaultMethod):

11065

Since bound objects are no longer treated as functions by default, we can return jsUndefined()

11066

here instead of in concrete subclasses that decide not to implement the default method

11067

functionality.

11068

11069

* bindings/runtime_object.cpp:

11070

(RuntimeObjectImp::implementsCall):

11071

Don't assume that the bound object is a function; instead, let the object instance decide whether

11072

it is callable.

11073

11074

* bindings/c/c_instance.h:

11075

* bindings/c/c_instance.cpp:

11076

(KJS::Bindings::CInstance::implementsCall):

11077

The object is callable if its class has an invokeDefault function.

11078

11079

* bindings/objc/objc_instance.h:

11080

* bindings/objc/objc_instance.mm:

11081

(ObjcInstance::implementsCall):

11082

The object is callable if the ObjC instance responds to -invokeDefaultMethodWithArguments:.

11083

11084

* bindings/jni/jni_instance.h:

11085

* bindings/jni/jni_instance.cpp:

11086

Moved bogus invokeDefaultMethod() to superclass.

11087

11088

2006-03-29 Geoffrey Garen <ggaren@apple.com>

11089

11090

Reviewed by Darin.

11091

11092

- JavaScriptCore side of fix for <rdar://problem/4308243> 8F36

11093

Regression: crash in malloc_consolidate if you use a .PAC file

11094

11095

The crash was a result of threaded deallocation of thread-unsafe

11096

objects. Pure JS objects are thread-safe because all JS execution

11097

is synchronized through JSLock. However, JS objects that wrap WebCore

11098

objects are thread-unsafe because JS and WebCore execution are not

11099

synchronized. That unsafety comes into play when the collector

11100

deallocates a JS object that wraps a WebCore object, thus causing the

11101

WebCore object to be deallocated.

11102

11103

The solution here is to have each JSCell know whether it is safe to

11104

collect on a non-main thread, and to avoid collecting unsafe cells

11105

when on a non-main thread.

11106

11107

We don't have a way to test PAC files yet, so there's no test

11108

attached to this patch.

11109

11110

* kjs/collector.cpp:

11111

(KJS::Collector::collect):

11112

(1) Added the test "currentThreadIsMainThread ||

11113

imp->m_destructorIsThreadSafe".

11114

11115

* kjs/protect.h:

11116

(KJS::gcProtectNullTolerant):

11117

(KJS::gcUnprotectNullTolerant):

11118

* kjs/value.h:

11119

(KJS::JSCell::JSCell): The bools here must be bitfields, otherwise

11120

m_destructorIsThreadSafe becomes another whole word, ruining the

11121

collector optimizations we've made based on the size of a JSObject.

11122

* kxmlcore/FastMalloc.cpp:

11123

(KXMLCore::currentThreadIsMainThread):

11124

(KXMLCore::fastMallocRegisterThread):

11125

* kxmlcore/FastMalloc.h:

11126

11127

2006-03-28 Darin Adler <darin@apple.com>

11128

11129

Reviewed by Geoff.

11130

11131

- change some code that resulted in init routines on Mac OS X -- if the framework has

11132

init routines it will use memory and slow down applications that link with WebKit

11133

even in cases where those applications don't use WebKit

11134

11135

* kjs/date_object.cpp: Changed constants that were derived by multiplying other constants

11136

to use immediate numbers instead. Apparently, double constant expressions of the type we

11137

had here are evaluated at load time.

11138

11139

* kjs/list.cpp: Can't use OwnArrayPtr in ListImp because of the global instances of

11140

ListImp, so go back to using a plain old pointer.

11141

(KJS::List::List): Set overflow to 0 when initializing ListImp.

11142

(KJS::List::release): Replace a clear call with a delete and explicit set to 0.

11143

(KJS::List::append): Use raw pointers, and do a delete [] instead of finessing it with

11144

a swap of OwnArrayPtr.

11145

(KJS::List::copyFrom): Remove now-unneeded get().

11146

(KJS::List::copyTail): Ditto.

11147

11148

* kjs/ustring.cpp: Changed UString::Rep::empty initializer a bit so that it doesn't get

11149

a static initializer routine. Had to get rid of one level of constant to get the compiler

11150

to understand it could initialize without any code.

11151

11152

- added a build step that checks for init routines

11153

11154

* JavaScriptCore.xcodeproj/project.pbxproj: Deleted now-unused custom build rule that

11155

was replaced by the generate-derived-sources script a while back. Added a custom build

11156

phase that invokes the check-for-global-initializers script.

11157

11158

2006-03-28 Timothy Hatcher <timothy@apple.com>

11159

11160

Reviewed by Eric.

11161

11162

fixes <rdar://problem/4458539> Unable to include Security(public) and WebKit(private) headers

11163

11164

* bindings/npapi.h: added #defines after the #ifndefs

11165

11166

2006-03-27 Maciej Stachowiak <mjs@apple.com>

11167

11168

Reviewed by Anders.

11169

11170

- fixed <rdar://problem/4489745> REGRESSION: Safari crashes at to display http://www.lgphilips-lcd.com/

11171

11172

* kjs/nodes.cpp:

11173

(Node::deref): take into account the case where the extra refcount table was never created

11174

11175

2006-03-23 David Carson <dacarson@gmail.com>

11176

11177

Reviewed by Darin.

11178

11179

- JSObject in LiveConnect not working.

11180

http://bugs.webkit.org/show_bug.cgi?id=7917

11181

11182

* bindings/jni_jsobject.cpp:

11183

(JavaJSObject::convertJObjectToValue): Was trying to retrieve the native pointer from the wrong base

11184

class, and the GetFieldID was using the wrong signature.

11185

11186

2006-03-23 Darin Adler <darin@apple.com>

11187

11188

Reviewed by Maciej.

11189

11190

- fix buildbot

11191

11192

* JavaScriptCore.xcodeproj/project.pbxproj: Change target name to JavaScriptCore (it was "include"!?).

11193

Also add -Y 3 option for linker.

11194

11195

2006-03-23 Darin Adler <darin@apple.com>

11196

11197

Reviewed by Maciej.

11198

11199

- fix http://bugs.webkit.org/show_bug.cgi?id=7726

11200

REGRESSION: orbitz calendar fails (JavaScript function serialization/parsing)

11201

11202

* kjs/object.h: Take function name, as well as source URL and line number, when

11203

using the special overloaded construct for making functions.

11204

* kjs/object.cpp: (KJS::JSObject::construct): Ditto.

11205

* kjs/function_object.h: Ditto.

11206

* kjs/function_object.cpp: (FunctionObjectImp::construct): Pass a name when

11207

constructing the function rather than null. Use "anonymous" when making a

11208

function using the default function constructor.

11209

11210

* kjs/nodes2string.cpp: (FuncDeclNode::streamTo): Put a line break just before

11211

a function declaration.

11212

11213

- unrelated fix

11214

11215

* kxmlcore/HashMapPtrSpec.h: Add missing needed friend declaration.

11216

11217

2006-03-23 Darin Adler <darin@apple.com>

11218

11219

Reviewed by Maciej.

11220

11221

- fix http://bugs.webkit.org/show_bug.cgi?id=7805

11222

LEAK: method name leaks in KJS::Bindings::CInstance::invokeMethod

11223

11224

* bindings/c/c_utility.h: Remove NPN_UTF16FromString declaration (not implemented).

11225

* bindings/c/c_utility.cpp:

11226

(KJS::Bindings::convertValueToNPVariant): Use DOUBLE_TO_NPVARIANT,

11227

BOOLEAN_TO_NPVARIANT, VOID_TO_NPVARIANT, NULL_TO_NPVARIANT, and

11228

OBJECT_TO_NPVARIANT. In the case of OBJECT, call _NPN_RetainObject in

11229

one case and remove a _NPN_ReleaseObject in another because this

11230

should return a retained value.

11231

(KJS::Bindings::convertNPVariantToValue): Use NPVARIANT_TO_BOOLEAN,

11232

NPVARIANT_TO_INT32, and NPVARIANT_TO_DOUBLE.

11233

11234

* bindings/c/c_runtime.h: Removed implementations of CMethod::name and

11235

CField::name that called _NPN_UTF8FromIdentifier and hence leaked.

11236

* bindings/c/c_runtime.cpp:

11237

(KJS::Bindings::CMethod::name): Added. Returns the string from inside the

11238

method object.

11239

(KJS::Bindings::CField::name): Added. Returns the string from inside the

11240

field object.

11241

(KJS::Bindings::CField::valueFromInstance): Added call to _NPN_ReleaseVariantValue

11242

on the result of getProperty after using it to fix a storage leak.

11243

(KJS::Bindings::CField::setValueToInstance): Added call to _NPN_ReleaseVariantValue

11244

after pasing a value to setProperty now that the conversion function does a retain.

11245

11246

* bindings/c/c_instance.cpp:

11247

(KJS::Bindings::CInstance::invokeMethod): Changed to use Vector for a local

11248

stack buffer. Removed special case for NPVARIANT_IS_VOID because the

11249

convertNPVariantToValue function handles that properly.

11250

(KJS::Bindings::CInstance::invokeDefaultMethod): Ditto.

11251

11252

* bindings/NP_jsobject.h: Formatting changes only.

11253

* bindings/NP_jsobject.cpp:

11254

(jsDeallocate): Changed parameter type so we don't need a function cast.

11255

(_NPN_InvokeDefault): Use VOID_TO_NPVARIANT.

11256

(_NPN_Invoke): Use NULL_TO_NPVARIANT and VOID_TO_NPVARIANT.

11257

(_NPN_Evaluate): Use VOID_TO_NPVARIANT.

11258

(_NPN_GetProperty): Use NULL_TO_NPVARIANT and VOID_TO_NPVARIANT.

11259

11260

* bindings/c/c_class.cpp: Formatting changes only.

11261

* bindings/c/c_class.h: Formatting changes only.

11262

11263

* bindings/npruntime_priv.h: Removed obsolete and now-unused functions:

11264

NPN_VariantIsVoid, NPN_VariantIsNull, NPN_VariantIsUndefined,

11265

NPN_VariantIsBool, NPN_VariantIsInt32, NPN_VariantIsDouble,

11266

NPN_VariantIsString, NPN_VariantIsObject, NPN_VariantToBool,

11267

NPN_VariantToInt32, NPN_VariantToDouble, NPN_VariantToString,

11268

NPN_VariantToStringCopy, NPN_VariantToObject, NPN_InitializeVariantAsVoid,

11269

NPN_InitializeVariantAsNull, NPN_InitializeVariantAsUndefined,

11270

NPN_InitializeVariantWithBool, NPN_InitializeVariantWithInt32,

11271

NPN_InitializeVariantWithDouble, NPN_InitializeVariantWithString,

11272

NPN_InitializeVariantWithObject, and NPN_InitializeVariantWithVariant.

11273

* bindings/npruntime.cpp:

11274

(getIntIdentifierDictionary): Don't bother creating custom callbacks for the

11275

integer dictionary since the default behavior is fine for integers.

11276

11277

2006-03-23 Mark Rowe <opendarwin.org@bdash.net.nz>

11278

11279

Reviewed and landed by Maciej.

11280

11281

- WebKit no longer builds with bison 2.1

11282

http://bugs.webkit.org/show_bug.cgi?id=7923

11283

11284

* generate-derived-sources: Handle generated header named either grammar.cpp.h

11285

or grammar.hpp.

11286

11287

2006-03-22 Maciej Stachowiak <mjs@apple.com>

11288

11289

- fix the build

11290

11291

* JavaScriptCore.xcodeproj/project.pbxproj:

11292

11293

2006-03-21 Maciej Stachowiak <mjs@apple.com>

11294

11295

* kjs/generate-derived-sources: Set executable property.

11296

11297

2006-03-21 Maciej Stachowiak <mjs@apple.com>

11298

11299

Reviewed by Darin.

11300

11301

Ensure that generated source dependencies are handled properly, as follows:

11302

11303

- Made an external script that generates the sources into a

11304

DerivedSources dir in the build products directory.

11305

- Added a new build target that builds all the generated sources

11306

if needed. Sadly it has to be a target, not a phase for Xcode to notice changes.

11307

- Added the DerivedSources dir in question to the include path.

11308

- Added the new DerivedSources dir and its contents to the project as build-relative.

11309

11310

* JavaScriptCore.xcodeproj/project.pbxproj:

11311

* kjs/generate-derived-sources: Added. Based on the windows version - maybe someday they

11312

can share more.

11313

11314

2006-03-11 Maciej Stachowiak <mjs@apple.com>

11315

11316

Reviewed by Darin.

11317

11318

- fixed "charAt layout test fails on intel macs; some NaNs are printed as -NaN"

11319

http://bugs.webkit.org/show_bug.cgi?id=7745

11320

11321

* kjs/ustring.cpp:

11322

(KJS::UString::from): Use "NaN" for all NaN values, regardless of sign.

11323

11324

2006-03-16 Maciej Stachowiak <mjs@apple.com>

11325

11326

Reviewed by Darin.

11327

11328

- tweaks to my change to redo KJS::Node refcounting

11329

11330

* kjs/nodes.cpp:

11331

(Node::ref):

11332

(Node::deref):

11333

(Node::refcount):

11334

(Node::clearNewNodes):

11335

* kjs/nodes.h:

11336

11337

2006-03-16 Darin Adler <darin@apple.com>

11338

11339

Reviewed by Maciej.

11340

11341

- fixed Vector so that you can pass a reference to something in the vector

11342

to the append or insert functions

11343

11344

* kxmlcore/Vector.h:

11345

(KXMLCore::Vector::expandCapacity): Added new overloads that take a pointer to adjust

11346

and return the adjusted value of the pointer.

11347

(KXMLCore::Vector::append): Pass a pointer when expanding the vector, and use it when

11348

adding the new element. Makes the case where the element moves when the vector

11349

is expanded work.

11350

(KXMLCore::Vector::insert): Ditto.

11351

11352

2006-03-15 Eric Seidel <eseidel@apple.com>

11353

11354

Reviewed by adele.

11355

11356

Build fix.

11357

11358

* kjs/date_object.cpp:

11359

(KJS::DateProtoFunc::callAsFunction): use size() not "len()"

11360

11361

2006-03-15 Eric Seidel <eseidel@apple.com>

11362

11363

Reviewed by mjs.

11364

11365

Fix CString copy constructor, fixes Date.parse("") on Win32.

11366

11367

* kjs/date_object.cpp:

11368

(KJS::DateProtoFunc::callAsFunction):

11369

* kjs/ustring.cpp:

11370

(KJS::CString::CString):

11371

(KJS::CString::operator=):

11372

11373

2006-03-13 Maciej Stachowiak <mjs@apple.com>

11374

11375

Reviewed by Anders.

11376

11377

- KJS::Node and KJS::StatementNode are bigger than they need to be

11378

http://bugs.webkit.org/show_bug.cgi?id=7775

11379

11380

The memory usage of Node was reduced by 2 machine words per node:

11381

11382

- sourceURL was removed and only kept on FunctionBodyNode. The

11383

source URL can only be distinct per function or top-level program node,

11384

and you always have one.

11385

11386

- refcount was removed and kept in a separate hashtable when

11387

greater than 1. newNodes set represents floating nodes with

11388

refcount of 0. This helps because almost all nodes have a refcount of 1

11389

for almost all of their lifetime.

11390

11391

* bindings/runtime_method.cpp:

11392

(RuntimeMethod::RuntimeMethod): Pass null body, added FIXME.

11393

* kjs/Parser.cpp:

11394

(KJS::clearNewNodes): New nodes are tracked in nodes.cpp now, but still clear

11395

them at the appropriate time.

11396

* kjs/context.h:

11397

(KJS::ContextImp::currentBody): added; used to retrieve source URL and sid

11398

for current code.

11399

(KJS::ContextImp::pushIteration): moved here from LabelStack

11400

(KJS::ContextImp::popIteration): ditto

11401

(KJS::ContextImp::inIteration): ditto

11402

(KJS::ContextImp::pushSwitch): ditto

11403

(KJS::ContextImp::popSwitch): ditto

11404

(KJS::ContextImp::inSwitch): ditto

11405

* kjs/function.cpp:

11406

(KJS::FunctionImp::FunctionImp): Add FunctionBodyNode* parameter.

11407

(KJS::FunctionImp::callAsFunction): Pass body to ContextImp.

11408

(KJS::FunctionImp::argumentsGetter): _context renamed to m_context.

11409

(KJS::DeclaredFunctionImp::DeclaredFunctionImp): Pass body to superclass

11410

constructor.

11411

(KJS::GlobalFuncImp::callAsFunction): Pass progNode as body for ContextImp in

11412

eval.

11413

* kjs/function.h: Move body field from DeclaredFunctionImp to

11414

FunctionImp.

11415

* kjs/grammar.y: Change DBG; statements no longer have a sourceid.

11416

* kjs/internal.cpp:

11417

(KJS::ContextImp::ContextImp): Initialize new m_currentBody, m_iterationDepth

11418

and m_switchDepth data members. New FunctionBodyNode* parameter - the

11419

function body provides source URL and SourceId.

11420

(KJS::InterpreterImp::mark): Use exception() function, not _exception directly.

11421

(KJS::InterpreterImp::evaluate): Pass progNode to ContextImp constructor

11422

to use as the body.

11423

* kjs/internal.h:

11424

(KJS::LabelStack::LabelStack): Remove iteration depth and switch depth;

11425

statement label stacks don't need these and it bloats their size. Put them

11426

in the ContextImp instead.

11427

* kjs/interpreter.cpp:

11428

(KJS::ExecState::lexicalInterpreter): Renamed _context to m_context.

11429

* kjs/interpreter.h:

11430

(KJS::ExecState::dynamicInterpreter): Renamed _context to m_context.

11431

(KJS::ExecState::context): ditto

11432

(KJS::ExecState::setException): Renamed _exception to m_exception

11433

(KJS::ExecState::clearException): ditto

11434

(KJS::ExecState::exception): ditto

11435

(KJS::ExecState::hadException): ditto

11436

(KJS::ExecState::ExecState): ditto both above renames

11437

* kjs/nodes.cpp:

11438

(Node::Node): Removed initialization of line, source URL and refcount. Add to

11439

local newNodes set instead of involving parser.

11440

(Node::ref): Instead of managing refcount directly, story refcount over 1 in a

11441

HashCountedSet, and keep a separate HashSet of "floating" nodes with refcount

11442

11443

(Node::deref): ditto

11444

(Node::refcount): ditto

11445

(Node::clearNewNodes): Destroy anything left in the new nodes set.

11446

(currentSourceId): Inline helper to get sourceId from function body via context.

11447

(currentSourceURL): ditto for sourceURL.

11448

(Node::createErrorCompletion): use new helper

11449

(Node::throwError): ditto

11450

(Node::setExceptionDetailsIfNeeded): ditto

11451

(StatementNode::StatementNode): remove initialization of l0 and sid, rename

11452

l1 to m_lastLine.

11453

(StatementNode::setLoc): Set own m_lastLine and Node's m_line.

11454

(StatementNode::hitStatement): Get sid, first line, last line in the proper new ways.

11455

(StatListNode::StatListNode): updated for setLoc changes

11456

(BlockNode::BlockNode): ditto

11457

(DoWhileNode::execute): excpect iteraton counts on ContextImp, not LabelStack

11458

(WhileNode::execute): ditto

11459

(ForNode::execute): ditto

11460

(ForInNode::execute): ditto

11461

(ContinueNode::execute): excpect inIteration on ContextImp, not LabelStack

11462

(BreakNode::execute): excpect inIteration and inSwitch on ContextImp, not LabelStack

11463

(SwitchNode::execute): expect switch counts on ContextImp, not LabelStack

11464

(FunctionBodyNode::FunctionBodyNode): update for new setLoc

11465

(FunctionBodyNode::processFuncDecl): reindent

11466

(SourceElementsNode::SourceElementsNode): update for new setLoc

11467

* kjs/nodes.h:

11468

(KJS::Node::lineNo): Renamed _line to m_line

11469

(KJS::StatementNode::firstLine): Use lineNo()

11470

(KJS::StatementNode::lastLine): Renamed l1 to m_lastLine

11471

(KJS::FunctionBodyNode::sourceId): added

11472

(KJS::FunctionBodyNode::sourceURL): added

11473

* kjs/testkjs.cpp:

11474

11475

2006-03-14 Geoffrey Garen <ggaren@apple.com>

11476

11477

- Fixed <rdar://problem/4478239> string sort puts "closed" before

11478

"close"

11479

11480

Reviewed by Eric.

11481

11482

* kjs/ustring.cpp:

11483

(KJS::compare): Inverted a < in order to treat the longer string as >

11484

the shorter string.

11485

11486

2006-03-12 Alexey Proskuryakov <ap@nypop.com>

11487

11488

Reviewed by Maciej.

11489

11490

- fix http://bugs.webkit.org/show_bug.cgi?id=7708

11491

REGRESSION: Flash callback to JavaScript function not working.

11492

11493

Test: plugins/invoke.html

11494

11495

* bindings/c/c_utility.cpp:

11496

(KJS::Bindings::convertUTF8ToUTF16): Return a correct string length.

11497

11498

2006-03-08 Eric Seidel <eseidel@apple.com>

11499

11500

Reviewed by darin.

11501

11502

Partially fix JS on win32 by fixing hash table generation.

11503

11504

* kjs/create_hash_table: limit << results to 32 bits.

11505

* kjs/testkjs.cpp:

11506

(TestFunctionImp::callAsFunction):

11507

11508

2006-03-07 Darin Adler <darin@apple.com>

11509

11510

* kxmlcore/Vector.h: Quick fix to try to get Windows compiling again.

11511

11512

2006-03-07 Darin Adler <darin@apple.com>

11513

11514

Reviewed by Anders.

11515

11516

- fix http://bugs.webkit.org/show_bug.cgi?id=7655

11517

unwanted output while running layout tests

11518

11519

* kjs/lexer.cpp: (Lexer::lex): Turn off the "yylex: ERROR" message.

11520

* kjs/regexp.cpp: (KJS::RegExp::RegExp): Remove the code to log errors from PCRE

11521

to standard output. I think we should arrange for the error text to be in JavaScript

11522

exceptions instead at some point.

11523

* kxmlcore/Vector.h: Add a check for overflow so that we'll abort if we pass a

11524

too-large size rather than allocating a buffer smaller than requested.

11525

11526

2006-03-06 David Carson <dacarson@gmail.com>

11527

11528

Reviewed by Darin, landed by ap.

11529

11530

- Fixed http://bugs.webkit.org/show_bug.cgi?id=7582

11531

c_utility.cpp contains CFString OS X platform-dependent code; should use ICU

11532

11533

Tested with test case from:

11534

http://bugs.webkit.org/show_bug.cgi?id=5163

11535

11536

* bindings/c_utility.cpp

11537

(convertUTF8ToUTF16): Changed to using Unicode converter from ICU, and manual Latin-1 conversion.

11538

* icu/unicode/ucnv.h: Copied from WebCore.

11539

* icu/unicode/ucnv_err.h: Ditto.

11540

* icu/unicode/uenum.h: Ditto.

11541

11542

2006-03-05 Darin Adler <darin@apple.com>

11543

11544

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: Updated.

11545

11546

2006-03-06 Mitz Pettel <opendarwin.org@mitzpettel.com>

11547

11548

Fix suggested by Alexey Proskuryakov <ap@nypop.com>, reviewed by Maciej and Hyatt.

11549

11550

- fix http://bugs.webkit.org/show_bug.cgi?id=7601

11551

REGRESSION (r13089): Reproducible crash dereferencing a deallocated element on google image search

11552

11553

* kxmlcore/Platform.h: Corrected the define to enable USE(MULTIPLE_THREADS) on Mac OS X.

11554

11555

2006-03-05 Darin Adler <darin@apple.com>

11556

11557

Reviewed by Maciej.

11558

11559

- http://bugs.webkit.org/show_bug.cgi?id=7616

11560

get all references to KJS::Node out of internal.h

11561

11562

* JavaScriptCore.xcodeproj/project.pbxproj: Updated for file changes.

11563

11564

* kjs/Parser.cpp: Added.

11565

* kjs/Parser.h: Added.

11566

11567

* kjs/internal.cpp: Removed the Parser class.

11568

* kjs/internal.h: Ditto. Also removed unnecessary declarations of classes

11569

not used in this header.

11570

11571

* kjs/nodes.h: Added an include of "Parser.h".

11572

* kjs/function.h: Added a declaration of FunctionBodyNode.

11573

11574

2006-03-05 Geoffrey Garen <ggaren@apple.com>

11575

11576

Reviewed by Maciej.

11577

11578

- JSC support for the fix for <rdar://problem/4467143> JavaScript

11579

enumeration of HTML element properties skips DOM node properties

11580

11581

* kjs/lookup.h:

11582

(1) Added the KJS_DEFINE_PROTOTYPE_WITH_PROTOTYPE macro. The

11583

class definiton macro needs to know about the prototype's prototype so

11584

that the class constructor properly sets it.

11585

(2) Removed the KJS_IMPLEMENT_PROTOTYPE_WITH_PARENT macro. The class

11586

implementation macro does not need to know about the prototype's

11587

prototype, since getOwnPropertySlot should only look in the current

11588

object's property map, and not its prototype's.

11589

11590

2006-03-05 Andrew Wellington <proton@wiretapped.net>

11591

11592

Reviewed by Eric, landed by ap.

11593

11594

- Remove unused breakpoint bool from StatementNodes. No test provided as

11595

there is no functionality change.

11596

11597

* kjs/nodes.cpp:

11598

(StatementNode::StatementNode):

11599

* kjs/nodes.h:

11600

11601

2006-03-03 Geoffrey Garen <ggaren@apple.com>

11602

11603

Reviewed by Darin.

11604

11605

- Fixed <rdar://problem/4465598> REGRESSION (TOT): Crash occurs at

11606

http://maps.google.com/?output=html ( KJS::Identifier::add(KJS::UString::Rep*)

11607

11608

This regression was caused by my fix for 4448098. I failed to account for the

11609

deleted entry sentinel in the mehtod that saves the contents of a property map to

11610

the back/forward cache.

11611

11612

Manual test in WebCore/manual-tests/property-map-save-crash.html

11613

11614

* kjs/property_map.cpp:

11615

(KJS::deletedSentinel): Use 1 instead of -1 to facilitate an easy bit mask

11616

(KJS::isValid): New function: checks if a key is null or the deleted sentinel

11617

(KJS::PropertyMap::~PropertyMap): Fixed up the branch logic here for readability

11618

and a slight performance win

11619

(KJS::PropertyMap::clear):

11620

(KJS::PropertyMap::rehash):

11621

(KJS::PropertyMap::addSparseArrayPropertiesToReferenceList):

11622

(KJS::PropertyMap::save): Check keys with isValid()

11623

11624

2006-03-02 Maciej Stachowiak <mjs@apple.com>

11625

11626

- now fix mac build again

11627

11628

* kjs/identifier.cpp:

11629

11630

2006-03-02 Maciej Stachowiak <mjs@apple.com>

11631

11632

Rubber stamped by Anders and Eric.

11633

11634

- add fpconst.cpp to win32 build, it is now needed

11635

11636

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

11637

* kjs/fpconst.cpp:

11638

11639

2006-03-02 Maciej Stachowiak <mjs@apple.com>

11640

11641

Reviewed by Eric.

11642

11643

- fix windows build, broken by my last patch

11644

11645

* kjs/JSImmediate.cpp:

11646

* kjs/identifier.cpp:

11647

* kxmlcore/FastMalloc.cpp:

11648

* kxmlcore/Platform.h:

11649

11650

2006-03-01 Maciej Stachowiak <mjs@apple.com>

11651

11652

Reviewed by Darin.

11653

11654

- Set up new prototype macros and avoid using #if without defined() in JSC

11655

11656

Added new PLATFORM macros and related, to make sure #if's all check if relevant macros

11657

are defined, and to separate core OS-level dependencies from operating environment

11658

dependencies so you can, e.g., build KDE on Mac or Windows.

11659

11660

* kxmlcore/Platform.h: Added.

11661

11662

- deploy them everywhere in JavaScriptCore

11663

11664

* JavaScriptCore.xcodeproj/project.pbxproj:

11665

* bindings/jni/jni_utility.cpp:

11666

(KJS::Bindings::convertValueToJValue):

11667

* bindings/objc/WebScriptObject.mm:

11668

* bindings/objc/objc_instance.mm:

11669

(ObjcInstance::end):

11670

* bindings/softlinking.h:

11671

* bindings/testbindings.mm:

11672

(main):

11673

* kjs/JSLock.cpp:

11674

* kjs/collector.cpp:

11675

(KJS::Collector::markCurrentThreadConservatively):

11676

(KJS::Collector::markOtherThreadConservatively):

11677

(KJS::Collector::markStackObjectsConservatively):

11678

* kjs/config.h:

11679

* kjs/date_object.cpp:

11680

(gmtoffset):

11681

(KJS::formatTime):

11682

(KJS::DateProtoFunc::callAsFunction):

11683

(KJS::DateObjectImp::construct):

11684

(KJS::makeTime):

11685

* kjs/dtoa.cpp:

11686

* kjs/fpconst.cpp:

11687

(KJS::sizeof):

11688

(KJS::):

11689

* kjs/grammar.y:

11690

* kjs/identifier.cpp:

11691

* kjs/internal.cpp:

11692

* kjs/interpreter.cpp:

11693

(KJS::Interpreter::evaluate):

11694

(KJS::Interpreter::createLanguageInstanceForValue):

11695

* kjs/interpreter.h:

11696

* kjs/lookup.cpp:

11697

* kjs/lookup.h:

11698

* kjs/math_object.cpp:

11699

* kjs/object.cpp:

11700

* kjs/object.h:

11701

* kjs/operations.cpp:

11702

(KJS::isNaN):

11703

(KJS::isInf):

11704

(KJS::isPosInf):

11705

(KJS::isNegInf):

11706

* kjs/operations.h:

11707

* kjs/regexp.cpp:

11708

(KJS::RegExp::RegExp):

11709

(KJS::RegExp::~RegExp):

11710

(KJS::RegExp::match):

11711

* kjs/regexp.h:

11712

* kjs/testkjs.cpp:

11713

(StopWatch::start):

11714

(StopWatch::stop):

11715

(StopWatch::getElapsedMS):

11716

* kjs/ustring.cpp:

11717

* kjs/ustring.h:

11718

* kxmlcore/AlwaysInline.h:

11719

* kxmlcore/Assertions.cpp:

11720

* kxmlcore/Assertions.h:

11721

* kxmlcore/FastMalloc.cpp:

11722

(KXMLCore::):

11723

* kxmlcore/FastMalloc.h:

11724

* kxmlcore/FastMallocInternal.h:

11725

* kxmlcore/HashTable.h:

11726

* kxmlcore/TCPageMap.h:

11727

* kxmlcore/TCSpinLock.h:

11728

(TCMalloc_SpinLock::Lock):

11729

(TCMalloc_SpinLock::Unlock):

11730

(TCMalloc_SlowLock):

11731

* kxmlcore/TCSystemAlloc.cpp:

11732

(TCMalloc_SystemAlloc):

11733

* os-win32/stdint.h:

11734

11735

2006-02-28 Geoffrey Garen <ggaren@apple.com>

11736

11737

Reviewed by Darin.

11738

11739

- Fixed <rdar://problem/4448098> Switch PropertyMap deleted entry

11740

placeholder to -1 from UString::Rep::null

11741

11742

This turned out to be only a small speedup (.12%). That's within the

11743

margin of error for super accurate JS iBench, but Shark confirms the

11744

same, so I think it's worth landing.

11745

11746

FYI, I also confirmed that the single entry optimization in

11747

PropertyMap is a 3.2% speedup.

11748

11749

* kjs/property_map.cpp:

11750

(KJS::PropertyMap::~PropertyMap):

11751

(KJS::PropertyMap::clear):

11752

(KJS::PropertyMap::put):

11753

(KJS::PropertyMap::insert):

11754

(KJS::PropertyMap::rehash):

11755

(KJS::PropertyMap::remove):

11756

(KJS::PropertyMap::addSparseArrayPropertiesToReferenceList):

11757

(KJS::PropertyMap::checkConsistency):

11758

* kjs/property_map.h:

11759

(KJS::PropertyMap::deletedSentinel):

11760

11761

2006-02-27 Eric Seidel <eseidel@apple.com>

11762

11763

Rubber-stamped by darin.

11764

11765

Remove fpconst.cpp, unused on win32 and the cause of linker warnings.

11766

11767

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

11768

11769

2006-02-27 Eric Seidel <eseidel@apple.com>

11770

11771

Reviewed by mjs.

11772

11773

Fix Assertions.cpp to compile on win32.

11774

11775

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

11776

* kxmlcore/Assertions.cpp:

11777

11778

2006-02-27 Eric Seidel <eseidel@apple.com>

11779

11780

Reviewed by mjs.

11781

11782

Made Assertions.cpp platform independent.

11783

Moved mac-specific logging logic up into WebCore.

11784

http://bugs.webkit.org/show_bug.cgi?id=7503

11785

11786

* JavaScriptCore.xcodeproj/project.pbxproj:

11787

* kxmlcore/Assertions.cpp: Added.

11788

* kxmlcore/Assertions.h:

11789

* kxmlcore/Assertions.mm: Removed.

11790

11791

2006-02-27 Darin Adler <darin@apple.com>

11792

11793

- fixed Mac Debug build, there was an unused parameter

11794

11795

* kxmlcore/FastMalloc.cpp: (KXMLCore::fastMallocRegisterThread):

11796

Remove parameter name.

11797

11798

* kjs/debugger.h: Fixed comment.

11799

11800

2006-02-27 Eric Seidel <eseidel@apple.com>

11801

11802

Reviewed by darin.

11803

11804

* kxmlcore/Vector.h:

11805

(KXMLCore::deleteAllValues): fix unused variable warning

11806

11807

2006-02-21 Maciej Stachowiak <mjs@apple.com>

11808

11809

Reviewed by Darin.

11810

11811

- Turn off -Wno-unused-param for JavaScriptCore and get rid of unused params

11812

http://bugs.webkit.org/show_bug.cgi?id=7384

11813

11814

* JavaScriptCore.xcodeproj/project.pbxproj:

11815

* bindings/NP_jsobject.cpp:

11816

(jsAllocate):

11817

(_NPN_InvokeDefault):

11818

(_NPN_Evaluate):

11819

(_NPN_GetProperty):

11820

(_NPN_SetProperty):

11821

(_NPN_RemoveProperty):

11822

(_NPN_HasProperty):

11823

(_NPN_HasMethod):

11824

* bindings/c/c_class.h:

11825

(KJS::Bindings::CClass::constructorAt):

11826

* bindings/c/c_utility.cpp:

11827

(KJS::Bindings::convertNPVariantToValue):

11828

* bindings/jni/jni_class.cpp:

11829

(JavaClass::methodsNamed):

11830

(JavaClass::fieldNamed):

11831

* bindings/jni/jni_instance.cpp:

11832

(JavaInstance::invokeDefaultMethod):

11833

* bindings/jni/jni_jsobject.cpp:

11834

* bindings/jni/jni_objc.mm:

11835

(-[NSObject KJS::Bindings::]):

11836

* bindings/objc/WebScriptObject.mm:

11837

(+[WebUndefined allocWithZone:]):

11838

(-[WebUndefined initWithCoder:]):

11839

(-[WebUndefined encodeWithCoder:]):

11840

(-[WebUndefined copyWithZone:]):

11841

* bindings/objc/objc_class.h:

11842

(KJS::Bindings::ObjcClass::constructorAt):

11843

* bindings/objc/objc_class.mm:

11844

(KJS::Bindings::ObjcClass::methodsNamed):

11845

(KJS::Bindings::ObjcClass::fallbackObject):

11846

* bindings/objc/objc_instance.mm:

11847

(ObjcInstance::getValueOfUndefinedField):

11848

* bindings/objc/objc_runtime.mm:

11849

(ObjcFallbackObjectImp::getOwnPropertySlot):

11850

(ObjcFallbackObjectImp::put):

11851

(ObjcFallbackObjectImp::canPut):

11852

(ObjcFallbackObjectImp::deleteProperty):

11853

(ObjcFallbackObjectImp::toBoolean):

11854

* bindings/runtime.cpp:

11855

(KJS::Bindings::Instance::createLanguageInstanceForValue):

11856

* bindings/runtime.h:

11857

(KJS::Bindings::Instance::getValueOfUndefinedField):

11858

(KJS::Bindings::Instance::setValueOfUndefinedField):

11859

* bindings/runtime_array.cpp:

11860

(RuntimeArray::lengthGetter):

11861

(RuntimeArray::indexGetter):

11862

(RuntimeArray::put):

11863

(RuntimeArray::deleteProperty):

11864

* bindings/runtime_method.cpp:

11865

(RuntimeMethod::lengthGetter):

11866

(RuntimeMethod::execute):

11867

* bindings/runtime_object.cpp:

11868

(RuntimeObjectImp::fallbackObjectGetter):

11869

(RuntimeObjectImp::fieldGetter):

11870

(RuntimeObjectImp::methodGetter):

11871

(RuntimeObjectImp::put):

11872

(RuntimeObjectImp::canPut):

11873

(RuntimeObjectImp::deleteProperty):

11874

(RuntimeObjectImp::defaultValue):

11875

(RuntimeObjectImp::callAsFunction):

11876

* bindings/runtime_root.cpp:

11877

(performJavaScriptAccess):

11878

* kjs/array_object.cpp:

11879

(ArrayInstance::lengthGetter):

11880

(ArrayInstance::getOwnPropertySlot):

11881

(ArrayPrototype::ArrayPrototype):

11882

(ArrayPrototype::getOwnPropertySlot):

11883

* kjs/bool_object.cpp:

11884

(BooleanObjectImp::BooleanObjectImp):

11885

* kjs/date_object.cpp:

11886

(KJS::DateObjectFuncImp::DateObjectFuncImp):

11887

(KJS::DateObjectFuncImp::callAsFunction):

11888

* kjs/error_object.cpp:

11889

(ErrorObjectImp::ErrorObjectImp):

11890

(NativeErrorPrototype::NativeErrorPrototype):

11891

(NativeErrorImp::NativeErrorImp):

11892

* kjs/function.cpp:

11893

(KJS::FunctionImp::argumentsGetter):

11894

(KJS::FunctionImp::lengthGetter):

11895

(KJS::Arguments::mappedIndexGetter):

11896

(KJS::ActivationImp::argumentsGetter):

11897

(KJS::ActivationImp::put):

11898

* kjs/function_object.cpp:

11899

(FunctionObjectImp::FunctionObjectImp):

11900

* kjs/internal.cpp:

11901

(KJS::GetterSetterImp::toPrimitive):

11902

(KJS::GetterSetterImp::toBoolean):

11903

* kjs/interpreter.cpp:

11904

(KJS::Interpreter::evaluate):

11905

* kjs/interpreter.h:

11906

(KJS::Interpreter::isGlobalObject):

11907

(KJS::Interpreter::interpreterForGlobalObject):

11908

(KJS::Interpreter::isSafeScript):

11909

* kjs/lexer.cpp:

11910

(Lexer::makeIdentifier):

11911

(Lexer::makeUString):

11912

* kjs/lookup.h:

11913

(KJS::staticFunctionGetter):

11914

(KJS::staticValueGetter):

11915

* kjs/nodes.cpp:

11916

(StatementNode::processFuncDecl):

11917

(PropertyNode::evaluate):

11918

(PropertyNameNode::evaluate):

11919

* kjs/number_object.cpp:

11920

(NumberObjectImp::NumberObjectImp):

11921

(NumberObjectImp::getOwnPropertySlot):

11922

* kjs/object.cpp:

11923

(KJS::JSObject::defineGetter):

11924

(KJS::JSObject::defineSetter):

11925

(KJS::JSObject::hasInstance):

11926

(KJS::JSObject::propertyIsEnumerable):

11927

* kjs/object_object.cpp:

11928

(ObjectObjectImp::ObjectObjectImp):

11929

* kjs/property_slot.cpp:

11930

(KJS::PropertySlot::undefinedGetter):

11931

(KJS::PropertySlot::functionGetter):

11932

* kjs/reference.cpp:

11933

(KJS::Reference::getPropertyName):

11934

* kjs/reference_list.cpp:

11935

(ReferenceListIterator::operator++):

11936

* kjs/regexp_object.cpp:

11937

(RegExpObjectImp::RegExpObjectImp):

11938

(RegExpObjectImp::getValueProperty):

11939

(RegExpObjectImp::putValueProperty):

11940

* kjs/string_object.cpp:

11941

(StringInstance::lengthGetter):

11942

(StringInstance::indexGetter):

11943

(StringPrototype::StringPrototype):

11944

* kxmlcore/Assertions.mm:

11945

* kxmlcore/FastMalloc.cpp:

11946

(KXMLCore::TCMalloc_PageHeap::CheckList):

11947

* kxmlcore/HashTable.h:

11948

(KXMLCore::HashTableConstIterator::checkValidity):

11949

(KXMLCore::IdentityHashTranslator::translate):

11950

* pcre/pcre_get.c:

11951

(pcre_get_stringnumber):

11952

11953

2006-02-23 Darin Adler <darin@apple.com>

11954

11955

- try to fix buildbot failure

11956

11957

* bindings/c/c_utility.cpp: Touch this file, which seems to not have been

11958

recompiled after additional inlining was introduced (Xcode bug?).

11959

11960

2006-02-23 Geoffrey Garen <ggaren@apple.com>

11961

11962

Reviewed by Darin, Maciej.

11963

11964

- Inline some functions suggested by Shark. 2.9% speedup on super

11965

accurate JS iBench.

11966

11967

http://bugs.webkit.org/show_bug.cgi?id=7411

11968

<rdar://problem/4448116>

11969

11970

* kjs/nodes.h:

11971

(KJS::ArgumentsNode::evaluateList):

11972

* kjs/object.cpp:

11973

* kjs/object.h:

11974

(KJS::ScopeChain::release):

11975

(KJS::JSObject::toPrimitive):

11976

* kjs/scope_chain.cpp:

11977

* kjs/ustring.cpp:

11978

* kjs/ustring.h:

11979

(KJS::UString::toArrayIndex):

11980

* kjs/value.cpp:

11981

* kjs/value.h:

11982

(KJS::JSValue::toObject):

11983

* kxmlcore/FastMalloc.cpp:

11984

(KXMLCore::TCMalloc_ThreadCache_FreeList::Push):

11985

(KXMLCore::TCMalloc_ThreadCache_FreeList::Pop):

11986

11987

2006-02-21 Eric Seidel <eseidel@apple.com>

11988

11989

Added *.user to ignore list.

11990

11991

2006-02-21 Eric Seidel <eseidel@apple.com>

11992

11993

Reviewed by ggaren.

11994

11995

Add grammarWrapper.cpp to work around visual studio bug plaguing buildbot.

11996

11997

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

11998

* JavaScriptCore.vcproj/JavaScriptCore/grammarWrapper.cpp: Added.

11999

12000

2006-02-21 Eric Seidel <eseidel@apple.com>

12001

12002

Reviewed by ggaren.

12003

12004

* kjs/testkjs.cpp: #if out timeval code on win32

12005

12006

2006-02-21 Michael Emmel <mike.emmel@gmail.com>

12007

12008

Reviewed by Darin.

12009

12010

- fix http://bugs.webkit.org/show_bug.cgi?id=7397

12011

TCPageMap.h would not compile for me because string.h was missing

12012

12013

* kxmlcore/TCPageMap.h: Added <string.h> include.

12014

12015

2006-02-21 Darin Adler <darin@apple.com>

12016

12017

Reviewed by John Sullivan.

12018

12019

- http://bugs.webkit.org/show_bug.cgi?id=7404

12020

remove a bunch of extra implementsCall overrides

12021

12022

* JavaScriptCore.xcodeproj/project.pbxproj: Sorted files.

12023

12024

* kjs/internal.h: Made InternalFunctionImp::callAsFunction pure virtual so that

12025

we'll get a compile error if some derived class neglects to implement it.

12026

12027

* kjs/function.cpp: (KJS::FunctionImp::FunctionImp): Remove unneeded initialization

12028

of param, which is an OwnPtr so it gets initialized by default.

12029

12030

* bindings/runtime_method.cpp:

12031

* bindings/runtime_method.h:

12032

* kjs/array_object.cpp:

12033

* kjs/array_object.h:

12034

* kjs/bool_object.cpp:

12035

* kjs/bool_object.h:

12036

* kjs/date_object.cpp:

12037

* kjs/date_object.h:

12038

* kjs/error_object.cpp:

12039

* kjs/error_object.h:

12040

* kjs/function.cpp:

12041

* kjs/function.h:

12042

* kjs/function_object.cpp:

12043

* kjs/function_object.h:

12044

* kjs/math_object.cpp:

12045

* kjs/math_object.h:

12046

* kjs/number_object.cpp:

12047

* kjs/number_object.h:

12048

* kjs/object_object.cpp:

12049

* kjs/object_object.h:

12050

* kjs/regexp_object.cpp:

12051

* kjs/regexp_object.h:

12052

* kjs/string_object.cpp:

12053

* kjs/string_object.h:

12054

Removed many rendundant implementations of implementsCall from subclasses of

12055

InternalFunctionImp.

12056

12057

2006-02-21 Darin Adler <darin@apple.com>

12058

12059

- fixed build

12060

12061

* kjs/internal.cpp: (KJS::InternalFunctionImp::implementsCall):

12062

Oops, fixed name.

12063

12064

2006-02-21 Darin Adler <darin@apple.com>

12065

12066

Change suggested by Mitz.

12067

12068

- http://bugs.webkit.org/show_bug.cgi?id=7402

12069

REGRESSION: Methods do not execute

12070

12071

* kjs/internal.h: Add implementsHasCall to InternalFunctionImp.

12072

* kjs/internal.cpp: (KJS::InternalFunctionImp::implementsHasCall):

12073

Return true. All the classes derived from InternalFunctionImp need

12074

to return true from this -- later we can remove all the extra

12075

implementations too.

12076

12077

2006-02-21 Maciej Stachowiak <mjs@apple.com>

12078

12079

- fix build breakage caused by last-minute change to my patch

12080

12081

* kjs/lookup.h:

12082

12083

2006-02-20 Maciej Stachowiak <mjs@apple.com>

12084

12085

Reviewed by Geoff and Darin.

12086

12087

Patch from Maks Orlovich, based on work by David Faure, hand-applied and

12088

significantly reworked by me.

12089

12090

- Patch: give internal function names (KJS merge)

12091

http://bugs.webkit.org/show_bug.cgi?id=6279

12092

12093

* tests/mozilla/expected.html: Updated for newly fixed test.

12094

12095

* kjs/array_object.cpp:

12096

(ArrayProtoFunc::ArrayProtoFunc):

12097

* kjs/array_object.h:

12098

* kjs/bool_object.cpp:

12099

(BooleanPrototype::BooleanPrototype):

12100

(BooleanProtoFunc::BooleanProtoFunc):

12101

* kjs/bool_object.h:

12102

* kjs/date_object.cpp:

12103

(KJS::DateProtoFunc::DateProtoFunc):

12104

(KJS::DateObjectImp::DateObjectImp):

12105

(KJS::DateObjectFuncImp::DateObjectFuncImp):

12106

* kjs/error_object.cpp:

12107

(ErrorPrototype::ErrorPrototype):

12108

(ErrorProtoFunc::ErrorProtoFunc):

12109

* kjs/error_object.h:

12110

* kjs/function.cpp:

12111

(KJS::FunctionImp::FunctionImp):

12112

(KJS::GlobalFuncImp::GlobalFuncImp):

12113

* kjs/function.h:

12114

* kjs/function_object.cpp:

12115

(FunctionPrototype::FunctionPrototype):

12116

(FunctionProtoFunc::FunctionProtoFunc):

12117

(FunctionProtoFunc::callAsFunction):

12118

* kjs/function_object.h:

12119

* kjs/internal.cpp:

12120

(KJS::InterpreterImp::initGlobalObject):

12121

(KJS::InternalFunctionImp::InternalFunctionImp):

12122

* kjs/internal.h:

12123

(KJS::InternalFunctionImp::functionName):

12124

* kjs/lookup.h:

12125

(KJS::staticFunctionGetter):

12126

(KJS::HashEntryFunction::HashEntryFunction):

12127

(KJS::HashEntryFunction::implementsCall):

12128

(KJS::HashEntryFunction::toBoolean):

12129

(KJS::HashEntryFunction::implementsHasInstance):

12130

(KJS::HashEntryFunction::hasInstance):

12131

* kjs/math_object.cpp:

12132

(MathFuncImp::MathFuncImp):

12133

* kjs/math_object.h:

12134

* kjs/number_object.cpp:

12135

(NumberPrototype::NumberPrototype):

12136

(NumberProtoFunc::NumberProtoFunc):

12137

* kjs/number_object.h:

12138

* kjs/object.cpp:

12139

(KJS::JSObject::putDirectFunction):

12140

(KJS::Error::create):

12141

* kjs/object.h:

12142

* kjs/object_object.cpp:

12143

(ObjectPrototype::ObjectPrototype):

12144

(ObjectProtoFunc::ObjectProtoFunc):

12145

* kjs/object_object.h:

12146

* kjs/regexp_object.cpp:

12147

(RegExpPrototype::RegExpPrototype):

12148

(RegExpProtoFunc::RegExpProtoFunc):

12149

* kjs/regexp_object.h:

12150

* kjs/string_object.cpp:

12151

(StringProtoFunc::StringProtoFunc):

12152

(StringObjectImp::StringObjectImp):

12153

(StringObjectFuncImp::StringObjectFuncImp):

12154

* kjs/string_object.h:

12155

12156

2006-02-20 Geoffrey Garen <ggaren@apple.com>

12157

12158

Reviewed by Darin, with help from Eric, Maciej.

12159

12160

- More changes to support super-accurate JS iBench. Doesn't work on

12161

Windows. (Doesn't break Windows, either.) I've filed

12162

[http://bugs.webkit.org/show_bug.cgi?id=7399] about that.

12163

12164

* kjs/interpreter.cpp:

12165

(KJS::Interpreter::evaluate): Print line numbers with exception output

12166

* kjs/testkjs.cpp: Changed " *" to "* " because Eric says that's the

12167

way we roll with .cpp files.

12168

(StopWatch::StopWatch): New class. Provides microsecond-accurate

12169

timings.

12170

(StopWatch::~StopWatch):

12171

(StopWatch::start):

12172

(StopWatch::stop):

12173

(StopWatch::getElapsedMS):

12174

(TestFunctionImp::callAsFunction): Added missing return statement.

12175

Fixed up "run" to use refactored helper functions. Removed bogus

12176

return statement from "quit" case. Made "print" output to stdout

12177

instead of stderr because that makes more sense, and PERL handles

12178

stdout better.

12179

(main): Factored out KXMLCore unit tests. Removed custom exception

12180

printing code because the interpreter prints exceptions for you. Added

12181

a "delete" call for the GlobalImp we allocate.

12182

(testIsInteger): New function, result of refacotring.

12183

(createStringWithContentsOfFile): New function, result of refactoring.

12184

Renamed "code" to "buffer" to match factored-out-ness.

12185

12186

2006-02-20 Eric Seidel <eseidel@apple.com>

12187

12188

Reviewed by hyatt.

12189

12190

Fix "Copy ICU DLLs..." phase.

12191

12192

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

12193

* JavaScriptCore.vcproj/JavaScriptCore/build-generated-files.sh:

12194

12195

2006-02-19 Darin Adler <darin@apple.com>

12196

12197

- renamed ERROR to LOG_ERROR to fix build

12198

presumably Maciej had this change and forgot to land it

12199

12200

* kjs/collector.cpp: Removed now-unneeded #undef ERROR.

12201

* kxmlcore/Assertions.h: Renamed ERROR to LOG_ERROR.

12202

* kxmlcore/FastMalloc.cpp: Changed MESSAGE macro to use LOG_ERROR.

12203

12204

2006-02-18 Mitz Pettel <opendarwin.org@mitzpettel.com>

12205

12206

Test: fast/js/toString-exception.html

12207

12208

Reviewed by Maciej.

12209

12210

- fix http://bugs.webkit.org/show_bug.cgi?id=7343

12211

REGRESSION: fast/js/toString-overrides.html fails when run multiple times

12212

12213

* kjs/array_object.cpp:

12214

(ArrayProtoFunc::callAsFunction): Remove the object from the visited elements set before

12215

returning an error.

12216

12217

2006-02-18 Darin Adler <darin@apple.com>

12218

12219

Reviewed by Maciej.

12220

12221

- http://bugs.webkit.org/show_bug.cgi?id=7345

12222

add insert and remove to KXMLCore::Vector

12223

12224

* kxmlcore/Vector.h: Added "moveOverlapping", which is used in both

12225

insert and remove to slide elements within the vector. Also added

12226

"insert" and "remove" functions.

12227

12228

2006-02-16 Geoffrey Garen <ggaren@apple.com>

12229

12230

Reviewed by John.

12231

12232

- Fixed <rdar://problem/4448534> TOT REGRESSION: crash in KJS::

12233

Bindings::Instance::deref when leaving page @ gigaom.com

12234

12235

* bindings/c/c_instance.cpp:

12236

(KJS::Bindings::CInstance::~CInstance): Since we cache the class object

12237

globally, we shouldn't delete it, so don't.

12238

12239

2006-02-16 Timothy Hatcher <timothy@apple.com>

12240

12241

Added -Wno-deprecated-declarations to all the ObjC binding files to prevent deprecation

12242

warnings. Using <rdar://problem/4448350> to track this.

12243

12244

* JavaScriptCore.xcodeproj/project.pbxproj:

12245

* bindings/objc/objc_jsobject.h: Removed empty file.

12246

* bindings/objc/objc_jsobject.mm: Removed empty file.

12247

12248

2006-02-16 Tim Omernick <timo@apple.com>

12249

12250

Reviewed by Geoff.

12251

12252

<rdar://problem/4428609> Flash Player 8.0.22 can crash Safari (and WebKit apps) with

12253

javascript disabled (7015)

12254

12255

* bindings/NP_jsobject.cpp:

12256

(_NPN_CreateNoScriptObject):

12257

Returns an NPObject which is not bound to a JavaScript object. This kind of NPObject

12258

can be given to a plugin as the "window script object" when JavaScript is disabled.

12259

The object has a custom NPClass, NPNoScriptObjectClass, which has no defined methods.

12260

Because of this, none of the NPN_* functions called by the plugin on this "no script

12261

object" will cause entry into JavaScript code.

12262

(_NPN_InvokeDefault):

12263

Make sure the NPVariant is filled before returning from this function. This never

12264

mattered before because we never reached this case, having only created NPObjects of

12265

the class NPScriptObjectClass.

12266

(_NPN_Invoke):

12267

ditto

12268

(_NPN_Evaluate):

12269

ditto

12270

(_NPN_GetProperty):

12271

ditto

12272

12273

* bindings/NP_jsobject.h:

12274

Declared _NPN_CreateNoScriptObject().

12275

12276

2006-02-16 Darin Adler <darin@apple.com>

12277

12278

Reviewed by me, change by Peter Kuemmel.

12279

12280

* kjs/operations.cpp: (KJS::isNegInf): Fix Windows code, which was

12281

checking for positive infinity (rolling in fix from KDE side).

12282

12283

2006-02-15 Geoffrey Garen <ggaren@apple.com>

12284

12285

Reviewed by Maciej, Eric.

12286

12287

- JavaScriptCore half of fix for <rdar://problem/4176077> CrashTracer: 6569

12288

crashes in DashboardClient at com.apple.JavaScriptCore:

12289

KJS::Bindings::ObjcFallbackObjectImp::type()

12290

12291

WebCore and JavaScriptCore weren't sharing Instance objects very

12292

nicely. I made them use RefPtrs, and sent them to bed without dessert.

12293

12294

* bindings/jni/jni_instance.cpp: Made _instance a RefPtr

12295

(JavaInstance::~JavaInstance):

12296

(JObjectWrapper::JObjectWrapper):

12297

* bindings/jni/jni_instance.h:

12298

(KJS::Bindings::JObjectWrapper::ref):

12299

(KJS::Bindings::JObjectWrapper::deref):

12300

* bindings/jni/jni_runtime.cpp: Made _array a RefPtr

12301

(JavaArray::~JavaArray):

12302

(JavaArray::JavaArray):

12303

* bindings/jni/jni_runtime.h:

12304

(KJS::Bindings::JavaArray::operator=):

12305

* bindings/objc/objc_runtime.h:

12306

- Prohibited copying because that would muss the ref count.

12307

- Prohibited construction without instance because an instance wrapper

12308

without an instance is almost certainly a bug.

12309

* bindings/objc/objc_runtime.mm:

12310

(ObjcFallbackObjectImp::ObjcFallbackObjectImp):

12311

* bindings/runtime.cpp:

12312

(KJS::Bindings::Instance::Instance):

12313

(KJS::Bindings::Instance::createBindingForLanguageInstance):

12314

(KJS::Bindings::Instance::createRuntimeObject):

12315

* bindings/runtime.h:

12316

(KJS::Bindings::Instance::ref):

12317

(KJS::Bindings::Instance::deref):

12318

* bindings/runtime_object.cpp:

12319

(RuntimeObjectImp::RuntimeObjectImp):

12320

(RuntimeObjectImp::fallbackObjectGetter):

12321

(RuntimeObjectImp::fieldGetter):

12322

(RuntimeObjectImp::methodGetter):

12323

(RuntimeObjectImp::getOwnPropertySlot):

12324

(RuntimeObjectImp::put):

12325

(RuntimeObjectImp::canPut):

12326

* bindings/runtime_object.h:

12327

- Removed ownsInstance data member because RefPtr takes care of

12328

instance lifetime now.

12329

- Prohibited copying because that would muss the ref count.

12330

- Prohibited construction without instance because an instance wrapper

12331

without an instance is almost certainly a bug.

12332

(KJS::RuntimeObjectImp::getInternalInstance):

12333

12334

2006-02-15 Geoffrey Garen <ggaren@apple.com>

12335

12336

Reviewed by John.

12337

12338

- Applied the 4330457 change to CClass and ObjcClass as well.

12339

12340

Once plugins work in DumpRenderTree, running run-webkit-tests --leaks

12341

will catch this.

12342

12343

This change isn't as critical because CClass and ObjcClass objects get

12344

cached globally and never deleted, but it's good practice, in case we

12345

ever do decide to delete CClass and ObjcClass objects.

12346

12347

This change requires prohibiting copying, because we don't do any

12348

intelligent ref-counting -- when a Class is destroyed, it destroys its

12349

methods and fields unconditionally. (Java classes already prohibited

12350

copying.)

12351

12352

* bindings/c/c_class.cpp:

12353

- Merged _commonInit and _commonDelete into constructor and destructor.

12354

(CClass::CClass):

12355

(CClass::~CClass):

12356

(CClass::methodsNamed): Added delete callbacks

12357

(CClass::fieldNamed): Added delete callbacks

12358

* bindings/c/c_class.h: Prohibited copying

12359

* bindings/c/c_instance.cpp:

12360

(KJS::Bindings::CInstance::getClass): Changed to use the preferred

12361

class factory method, to take advantage of the global cache.

12362

12363

[ Repeated changes applied to CClass for ObjcClass: ]

12364

12365

* bindings/objc/objc_class.h:

12366

* bindings/objc/objc_class.mm:

12367

(KJS::Bindings::ObjcClass::ObjcClass):

12368

(KJS::Bindings::ObjcClass::~ObjcClass):

12369

(KJS::Bindings::ObjcClass::methodsNamed):

12370

(KJS::Bindings::ObjcClass::fieldNamed):

12371

* bindings/objc/objc_runtime.h:

12372

(KJS::Bindings::ObjcMethod::ObjcMethod): Initialized uninitialized

12373

variable to prevent bad CFRelease.

12374

(KJS::Bindings::ObjcMethod::~ObjcMethod): Removed erroneous ';' from

12375

if statement to prevent bad CFRelease.

12376

* bindings/objc/objc_runtime.cpp: Changed to use the preferred

12377

ObjectStructPtr, for clarity.

12378

12379

2006-02-14 Geoffrey Garen <ggaren@apple.com>

12380

12381

Reviewed by John.

12382

12383

- Fixed <rdar://problem/4330457> CrashTracer: [REGRESSION] 3763 crashes

12384

in Safari at com.apple.JavaScriptCore: KJS::Bindings::JavaInstance::

12385

getClass const + 56

12386

12387

Once plugins work in DumpRenderTree, running run-webkit-tests --leaks

12388

will catch this.

12389

12390

This was a memory leak in the bindings code. The leak was so extreme

12391

that it would cause Safari or the JVM to abort from lack of memory.

12392

Upon construction, Class objects create field and method objects,

12393

storing them in CFDictionaries. The bug was that upon destruction, the

12394

class objects released the dictionaries but didn't destroy the stored

12395

objects.

12396

12397

The fix is to supply CFDictionary callbacks for destroying the values

12398

added to the dictionary.

12399

12400

* bindings/jni/jni_class.cpp:

12401

(JavaClass::JavaClass): Added delete callbacks

12402

* bindings/runtime.cpp: Added definitions for delete callbacks

12403

(KJS::Bindings::deleteMethodList):

12404

(KJS::Bindings::deleteMethod):

12405

(KJS::Bindings::deleteField):

12406

* bindings/runtime.h: Added declarations for delete callbacks

12407

12408

2006-02-14 Timothy Hatcher <timothy@apple.com>

12409

12410

Reviewed by Justin.

12411

12412

Fixed <rdar://problem/4415050> STD: WebCore build steps use echo -n, which will change

12413

behavior due to POSIX version of sh

12414

12415

* JavaScriptCore.xcodeproj/project.pbxproj: removed the use of echo -n, replaced with printf ""

12416

12417

2006-02-13 Dave Hyatt <hyatt@apple.com>

12418

12419

Fix Win32 bustage in JavaScriptCore.

12420

12421

Reviewed by darin

12422

12423

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

12424

Add JSImmediate to the Win32 project.

12425

12426

* kjs/JSImmediate.h:

12427

(KJS::JSImmediate::fromDouble):

12428

(KJS::JSImmediate::toDouble):

12429

(KJS::JSImmediate::NanAsBits):

12430

(KJS::JSImmediate::oneAsBits):

12431

Win32 needs explicit returns after abort() for non-void functions.

12432

12433

* kjs/testkjs.cpp:

12434

(run):

12435

Win32 catches a bug in testkjs! The "return 2" should actually

12436

have been a return false.

12437

12438

* kjs/value.h:

12439

The extern decls of NaN and Inf need to be const.

12440

12441

=== JavaScriptCore-521.7 ===

12442

12443

2006-02-13 Timothy Hatcher <timothy@apple.com>

12444

12445

Reviewed by Darin.

12446

12447

Replaced the old NS_DURING exception blocking with @try/@catch.

12448

12449

* JavaScriptCorePrefix.h: undef try and catch to workaround a C++ conflict

12450

* bindings/objc/objc_instance.mm:

12451

(ObjcInstance::invokeMethod):

12452

(ObjcInstance::invokeDefaultMethod):

12453

(ObjcInstance::setValueOfUndefinedField):

12454

(ObjcInstance::getValueOfUndefinedField):

12455

* bindings/objc/objc_runtime.mm:

12456

(ObjcField::valueFromInstance):

12457

(ObjcField::setValueToInstance):

12458

(ObjcArray::setValueAt):

12459

(ObjcArray::valueAt):

12460

12461

2006-02-13 Darin Adler <darin@apple.com>

12462

12463

- fix a couple problems building on Windows, based on requests

12464

from Krzysztof Kowalczyk <kkowalczyk@gmail.com>

12465

12466

* kjs/JSImmediate.h: Change code using non-standard u_int32/64_t types

12467

to the standard uint32/64_t. Also removed curious "isIEEE()" function

12468

that checked the sizes of some types (and type sizes alone don't tell you if

12469

the floating point conforms to the IEEE-standard). Added missing include

12470

of <stdint.h>.

12471

12472

* kjs/property_slot.h: Added missing include of <assert.h>.

12473

12474

2006-02-12 Geoffrey Garen <ggaren@apple.com>

12475

12476

Reviewed by darin.

12477

12478

Cleaned up testkjs, added new "run" functionality to allow scripting

12479

tests from within JS. ("run" is a part of my new super-accurate

12480

JS iBench.)

12481

12482

No regressions in run-javascriptcore-tests.

12483

12484

* kjs/testkjs.cpp:

12485

(GlobalImp::className):

12486

(TestFunctionImp::):

12487

(TestFunctionImp::callAsFunction):

12488

(main):

12489

(run):

12490

12491

2006-02-11 Alexey Proskuryakov <ap@nypop.com>

12492

12493

Reviewed by Darin.

12494

12495

- improve fix for http://bugs.webkit.org/show_bug.cgi?id=5163

12496

RealPlayer.GetTitle() Crashes Safari/Dashboard

12497

12498

* bindings/c/c_utility.cpp: (KJS::Bindings::convertUTF8ToUTF16):

12499

Use kCFStringEncodingISOLatin1 rather than kCFStringEncodingWindowsLatin1,

12500

because the latter encoding has holes, and conversion can still fail.

12501

12502

2006-02-10 Geoffrey Garen <ggaren@apple.com>

12503

12504

Reviewed by Darin.

12505

12506

- Inlined RefPtr assignment operators. .7% performance win on

12507

super-accurate JS iBench.

12508

12509

* kxmlcore/RefPtr.h:

12510

(KXMLCore::::operator):

12511

12512

2006-02-10 Geoffrey Garen <ggaren@apple.com>

12513

12514

No review needed, just a build fix. This time for sure.

12515

12516

* kjs/JSType.h:

12517

12518

2006-02-10 Geoffrey Garen <ggaren@apple.com>

12519

12520

Reviewed by eric.

12521

12522

- Fixed build. As it goes without saying, I will not mention that I

12523

blame Kevin.

12524

12525

* JavaScriptCore.xcodeproj/project.pbxproj:

12526

* kjs/JSImmediate.cpp:

12527

(KJS::JSImmediate::toObject):

12528

12529

2006-02-09 Geoffrey Garen <ggaren@apple.com>

12530

12531

Reviewed by mjs.

12532

12533

- Fixed <rdar://problem/4343730> Should switch ConstantValues (null,

12534

undefined, true, false) from JS objects to immediate values similar to

12535

SimpleNumber

12536

12537

2.0% performance gain on my new super-accurate version of JS iBench.

12538

(I promise to land a version of it soon.)

12539

12540

The gist of the change:

12541

(1) The SimpleNumber class (simple_number.h) is now the JSImmediate

12542

class (JSImmediate.h/.cpp), and it handles not only numbers but also

12543

null, undefined, true, and false.

12544

(2) JSImmediate provides convenience methods for the bit masking

12545

necessary to encode and decode immediate values.

12546

(3) ConstantValues, BooleanImp, NullImp, and UndefinedImp are gone.

12547

(4) JSCell no longer implements functions like getBoolean, because

12548

only a JSImmediate can be a boolean.

12549

(5) JSImmediate no longer uses ALWAYS_INLINE because there's no need,

12550

and ALWAYS_INLINE is a non-portable option of last resort.

12551

(6) Type is now JSType, and it resides in its own file, JSType.h.

12552

Since I was there, I did some header include sorting as part of this

12553

change.

12554

12555

The rest pretty much explains itself.

12556

12557

* JavaScriptCore.xcodeproj/project.pbxproj: Removed simple_number.h,

12558

added JSImmediate.h/.cpp.

12559

* bindings/c/c_instance.cpp:

12560

(KJS::Bindings::CInstance::defaultValue):

12561

* bindings/c/c_instance.h:

12562

* bindings/c/c_utility.cpp:

12563

(KJS::Bindings::convertValueToNPVariant):

12564

* bindings/jni/jni_instance.cpp:

12565

(JavaInstance::defaultValue):

12566

* bindings/jni/jni_instance.h:

12567

* bindings/jni/jni_jsobject.cpp:

12568

(JavaJSObject::convertValueToJObject):

12569

* bindings/objc/WebScriptObject.mm:

12570

(+[WebScriptObject _convertValueToObjcValue:originExecutionContext:executionContext:]):

12571

Standardized calls to use getXXX instead of hand-rolling JSValue

12572

functionality.

12573

* bindings/objc/objc_instance.h:

12574

* bindings/objc/objc_instance.mm:

12575

(ObjcInstance::getValueOfUndefinedField):

12576

(ObjcInstance::defaultValue):

12577

* bindings/objc/objc_runtime.h:

12578

* bindings/objc/objc_runtime.mm:

12579

(ObjcFallbackObjectImp::type):

12580

(ObjcFallbackObjectImp::defaultValue):

12581

* bindings/runtime.h:

12582

(KJS::Bindings::Instance::getValueOfUndefinedField):

12583

* bindings/runtime_object.cpp:

12584

(RuntimeObjectImp::defaultValue):

12585

* bindings/runtime_object.h:

12586

* kjs/JSImmediate.h: Added.

12587

(KJS::JSImmediate::isImmediate):

12588

(KJS::JSImmediate::isNumber):

12589

(KJS::JSImmediate::isBoolean):

12590

(KJS::JSImmediate::isUndefinedOrNull):

12591

(KJS::JSImmediate::fromDouble):

12592

(KJS::JSImmediate::toDouble):

12593

(KJS::JSImmediate::toBoolean):

12594

(KJS::JSImmediate::trueImmediate):

12595

(KJS::JSImmediate::falseImmediate):

12596

(KJS::JSImmediate::NaNImmediate):

12597

(KJS::JSImmediate::undefinedImmediate):

12598

(KJS::JSImmediate::nullImmediate):

12599

(KJS::JSImmediate::tag):

12600

(KJS::JSImmediate::unTag):

12601

(KJS::JSImmediate::getTag):

12602

(KJS::JSImmediate::):

12603

(KJS::JSImmediate::isIEEE):

12604

(KJS::JSImmediate::is32bit):

12605

(KJS::JSImmediate::is64bit):

12606

(KJS::JSImmediate::NanAsBits):

12607

(KJS::JSImmediate::zeroAsBits):

12608

(KJS::JSImmediate::oneAsBits):

12609

* kjs/JSLock.cpp:

12610

(KJS::JSLock::lock): Removed hack-o-rama to initialize ConstantValues.

12611

* kjs/JSType.h: Added.

12612

* kjs/collector.cpp:

12613

(KJS::Collector::protect):

12614

(KJS::Collector::unprotect):

12615

(KJS::Collector::collect):

12616

* kjs/internal.cpp:

12617

(KJS::StringImp::toPrimitive):

12618

(KJS::NumberImp::toPrimitive):

12619

(KJS::NumberImp::toBoolean):

12620

(KJS::GetterSetterImp::toPrimitive):

12621

* kjs/internal.h:

12622

(KJS::StringImp::type):

12623

(KJS::NumberImp::type):

12624

* kjs/object.cpp:

12625

(KJS::JSObject::type):

12626

(KJS::tryGetAndCallProperty): Replaced "Are you one of the six things

12627

I'm looking for?" test with "Are you not the one thing I'm not looking

12628

for" test.

12629

(KJS::JSObject::defaultValue):

12630

(KJS::JSObject::toPrimitive):

12631

* kjs/object.h:

12632

(KJS::GetterSetterImp::type):

12633

(KJS::JSValue::isObject):

12634

* kjs/operations.cpp:

12635

(KJS::equal):

12636

(KJS::strictEqual):

12637

(KJS::add):

12638

* kjs/reference.cpp:

12639

(KJS::Reference::deleteValue):

12640

* kjs/simple_number.h: Removed.

12641

* kjs/string_object.cpp:

12642

(StringInstance::getOwnPropertySlot): fixed indentation

12643

* kjs/value.cpp:

12644

(KJS::JSValue::toObject):

12645

(KJS::jsNumberCell): New function to quarantine a PIC branch -- allows

12646

us to inline jsNumber without adding PIC branches to callers.

12647

* kjs/value.h:

12648

(KJS::jsUndefined):

12649

(KJS::jsNull):

12650

(KJS::jsNaN):

12651

(KJS::jsBoolean):

12652

(KJS::jsNumber):

12653

(KJS::JSValue::downcast):

12654

(KJS::JSValue::isUndefinedOrNull):

12655

(KJS::JSValue::isBoolean):

12656

(KJS::JSValue::isNumber):

12657

(KJS::JSValue::isString):

12658

(KJS::JSValue::isObject):

12659

(KJS::JSValue::getBoolean):

12660

(KJS::JSValue::getNumber):

12661

(KJS::JSValue::getString):

12662

(KJS::JSValue::getObject):

12663

(KJS::JSValue::getUInt32):

12664

(KJS::JSValue::mark): Replaced !JSImmediate::is() test with assertion,

12665

resulting in a slight performance gain. Callers should always check

12666

!marked() before calling mark(), so it's impossible to call mark on

12667

a JSImmediate.

12668

(KJS::JSValue::marked):

12669

(KJS::JSValue::type):

12670

(KJS::JSValue::toPrimitive):

12671

(KJS::JSValue::toBoolean):

12672

(KJS::JSValue::toNumber):

12673

(KJS::JSValue::toString):

12674

12675

2006-02-06 Eric Seidel <eseidel@apple.com>

12676

12677

Add svn:ignore properties for visual studio internals.

12678

12679

2006-02-06 Alexey Proskuryakov <ap@nypop.com>

12680

12681

Reviewed by Darin.

12682

12683

- Refactor DateInstance to provide direct access to data. Several WIN32 modifications.

12684

http://bugs.webkit.org/show_bug.cgi?id=7107

12685

12686

- No tests added - only changed functionality on WIN32, which should be covered by

12687

existing tests.

12688

12689

* kjs/date_object.cpp:

12690

(gmtoffset): On WIN32, use the recommended global (_timezone rather than timezone).

12691

Updated comments.

12692

(KJS::timeZoneOffset): Removed, was basically the same as the above.

12693

(KJS::formatTime): Pass an UTC flag - UTC/local cannot be correctly selected on

12694

Windows based on struct tm itself.

12695

(KJS::DateInstance::getTime): Added.

12696

(KJS::DateInstance::getUTCTime): Added.

12697

(KJS::millisecondsToTM): Factored out from DateProtoFunc::callAsFunction().

12698

(KJS::DateObjectImp::callAsFunction): Use the new parameter to formatTime().

12699

(KJS::DateProtoFunc::callAsFunction): Updated for the other changes. The code for

12700

GetTimezoneOffset was incorrect on WIN32 - _daylight global has nothing to do

12701

with daylight savings time being in effect.

12702

12703

* kjs/date_object.h: Added prototypes for new functions.

12704

12705

2006-02-05 Maciej Stachowiak <mjs@apple.com>

12706

12707

Reviewed by Anders.

12708

12709

- fixed ~1100 KJS::Node leaked on layout tests

12710

http://bugs.webkit.org/show_bug.cgi?id=7097

12711

12712

* kjs/internal.cpp:

12713

(KJS::Parser::noteNodeCycle):

12714

(KJS::Parser::removeNodeCycle):

12715

(KJS::clearNewNodes):

12716

* kjs/internal.h:

12717

* kjs/nodes.cpp:

12718

(ElementNode::breakCycle):

12719

(PropertyListNode::breakCycle):

12720

(ArgumentListNode::breakCycle):

12721

(StatListNode::StatListNode):

12722

(StatListNode::breakCycle):

12723

(VarDeclListNode::breakCycle):

12724

(BlockNode::BlockNode):

12725

(ClauseListNode::breakCycle):

12726

(CaseBlockNode::CaseBlockNode):

12727

(ParameterNode::breakCycle):

12728

(SourceElementsNode::SourceElementsNode):

12729

(SourceElementsNode::breakCycle):

12730

* kjs/nodes.h:

12731

(KJS::Node::breakCycle):

12732

(KJS::ElementNode::ElementNode):

12733

(KJS::ArrayNode::ArrayNode):

12734

(KJS::PropertyListNode::PropertyListNode):

12735

(KJS::ObjectLiteralNode::ObjectLiteralNode):

12736

(KJS::ArgumentListNode::ArgumentListNode):

12737

(KJS::ArgumentsNode::ArgumentsNode):

12738

(KJS::VarDeclListNode::VarDeclListNode):

12739

(KJS::VarStatementNode::VarStatementNode):

12740

(KJS::ForNode::ForNode):

12741

(KJS::CaseClauseNode::CaseClauseNode):

12742

(KJS::ClauseListNode::ClauseListNode):

12743

(KJS::ParameterNode::ParameterNode):

12744

(KJS::FuncExprNode::FuncExprNode):

12745

(KJS::FuncDeclNode::FuncDeclNode):

12746

12747

2006-02-05 Maciej Stachowiak <mjs@apple.com>

12748

12749

Reviewed by Hyatt.

12750

12751

- fix default traits for classes to make sure default constructors get called

12752

12753

* kxmlcore/VectorTraits.h:

12754

(KXMLCore::):

12755

12756

2006-02-04 Darin Adler <darin@apple.com>

12757

12758

Reviewed by Maciej.

12759

12760

- fix http://bugs.webkit.org/show_bug.cgi?id=5210

12761

REGRESSION: for/in loop with var changes global variable instead of making local

12762

12763

Test: fast/js/for-in-var-scope.html

12764

12765

* kjs/nodes.cpp:

12766

(valueForReadModifyAssignment): Use ALWAYS_INLINE macro.

12767

(ForInNode::execute): Break out of the scope chain loop once we find and set the

12768

loop variable. We don't want to set multiple loop variables.

12769

(ForInNode::processVarDecls): Process the declaration of the loop variable.

12770

12771

- other cleanup

12772

12773

* kjs/object.cpp: (KJS::tryGetAndCallProperty): Use ALWAYS_INLINE macro.

12774

* kxmlcore/FastMalloc.cpp: Change to use ALWAYS_INLINE macro from AlwaysInline.h

12775

instead of defining it here a second time.

12776

12777

2006-02-04 Maciej Stachowiak <mjs@apple.com>

12778

12779

Reviewed by Hyatt.

12780

12781

- change JavaScript collector statistics calls to use HashCountedSet instead

12782

of CFSet; other misc cleanup

12783

http://bugs.webkit.org/show_bug.cgi?id=7072

12784

12785

* kjs/collector.cpp:

12786

(KJS::Collector::numProtectedObjects): renamed from numReferencedObjects

12787

(KJS::typeName):

12788

(KJS::Collector::rootObjectTypeCounts): renamed from rootObjectClasses,

12789

use HashSet

12790

* kjs/collector.h:

12791

(KJS::Collector::isOutOfMemory): Renamed from outOfMemory.

12792

* kjs/nodes.cpp:

12793

12794

2006-02-03 Timothy Hatcher <timothy@apple.com>

12795

12796

Reviewed by Justin.

12797

12798

Renamed configuration names to Debug, Release and Production.

12799

12800

* JavaScriptCore.xcodeproj/project.pbxproj:

12801

12802

2006-02-02 George Staikos <staikos@opensource.apple.com>

12803

12804

Reviewed by Maciej.

12805

12806

* kjs/lookup.h: Fix compile, merged from KDE.

12807

12808

2006-02-02 Darin Adler <darin@apple.com>

12809

12810

Reviewed by Maciej.

12811

12812

- http://bugs.webkit.org/show_bug.cgi?id=7005

12813

add Noncopyable, OwnPtr, OwnArrayPtr to KXMLCore

12814

12815

* kxmlcore/Noncopyable.h: Added.

12816

* kxmlcore/OwnArrayPtr.h: Added.

12817

* kxmlcore/OwnPtr.h: Added.

12818

12819

* JavaScriptCore.xcodeproj/project.pbxproj: Added new files.

12820

12821

* kjs/function.h:

12822

* kjs/function.cpp: Use OwnPtr for Parameter pointers.

12823

12824

* kjs/internal.h: Use Noncopyable for LabelStack.

12825

12826

* kjs/list.cpp: Use OwnArrayPtr for overflow.

12827

12828

* kjs/property_map.h:

12829

* kjs/property_map.cpp: Use OwnArrayPtr for SavedProperties.

12830

Use Vector for some stack buffers.

12831

12832

* kjs/regexp_object.h:

12833

* kjs/regexp_object.cpp: Use OwnArrayPtr for lastOvector.

12834

12835

2006-01-31 Maciej Stachowiak <mjs@apple.com>

12836

12837

Reviewed by Darin.

12838

12839

- fixed leak of hundreds of thousands of JS parser nodes on the layout tests, and added an exit counter

12840

that would catch them

12841

12842

* kjs/nodes.cpp:

12843

(NodeCounter::~NodeCounter): Added debug-only node counter.

12844

(Node::Node):

12845

(Node::~Node):

12846

* kxmlcore/VectorTraits.h: Simple classes like RefPtr do in fact need destruction.

12847

12848

2006-01-31 Darin Adler <darin@apple.com>

12849

12850

Reviewed by Maciej.

12851

12852

- added deleteAllValues for HashSet as well as HashMap

12853

- fixed conversion from const_iterator to iterator, which I broke a while back

12854

12855

* kxmlcore/HashMap.h: Updated copyright date.

12856

* kxmlcore/HashSet.h: (KXMLCore::deleteAllValues): Added.

12857

* kxmlcore/HashTable.h: (KXMLCore::HashTableIterator::operator const_iterator): Added.

12858

12859

2006-01-31 Tim Omernick <timo@apple.com>

12860

12861

Reviewed by Geoff Garen.

12862

12863

* bindings/c/c_utility.cpp:

12864

(KJS::Bindings::convertUTF8ToUTF16):

12865

Fixed an invalid assertion that UTF8Chars is not NULL. It is valid for it to be NULL as long as

12866

UTF8Length is 0.

12867

This fixes an assertion failure on TOT at <http://www.musicindiaonline.com/p/x/tJO0OOBME9.As1NMvHdW/>,

12868

where JavaScript is getting a NULL string back from some call on the Real Player plugin.

12869

12870

2006-01-30 Anders Carlsson <andersca@mac.com>

12871

12872

Reviewed by Darin.

12873

12874

Fix http://bugs.webkit.org/show_bug.cgi?id=6907

12875

REGRESSION: United.com menus messed up due to document.all/MSIE sniff

12876

12877

* kjs/nodes.cpp:

12878

(typeStringForValue):

12879

Return "undefined" if the given object should masquerade as undefined.

12880

12881

* kjs/object.h:

12882

(KJS::JSObject::masqueradeAsUndefined):

12883

Rename from isEqualToNull.

12884

12885

* kjs/operations.cpp:

12886

(KJS::equal):

12887

Update for name change.

12888

12889

2006-01-29 Maciej Stachowiak <mjs@apple.com>

12890

12891

Reviewed by Darin.

12892

12893

- properly define Vector assignment operator; the private version was accidentally left

12894

in, and the template version is not enough to replace the default

12895

12896

* kxmlcore/Vector.h:

12897

(KXMLCore::Vector::operator=):

12898

12899

2006-01-29 Eric Seidel <eseidel@apple.com>

12900

12901

Reviewed by darin.

12902

12903

Fix the build by applying a GCC-specific namespace hack.

12904

12905

* kjs/lookup.h:

12906

12907

2006-01-29 Eric Seidel <eseidel@apple.com>

12908

12909

Reviewed by hyatt.

12910

12911

Fix build on Win32.

12912

12913

* kjs/lookup.h: fixed ::cacheGlobalObject

12914

* kxmlcore/Vector.h:

12915

(KXMLCore::Vector::operator[]): use unsigned long

12916

12917

2006-01-29 Maciej Stachowiak <mjs@apple.com>

12918

12919

Reviewed by Dave Hyatt.

12920

12921

* kxmlcore/Vector.h:

12922

(KXMLCore::Vector::operator[]): Add unsigned overload

12923

12924

2006-01-28 Darin Adler <darin@apple.com>

12925

12926

Reviewed by John Sullivan.

12927

12928

- http://bugs.webkit.org/show_bug.cgi?id=6895

12929

include exception names in JavaScript form of DOM exception

12930

12931

* khtml/ecma/kjs_binding.cpp: (KJS::setDOMException): Include the name of the

12932

exception in the error message.

12933

12934

2006-01-28 Maciej Stachowiak <mjs@apple.com>

12935

12936

Reviewed by Darin.

12937

12938

- miscellaneous Vector improvements

12939

12940

* kxmlcore/Vector.h:

12941

(KXMLCore::Vector::at): Add range-checking asserts.

12942

(KXMLCore::Vector::first): Added as a convenience.

12943

(KXMLCore::Vector::last): Convenience for stack-style use.

12944

(KXMLCore::Vector::removeLast): ditto

12945

12946

2006-01-28 Darin Adler <darin@apple.com>

12947

12948

Reviewed by John Sullivan

12949

12950

- fix http://bugs.webkit.org/show_bug.cgi?id=6870

12951

REGRESSION: JavaScript Date constructor won't accept another Date object

12952

12953

Test: fast/js/date-constructor.html

12954

12955

* kjs/date_object.cpp: (KJS::DateObjectImp::construct):

12956

Added a special case for constructing one date from another (to avoid losing

12957

milliseconds, which are not in the text form, to match Firefox), and changed

12958

the base code to convert to primitive before checking for string to match

12959

the standard. Also corrected a couple silly things in the "construct from

12960

current time" code path (removed a floor that does no good, and changed

12961

the constant used to convert microseconds to milliseconds to be a 1000

12962

rather than "msPerSecond").

12963

12964

2006-01-28 Darin Adler <darin@apple.com>

12965

12966

* kjs/create_hash_table: Added missing license.

12967

12968

2006-01-28 Maciej Stachowiak <mjs@apple.com>

12969

12970

Reviewed by Dave Hyatt.

12971

12972

- added a Vector class

12973

http://bugs.webkit.org/show_bug.cgi?id=6894

12974

12975

* JavaScriptCore.xcodeproj/project.pbxproj:

12976

* kjs/internal.cpp:

12977

(KJS::Parser::saveNewNode): Apply Vector.

12978

(KJS::clearNewNodes): ditto

12979

* kjs/number_object.cpp:

12980

(integer_part_noexp): ditto

12981

(char_sequence): ditto

12982

* kjs/ustring.cpp:

12983

(KJS::UString::UTF8String): ditto

12984

* kxmlcore/HashMap.h:

12985

(KXMLCore::deleteAllValues): Tweaked this to only apply to HashMap,

12986

other versions are useful for other containers.

12987

* kxmlcore/Vector.h: Added. Implemented a Vector class, which should

12988

be usable for all Array/QVector style purposes, and also as a stack buffer

12989

with oversize handling. Also some helper classes to make vector operations

12990

as efficient as possible for POD types and for simple non-PODs like RefPtr.

12991

(KXMLCore::):

12992

(KXMLCore::VectorTypeOperations::destruct):

12993

(KXMLCore::VectorTypeOperations::initialize):

12994

(KXMLCore::VectorTypeOperations::move):

12995

(KXMLCore::VectorTypeOperations::uninitializedCopy):

12996

(KXMLCore::VectorTypeOperations::uninitializedFill):

12997

(KXMLCore::VectorBuffer::VectorBuffer):

12998

(KXMLCore::VectorBuffer::~VectorBuffer):

12999

(KXMLCore::VectorBuffer::deallocateBuffer):

13000

(KXMLCore::VectorBuffer::inlineBuffer):

13001

(KXMLCore::Vector::Vector):

13002

(KXMLCore::Vector::~Vector):

13003

(KXMLCore::Vector::size):

13004

(KXMLCore::Vector::capacity):

13005

(KXMLCore::Vector::isEmpty):

13006

(KXMLCore::Vector::at):

13007

(KXMLCore::Vector::operator[]):

13008

(KXMLCore::Vector::data):

13009

(KXMLCore::Vector::operator T*):

13010

(KXMLCore::Vector::operator const T*):

13011

(KXMLCore::Vector::begin):

13012

(KXMLCore::Vector::end):

13013

(KXMLCore::Vector::clear):

13014

(KXMLCore::Vector::fill):

13015

(KXMLCore::Vector::operator=):

13016

(KXMLCore::::Vector):

13017

(KXMLCore::::operator):

13018

(KXMLCore::::fill):

13019

(KXMLCore::::expandCapacity):

13020

(KXMLCore::::resize):

13021

(KXMLCore::::reserveCapacity):

13022

(KXMLCore::::append):

13023

(KXMLCore::deleteAllValues):

13024

* kxmlcore/VectorTraits.h: Added.

13025

(KXMLCore::VectorTraits): Traits to enable making Vector efficient for

13026

simple types.

13027

13028

2006-01-28 Alexey Proskuryakov <ap@nypop.com>

13029

13030

Reviewed by Darin.

13031

13032

- fix http://bugs.webkit.org/show_bug.cgi?id=5163

13033

RealPlayer.GetTitle() Crashes Safari/Dashboard

13034

13035

* bindings/c/c_utility.cpp: (KJS::Bindings::convertUTF8ToUTF16):

13036

Fallback to kCFStringEncodingWindowsLatin1 if the passed buffer is not valid UTF-8, preventing crashes.

13037

13038

2006-01-25 George Staikos <staikos@opensource.apple.com>

13039

13040

Reviewed by Darin.

13041

13042

* kxmlcore/HashFunctions.h: Merge build fix from KDE.

13043

13044

2006-01-25 Darin Adler <darin@apple.com>

13045

13046

- removed an unused source file

13047

13048

* kjs/pointer_hash.h: Removed.

13049

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: Removed reference to pointer_hash.h.

13050

13051

2006-01-23 Anders Carlsson <andersca@mac.com>

13052

13053

Reviewed by Maciej.

13054

13055

- fix http://bugs.webkit.org/show_bug.cgi?id=6737

13056

KJS_DEFINE_PROTOTYPE should work outside of the KJS namespace

13057

13058

* kjs/lookup.h:

13059

Prefix all KJS types with KJS:: in KJS_DEFINE_PROTOTYPE.

13060

13061

(cacheGlobalObject):

13062

Move this out of the KJS namespace.

13063

13064

2006-01-23 Maciej Stachowiak <mjs@apple.com>

13065

13066

Reviewed by Eric.

13067

13068

- renamed PointerHash to PtrHash

13069

- made PtrHash the default hash function for int and pointer types that aren't further specialized

13070

- added an AtomicStringImpl class to make it easier and more typesafe to identity hash atomic strings

13071

- did appropriate consequent cleanup (very few places now need to declare a hash function)

13072

http://bugs.webkit.org/show_bug.cgi?id=6752

13073

13074

* kjs/array_object.cpp:

13075

(ArrayProtoFunc::callAsFunction): no need to mention PointerHash

13076

* kjs/collector.cpp: ditto

13077

* kjs/identifier.cpp:

13078

(KXMLCore::): declare DefaultHash the new way

13079

* kjs/internal.cpp: no need to mention PointerHash

13080

* kjs/ustring.h:

13081

* kxmlcore/HashCountedSet.h: change how we get the default hash to make it

13082

easier to specialize on PtrHash

13083

* kxmlcore/HashFunctions.h:

13084

(KXMLCore::): renamed PointerHash to PtrHash; changed DefaultHash so that it has

13085

a Hash typedef rather than being a hash function class itself; declared DefaultHash

13086

for int and partializy specialized for pointer types

13087

* kxmlcore/HashMapPtrSpec.h:

13088

(KXMLCore::PtrHashIteratorAdapter::PtrHashIteratorAdapter): Slight tweaks for new

13089

way of handling pointer hash

13090

(KXMLCore::PtrHashConstIteratorAdapter::PtrHashConstIteratorAdapter): ditto

13091

(KXMLCore::): ditto

13092

* kxmlcore/HashMap.h: ditto

13093

* kxmlcore/HashSet.h: ditto

13094

13095

2006-01-23 Maciej Stachowiak <mjs@apple.com>

13096

13097

Reviewed by Tim Omernick.

13098

13099

- use classes instead of free functions for extractors, this better matches how other

13100

things work and should avoid the need for hacky workarounds on other compilers

13101

http://bugs.webkit.org/show_bug.cgi?id=6748

13102

13103

* kjs/array_object.cpp:

13104

* kjs/identifier.cpp:

13105

* kjs/internal.cpp:

13106

* kxmlcore/HashMap.h:

13107

(KXMLCore::PairFirstExtractor::extract):

13108

* kxmlcore/HashMapPtrSpec.h:

13109

(KXMLCore::):

13110

* kxmlcore/HashSet.h:

13111

(KXMLCore::IdentityExtractor::extract):

13112

* kxmlcore/HashTable.h:

13113

(KXMLCore::addIterator):

13114

(KXMLCore::removeIterator):

13115

(KXMLCore::HashTable::add):

13116

(KXMLCore::HashTable::isEmptyBucket):

13117

(KXMLCore::HashTable::isDeletedBucket):

13118

(KXMLCore::HashTable::HashTable):

13119

(KXMLCore::HashTable::lookup):

13120

(KXMLCore::HashTable::add):

13121

(KXMLCore::HashTable::reinsert):

13122

(KXMLCore::HashTable::find):

13123

(KXMLCore::HashTable::contains):

13124

(KXMLCore::HashTable::remove):

13125

(KXMLCore::HashTable::allocateTable):

13126

(KXMLCore::HashTable::deallocateTable):

13127

(KXMLCore::HashTable::expand):

13128

(KXMLCore::HashTable::rehash):

13129

(KXMLCore::HashTable::clear):

13130

(KXMLCore::HashTable::swap):

13131

(KXMLCore::HashTable::operator):

13132

(KXMLCore::HashTable::checkTableConsistency):

13133

(KXMLCore::HashTable::checkTableConsistencyExceptSize):

13134

(KXMLCore::HashTable::invalidateIterators):

13135

13136

2006-01-23 Maciej Stachowiak <mjs@apple.com>

13137

13138

Rubber stamped by Tim Hatcher.

13139

13140

- renamed inert() operation on HashSet, HashCountedSet and HashTable to add()

13141

for consistency with HashMap

13142

13143

* kjs/array_object.cpp:

13144

(ArrayProtoFunc::callAsFunction):

13145

* kjs/collector.cpp:

13146

(KJS::Collector::protect):

13147

* kjs/identifier.cpp:

13148

(KJS::Identifier::add):

13149

* kxmlcore/HashCountedSet.h:

13150

(KXMLCore::::add):

13151

* kxmlcore/HashMap.h:

13152

(KXMLCore::::inlineAdd):

13153

* kxmlcore/HashSet.h:

13154

(KXMLCore::::add):

13155

* kxmlcore/HashTable.h:

13156

(KXMLCore::HashTable::add):

13157

(KXMLCore::::add):

13158

(KXMLCore::::HashTable):

13159

13160

2006-01-23 Justin Garcia <justin.garcia@apple.com>

13161

13162

Reviewed by thatcher

13163

13164

Turned on -O2 for B&I build.

13165

13166

* JavaScriptCore.xcodeproj/project.pbxproj:

13167

13168

2006-01-23 Maciej Stachowiak <mjs@apple.com>

13169

13170

Reviewed by Tim Hatcher.

13171

13172

- it's "Franklin Street", not "Franklin Steet"

13173

13174

* kjs/array_instance.h:

13175

* kjs/array_object.cpp:

13176

* kjs/array_object.h:

13177

* kjs/bool_object.cpp:

13178

* kjs/bool_object.h:

13179

* kjs/collector.cpp:

13180

* kjs/collector.h:

13181

* kjs/completion.h:

13182

* kjs/context.h:

13183

* kjs/date_object.cpp:

13184

* kjs/date_object.h:

13185

* kjs/debugger.cpp:

13186

* kjs/debugger.h:

13187

* kjs/dtoa.h:

13188

* kjs/error_object.cpp:

13189

* kjs/error_object.h:

13190

* kjs/function.cpp:

13191

* kjs/function.h:

13192

* kjs/function_object.cpp:

13193

* kjs/function_object.h:

13194

* kjs/grammar.y:

13195

* kjs/identifier.cpp:

13196

* kjs/identifier.h:

13197

* kjs/internal.cpp:

13198

* kjs/internal.h:

13199

* kjs/interpreter.cpp:

13200

* kjs/interpreter.h:

13201

* kjs/lexer.cpp:

13202

* kjs/lexer.h:

13203

* kjs/list.cpp:

13204

* kjs/list.h:

13205

* kjs/lookup.cpp:

13206

* kjs/lookup.h:

13207

* kjs/math_object.cpp:

13208

* kjs/math_object.h:

13209

* kjs/nodes.cpp:

13210

* kjs/nodes.h:

13211

* kjs/nodes2string.cpp:

13212

* kjs/number_object.cpp:

13213

* kjs/number_object.h:

13214

* kjs/object.cpp:

13215

* kjs/object.h:

13216

* kjs/object_object.cpp:

13217

* kjs/object_object.h:

13218

* kjs/operations.cpp:

13219

* kjs/operations.h:

13220

* kjs/property_map.cpp:

13221

* kjs/property_map.h:

13222

* kjs/property_slot.cpp:

13223

* kjs/property_slot.h:

13224

* kjs/reference.cpp:

13225

* kjs/reference.h:

13226

* kjs/reference_list.cpp:

13227

* kjs/reference_list.h:

13228

* kjs/regexp.cpp:

13229

* kjs/regexp.h:

13230

* kjs/regexp_object.cpp:

13231

* kjs/regexp_object.h:

13232

* kjs/scope_chain.cpp:

13233

* kjs/scope_chain.h:

13234

* kjs/simple_number.h:

13235

* kjs/string_object.cpp:

13236

* kjs/string_object.h:

13237

* kjs/testkjs.cpp:

13238

* kjs/types.h:

13239

* kjs/ustring.cpp:

13240

* kjs/ustring.h:

13241

* kjs/value.cpp:

13242

* kjs/value.h:

13243

* kxmlcore/AlwaysInline.h:

13244

* kxmlcore/ListRefPtr.h:

13245

* kxmlcore/PassRefPtr.h:

13246

* kxmlcore/RefPtr.h:

13247

13248

2006-01-23 Darin Adler <darin@apple.com>

13249

13250

Reviewed by John Sullivan.

13251

13252

- change needed for fix to http://bugs.webkit.org/show_bug.cgi?id=6617

13253

REGRESSION: Crash in cloneChildNodes when clicking element

13254

13255

* kxmlcore/PassRefPtr.h: Fix assignment operator from RefPtr of a different

13256

type by calling get() instead of going directly at m_ptr.

13257

* kxmlcore/RefPtr.h: Ditto.

13258

13259

- other changes

13260

13261

* JavaScriptCore.xcodeproj/project.pbxproj: Xcode decided to change this file.

13262

It's just a resorted list of keys in a dictionary.

13263

13264

* kjs/fpconst.cpp: Wrap this file in #if __APPLE__ since the alternate version

13265

in internal.cpp is in #if !__APPLE__. This file is to give us the "no init

13266

routine" property we want to have on OS X.

13267

13268

2006-01-22 Maciej Stachowiak <mjs@apple.com>

13269

13270

Reviewed by Darin.

13271

13272

- Set up Page class and invert Frame / WebCoreFrameBridge ownership

13273

http://bugs.webkit.org/show_bug.cgi?id=6577

13274

13275

* kjs/interpreter.h: make globalExec virtual so ScriptInterpreter can

13276

override it

13277

13278

2006-01-23 George Staikos <staikos@opensource.apple.com>

13279

13280

Reviewed by Maciej and Darin.

13281

13282

* kxmlcore/Assertions.h: This file only works with __APPLE__ right now

13283

* kjs/interpreter.cpp: ditto

13284

* kjs/simple_number.h: Add assert.h and remove from config.h

13285

* kjs/array_object.cpp: Use relative paths for kxmlcore includes

13286

* kjs/testkjs.cpp: Use relative paths for kxmlcore includes

13287

13288

2006-01-23 George Staikos <staikos@opensource.apple.com>

13289

13290

Reviewed by Maciej.

13291

13292

* kjs/config.h: unbreak preprocessor change

13293

13294

2006-01-23 George Staikos <staikos@opensource.apple.com>

13295

13296

Approved by Maciej and Darin.

13297

13298

* kjs/:

13299

* kxmlcore/:

13300

Update FSF address in license to make merging easier

13301

13302

2006-01-22 George Staikos <staikos@opensource.apple.com>

13303

13304

Reviewed by Maciej.

13305

13306

* kjs/collector.cpp: merge major speedup from KDE on Linux

13307

patch by Maks Orlovich, bug #6145

13308

Also unify cpu detection

13309

* kjs/config.h: define simpler CPU macros

13310

13311

2006-01-22 George Staikos <staikos@opensource.apple.com>

13312

13313

Reviewed by Maciej.

13314

13315

* kjs/collector.cpp: merge FreeBSD compile fix from KDE

13316

-> requires build magic for use

13317

13318

2006-01-21 George Staikos <staikos@opensource.apple.com>

13319

13320

Reviewed by Maciej.

13321

13322

* kjs/nodes2string.cpp

13323

* kjs/operations.h

13324

* kjs/debugger.h

13325

Fix pedantic compile with some gcc versions (Merge from KDE)

13326

13327

* kjs/create_hash_table:

13328

Fix build with Perl 5.8.0 (Merge from KDE)

13329

13330

2006-01-18 Darin Adler <darin@apple.com>

13331

13332

Reviewed by Hyatt.

13333

13334

- hash table fixes needed for my WebCore changes

13335

13336

* kxmlcore/HashTable.h: (KXMLCore::HashTableConstIterator::operator=):

13337

Added a missing return statement.

13338

13339

* kxmlcore/HashTraits.h: Fix traits so they work properly for classes where you

13340

can't instantiate with a 0 by using traits rather than ? : to select the default

13341

emtpy value of hash table keys.

13342

13343

- small cleanup of "runtime" code left over from recent JavaScript crash fix

13344

13345

* bindings/runtime_root.h:

13346

(KJS::Bindings::RootObject::RootObject): No explicit initialization of _imp needed

13347

since it's now a ProtectedPtr.

13348

(KJS::Bindings::RootObject::setRootObjectImp): Remove old code that relied on the

13349

fact that _imp was 0 and replaced with use of ProtectedPtr.

13350

(KJS::Bindings::RootObject::rootObjectImp): Updated since _imp is a ProtectedPtr.

13351

13352

2006-01-17 Darin Adler <darin@apple.com>

13353

13354

Reviewed by Anders.

13355

13356

- http://bugs.webkit.org/show_bug.cgi?id=6611

13357

add assertions to check correct use of hash table iterators

13358

13359

* kxmlcore/HashTable.h:

13360

(KXMLCore::addIterator): Added. Helper function that adds an iterator to the list

13361

maintained by the specified hash table.

13362

(KXMLCore::removeIterator): Added. Helper function that removes an iterator from

13363

the list maintained by the hash table it's in.

13364

(KXMLCore::HashTableConstIterator::HashTableConstIterator): Added a HashTable

13365

parameter, ignored when not debugging. Call addIterator.

13366

(KXMLCore::HashTableConstIterator::~HashTableConstIterator):

13367

(KXMLCore::HashTableConstIterator::operator=): Call removeIterator.

13368

(KXMLCore::HashTableConstIterator::operator*): Call checkValidity.

13369

(KXMLCore::HashTableConstIterator::operator->): Ditto.

13370

(KXMLCore::HashTableConstIterator::operator++): Ditto.

13371

(KXMLCore::HashTableConstIterator::operator==): Ditto.

13372

(KXMLCore::HashTableConstIterator::operator!=): Ditto.

13373

(KXMLCore::HashTableConstIterator::checkValidity): Checks that the hash table

13374

pointer is not 0 and if there are two iterators that both point at the same table.

13375

(KXMLCore::HashTableIterator::HashTableIterator): Changed to use the const iterator

13376

as an implementation detail, to avoid having two separate iterator implementations.

13377

(KXMLCore::HashTableIterator::operator*): Ditto.

13378

(KXMLCore::HashTableIterator::operator->): Ditto.

13379

(KXMLCore::HashTableIterator::operator++): Ditto.

13380

(KXMLCore::HashTableIterator::operator==): Ditto.

13381

(KXMLCore::HashTableIterator::operator!=): Ditto.

13382

(KXMLCore::HashTable::HashTable): Initialize pointer to head of iterators list.

13383

(KXMLCore::HashTable::~HashTable): Added call to invalidateIterators.

13384

(KXMLCore::HashTable::makeIterator): Pass this pointer.

13385

(KXMLCore::HashTable::makeConstIterator): Ditto.

13386

(KXMLCore::HashTable::insert): Call invalidateIterators, since this is a public

13387

entry point that modifies the hash table.

13388

(KXMLCore::HashTable::remove): Ditto.

13389

(KXMLCore::HashTable::clear): Ditto.

13390

(KXMLCore::HashTable::swap): Ditto.

13391

(KXMLCore::HashTable::invalidateIterators): Added. Walks the iterators list and

13392

clears out the table, next, and previous pointers in all of them, and then clears

13393

the head so we have an empty list.

13394

(KXMLCore::addIterator): Added. Adds the iterator the the linked list in the

13395

passed-in table, and points the iterator at the table.

13396

(KXMLCore::removeIterator): Added. Removes the iterator from the linked list in

13397

the passed-in table.

13398

13399

* kxmlcore/HashTraits.h: A bit of tweaking and formatting.

13400

13401

2006-01-17 Justin Garcia <justin.garcia@apple.com>

13402

13403

Reviewed by eric

13404

13405

Deployment builds now use -O2

13406

13407

* JavaScriptCore.xcodeproj/project.pbxproj:

13408

13409

2006-01-17 Darin Adler <darin@apple.com>

13410

13411

Reviewed by Anders.

13412

13413

- fix http://bugs.webkit.org/show_bug.cgi?id=6610

13414

change RefPtr so that it works when deref ends up deleting the RefPtr

13415

13416

* kxmlcore/PassRefPtr.h: Always set m_ptr before calling deref.

13417

* kxmlcore/RefPtr.h: Ditto.

13418

13419

2006-01-16 Geoffrey Garen <ggaren@apple.com>

13420

13421

Reviewed by darin.

13422

13423

- Fixed http://bugs.webkit.org/show_bug.cgi?id=6322

13424

DateProtoFuncImp::callAsFunction can crash due to lack of type checking

13425

13426

* kjs/date_object.cpp:

13427

(KJS::DateProtoFunc::callAsFunction): Type check calls to all methods.

13428

This matches section 15.9.5 in the spec.

13429

13430

2006-01-16 Tim Omernick <timo@apple.com>

13431

13432

Reviewed by John Sullivan.

13433

13434

JavaScriptCore part of <rdar://problem/4211707> NPAPI ref count behavior differs with Mozilla

13435

13436

* bindings/npruntime.cpp:

13437

(_NPN_ReleaseObject):

13438

Refactored part of this function out into _NPN_DeallocateObject.

13439

(_NPN_DeallocateObject):

13440

Forcibly deallocates the passed object, even if its refcount is

13441

greater than zero.

13442

13443

* bindings/npruntime_impl.h:

13444

Declared _NPN_DeallocateObject().

13445

13446

2006-01-16 Darin Adler <darin@apple.com>

13447

13448

Reviewed by Maciej.

13449

13450

- fix problem with ++, ==, and != on const iterators in

13451

HashMaps that are using the pointer specialization

13452

13453

* kxmlcore/HashMapPtrSpec.h:

13454

(KXMLCore::PointerHashConstIteratorAdapter::operator++): Change type to const_iterator.

13455

(KXMLCore::PointerHashConstIteratorAdapter::operator==): Ditto.

13456

(KXMLCore::PointerHashConstIteratorAdapter::operator!=): Ditto.

13457

13458

2006-01-15 Alexey Proskuryakov <ap@nypop.com>

13459

13460

Reviewed by Anders.

13461

13462

- fix http://bugs.webkit.org/show_bug.cgi?id=6561

13463

run-javascriptcore-tests doesn't work

13464

13465

* JavaScriptCore/tests/mozilla/Getopt/Mixed.pm:

13466

Changed revision number to 1.8 (broken by svn migration).

13467

13468

2006-01-14 David Kilzer <ddkilzer@kilzer.net>

13469

13470

Reviewed and landed by Anders.

13471

13472

* kjs/create_hash_table: Fixed comment typo.

13473

13474

2006-01-13 Maks Orlovich <maksim@kde.org>

13475

13476

Mostly merging work by Peter Kelly. Reviewed by Maciej, landed by ap.

13477

13478

- fix http://bugs.webkit.org/show_bug.cgi?id=6261

13479

Misc. array object fixes from KJS

13480

13481

* kjs/array_object.cpp: Don't treat 2^32-1 as a real array index property.

13482

(ArrayInstance::getOwnPropertySlot): Ditto.

13483

(ArrayInstance::deleteProperty): Ditto.

13484

(ArrayInstance::put): Ditto.

13485

(ArrayInstance::propList): Added a FIXME comment.

13486

(ArrayInstance::put): Throw exception on trying to set invalid array length.

13487

(ArrayProtoFunc::callAsFunction): Do not use a separator argument when doing toString/toLocalString.

13488

* kjs/array_object.h: Added MAX_ARRAY_INDEX.

13489

13490

2006-01-13 Darin Adler <darin@apple.com>

13491

13492

- Replaced tabs with spaces in source files that had less than 10 lines with tabs.

13493

- Set allow-tabs Subversion property in source files that have more than 10 lines with tabs.

13494

13495

2006-01-13 Anders Carlsson <andersca@mac.com>

13496

13497

Reviewed by Eric.

13498

13499

* kjs/create_hash_table:

13500

Use correct size variables.

13501

13502

2006-01-13 Anders Carlsson <andersca@mac.com>

13503

13504

Reviewed by Darin.

13505

13506

* kjs/create_hash_table:

13507

Don't create an empty entry array, instead add a entry with all fields

13508

set to null and set the hash table size to 1.

13509

13510

* kjs/lookup.cpp:

13511

(findEntry):

13512

Remove the hash table size check

13513

13514

2006-01-12 Anders Carlsson <andersca@mac.com>

13515

13516

Reviewed by Maciej.

13517

13518

- fix http://bugs.webkit.org/show_bug.cgi?id=6494

13519

Crash when assigning a new function to a DOMParser object

13520

13521

* JavaScriptCore.xcodeproj/project.pbxproj:

13522

Move lookup.cpp before lookup.h

13523

13524

* kjs/lookup.cpp:

13525

(findEntry):

13526

If the hash table is empty, return 0 early.

13527

13528

2006-01-12 George Staikos <staikos@kde.org>

13529

13530

Reviewed by Darin.

13531

13532

* kjs/interpreter.cpp:

13533

* kjs/testkjs.cpp:

13534

* kjs/interpreter.h:

13535

Add helper to interpreter to call the collector in order to facilitate

13536

visibility rules in KDE.

13537

13538

2006-01-12 George Staikos <staikos@kde.org>

13539

13540

Reviewed by Maciej.

13541

13542

* kjs/kjs.pro: Updates to build the whole thing on Linux at least.

13543

13544

* kxmlcore/HashTable.h: Add missing assert.h

13545

13546

2006-01-12 Darin Adler <darin@apple.com>

13547

13548

Reviewed by Geoff.

13549

13550

- fix http://bugs.webkit.org/show_bug.cgi?id=6505

13551

retire APPLE_CHANGES from JavaScriptCore

13552

13553

* JavaScriptCore.xcodeproj/project.pbxproj: Removed both

13554

APPLE_CHANGES and HAVE_CONFIG_H from all targets.

13555

13556

* README: Removed. This had obsolete information in it

13557

and it wasn't clear what to replace it with.

13558

13559

* kjs/collector.h: Removed an APPLE_CHANGES if around something

13560

that's not really platform-specific (although it does use a

13561

platform-specific API at the moment).

13562

* kjs/collector.cpp: Removed a mistaken comment.

13563

13564

* kjs/grammar.y:

13565

* kjs/internal.cpp:

13566

* kjs/object.h:

13567

* kjs/operations.cpp:

13568

* kjs/operations.h:

13569

* kjs/ustring.h:

13570

Use __APPLE__ instead of APPLE_CHANGES for code that should be

13571

used only on Mac OS X.

13572

13573

* kjs/interpreter.cpp: Removed APPLE_CHANGES ifdef around the include

13574

of the runtime.h header. Even though that header isn't needed at the

13575

moment on platforms other than Mac OS X, the conditional stuff should

13576

be in the header itself, not in this one client.

13577

13578

* kjs/math_object.cpp: (MathFuncImp::callAsFunction): Removed some

13579

code inside APPLE_CHANGES. I'm pretty sure this code isn't needed on

13580

any platform where pow is implemented corrrectly according to the IEEE

13581

standard. If it is needed on some, we can add it back with an appropriate

13582

#if for the platforms where it is needed.

13583

13584

2006-01-12 Justin Haygood <justin@xiondigital.net>

13585

13586

Reviewed, tweaked, and landed by Darin.

13587

13588

- fix http://bugs.webkit.org/show_bug.cgi?id=6416

13589

lexer.cpp, grammar.y protect include of config.h with "HAVE_CONFIG_H"

13590

13591

* kjs/dtoa.cpp: Removed HAVE_CONFIG_H, changed config.h to use

13592

quotes instead of angle brackets. Moved dtoa.h include to the top.

13593

Changed system header includes to use angle brackets instead of quotes.

13594

* kjs/grammar.y: Removed HAVE_CONFIG_H, changed config.h to use

13595

quotes instead of angle brackets.

13596

* kjs/lexer.cpp: Removed HAVE_CONFIG_H, changed config.h to use

13597

quotes instead of angle brackets. Moved lexer.h include to the top.

13598

* kjs/ustring.cpp: Removed HAVE_CONFIG_H, changed config.h to use

13599

quotes instead of angle brackets. Moved ustring.h include to the top.

13600

13601

2006-01-12 George Staikos <staikos@kde.org>

13602

13603

Reviewed by Maciej

13604

13605

- Import initial QMake file. Doesn't fully work yet.

13606

13607

2006-01-11 Ricci Adams <ricciadams@apple.com>

13608

13609

Reviewed by Maciej and Darin, landed by Darin.

13610

13611

- fix http://bugs.webkit.org/show_bug.cgi?id=5939

13612

final comma in javascript object prevents parsing

13613

13614

* kjs/grammar.y: Added rule to allow trailing comma in

13615

object construction.

13616

13617

2006-01-11 Ricci Adams <ricciadams@apple.com>

13618

13619

Reviewed by Geoff, landed by Darin.

13620

13621

- fix http://bugs.webkit.org/show_bug.cgi?id=5308

13622

Number.toFixed doesn't include leading 0

13623

13624

* kjs/number_object.cpp: (NumberProtoFunc::callAsFunction):

13625

Fixed a "<" that should have been a "<=".

13626

13627

2006-01-11 Ricci Adams <ricciadams@apple.com>

13628

13629

Reviewed by Geoff, landed by Darin.

13630

13631

- fix http://bugs.webkit.org/show_bug.cgi?id=5307

13632

Number.toFixed doesn't round 0.5 up

13633

13634

* kjs/number_object.cpp: (NumberProtoFunc::callAsFunction):

13635

Fixed a ">" that should have been a ">=".

13636

13637

2006-01-11 Justin Haygood <justin@xiondigital.net>

13638

13639

Reviewed and landed by Darin.

13640

13641

- fix http://bugs.webkit.org/show_bug.cgi?id=6486

13642

JavaScriptCore should use system malloc on Windows

13643

13644

* kjs/config.h: Add USE_SYSTEM_MALLOC to the Win32 section.

13645

13646

2006-01-10 Darin Adler <darin@apple.com>

13647

13648

* Makefile: Took out unneeded "export" line.

13649

* <many-files>: Changed a lot of flags (cleared bogus executable bits, set

13650

MIME types, other small corrections).

13651

13652

2006-01-09 Darin Adler <darin@apple.com>

13653

13654

* Makefile.am: Removed.

13655

13656

2006-01-07 Anders Carlsson <andersca@mac.com>

13657

13658

Reviewed by Maciej.

13659

13660

- fix http://bugs.webkit.org/show_bug.cgi?id=6373

13661

REGRESSION: JavaScript hang when comparing large array to null

13662

13663

* kjs/object.h:

13664

(KJS::JSObject::isEqualToNull):

13665

Add new function which returns true if an object should be treated as null when

13666

doing comparisons.

13667

13668

* kjs/operations.cpp:

13669

(KJS::equal):

13670

Use isEqualToNull.

13671

13672

2006-01-07 Alexey Proskuryakov <ap@nypop.com>

13673

13674

Reviewed by Maciej.

13675

13676

- Fix WebCore development build

13677

http://bugs.webkit.org/show_bug.cgi?id=6408

13678

13679

* kxmlcore/Assertions.h: Use __VA_ARGS__ in variadic macros.

13680

13681

2006-01-06 Maciej Stachowiak <mjs@apple.com>

13682

13683

Reviewed by Darin.

13684

13685

- miscellaneous changes for 4% speedup on the JavaScript iBench

13686

http://bugs.webkit.org/show_bug.cgi?id=6396

13687

13688

Changes mostly thanks to Maks Orlovich, tweaked a little by me.

13689

13690

* kjs/create_hash_table: Use the same hash as the one used by Identifier.

13691

* kjs/function.cpp:

13692

(KJS::FunctionImp::processParameters): Use the new List::copyFrom

13693

(KJS::ActivationImp::ActivationImp): track variable while iterating

13694

* kjs/internal.cpp:

13695

(KJS::StringImp::toObject): create StringInstance directly

13696

* kjs/list.cpp:

13697

(KJS::List::copy): implement in terms of copyFrom

13698

(KJS::List::copyFrom): more efficient way to copy in another list

13699

* kjs/list.h:

13700

* kjs/lookup.cpp:

13701

(keysMatch): updated to work with identifier hash

13702

(findEntry): ditto

13703

(Lookup::findEntry): ditto

13704

(Lookup::find): ditto

13705

* kjs/lookup.h:

13706

13707

2006-01-06 Maciej Stachowiak <mjs@apple.com>

13708

13709

- fix development build failure from the previous checkin

13710

13711

* kjs/function.cpp:

13712

(KJS::ActivationImp::put): Use prototype() accessor in assert.

13713

13714

2006-01-05 Maciej Stachowiak <mjs@apple.com>

13715

13716

Reviewed by Eric.

13717

13718

- fix remaining performance regression from Getter/Setter change

13719

http://bugs.webkit.org/show_bug.cgi?id=6249

13720

13721

- Activation objects should not have __proto__ property

13722

http://bugs.webkit.org/show_bug.cgi?id=6395

13723

13724

* kjs/function.cpp:

13725

(KJS::ActivationImp::getOwnPropertySlot): Implement directly, thus

13726

skipping getter/setter handling and __proto__ handling, as well

13727

as inlining needed superclass stuff.

13728

(KJS::ActivationImp::put): Implement directly, skipping getter/setter,

13729

__proto__, and do canPut directly in PropertyMap::put since there's no

13730

static property table either.

13731

* kjs/function.h:

13732

* kjs/property_map.cpp:

13733

(KJS::PropertyMap::put): Allow optionally inlining canPut check.

13734

* kjs/property_map.h:

13735

13736

2006-01-04 Geoffrey Garen <ggaren@apple.com>

13737

13738

Patch by kimmo.t.kinnunen@nokia.com, reviewed by darin, tweaked by me.

13739

13740

- Fixed http://bugs.webkit.org/show_bug.cgi?id=4921

13741

\u escape sequences in JavaScript identifiers

13742

13743

* kjs/function_object.cpp:

13744

(FunctionObjectImp::construct):

13745

* kjs/lexer.cpp:

13746

(Lexer::shift):

13747

(Lexer::lex):

13748

(Lexer::isWhiteSpace):

13749

(Lexer::isLineTerminator):

13750

(Lexer::isIdentStart):

13751

(Lexer::isIdentPart):

13752

(isDecimalDigit):

13753

(Lexer::scanRegExp):

13754

* kjs/lexer.h:

13755

(KJS::Lexer::):

13756

13757

* tests/mozilla/expected.html: Updated test results.

13758

13759

2005-12-30 Maciej Stachowiak <mjs@apple.com>

13760

13761

No review, just test result update.

13762

13763

* tests/mozilla/expected.html: Updated for newly passing test from recent fixes.

13764

13765

2005-12-30 Anders Carlsson <andersca@mac.com>

13766

13767

Reviewed by Maciej.

13768

13769

- Fix http://bugs.webkit.org/show_bug.cgi?id=6298

13770

Getter setter test is failing

13771

13772

* kjs/object.cpp:

13773

(KJS::JSObject::put):

13774

Rework the getter setter part. We now walk the prototype chain, checking for

13775

getter/setter properties and only take the slow path if any are found.

13776

13777

2005-12-30 Maks Orlovich <maksim@kde.org>

13778

13779

Reviewed and committed by Maciej.

13780

13781

- Handle negative, FP numbers with non-10 radix in toString

13782

http://bugs.webkit.org/show_bug.cgi?id=6259

13783

13784

(Merged from KJS, original work by Harri Porten)

13785

13786

* kjs/number_object.cpp:

13787

(NumberProtoFunc::callAsFunction): rewrote Number.toString(radix) to work with

13788

negative numbers, floating point and very large numbers.

13789

13790

2005-12-29 Geoffrey Garen <ggaren@apple.com>

13791

13792

Patch by Maks Orlovich, reviewed and landed by me.

13793

13794

- http://bugs.webkit.org/show_bug.cgi?id=6267

13795

Fix Number.prototype.toFixed/toExponential(undefined)

13796

13797

* kjs/number_object.cpp:

13798

(NumberProtoFunc::callAsFunction):

13799

13800

2005-12-29 Geoffrey Garen <ggaren@apple.com>

13801

13802

Patch by Maks Orlovich, Reviewed and landed by me.

13803

13804

- http://bugs.webkit.org/show_bug.cgi?id=6266

13805

Minor object naming updates (to match Mozilla, KJS)

13806

13807

* kjs/number_object.cpp:

13808

* kjs/regexp_object.cpp:

13809

13810

2005-12-29 Geoffrey Garen <ggaren@apple.com>

13811

13812

Patch by Maks Orlovich, reviewed by mjs.

13813

13814

This has 2 very minor fixes, covered by KJS testsuite:

13815

1. Enumerates string indices in property list (with the same bug as array

13816

object has in corresponding code). This is a mozilla emulation thing.

13817

2. Permits properties with integer names in prototypes to be found

13818

13819

* kjs/string_object.cpp:

13820

(StringInstance::getOwnPropertySlot):

13821

(StringInstanceImp::propList):

13822

* kjs/string_object.h:

13823

13824

2005-12-26 Geoffrey Garen <ggaren@apple.com>

13825

13826

Reviewed by mjs.

13827

13828

- Fixed <rdar://problem/4364705> run-javascriptcore-tests crashes in

13829

KJS::BlockNode::deref

13830

AKA

13831

http://bugs.webkit.org/show_bug.cgi?id=6233

13832

Reproducible stack-overflow crash in ~RefPtr<T> due to RefPtr<T> use in

13833

linked lists

13834

13835

This patch does four things:

13836

(1) Standardizes all our linked list nodes to use "next" as their next

13837

pointers.

13838

(2) Creates the ListRefPtr<T> class, a subclass of RefPtr<T> specialized

13839

to iteratively deref "next" pointers.

13840

(3) Standardizes our linked list nodes to use ListRefPtr<T> and

13841

implement the releaseNext() function used by ~ListRefPtr<T>().

13842

(4) Adds to RefPtr<T> the release() method used by releaseNext().

13843

13844

- Modified existing mozilla test to ensure it would make deployment

13845

builds crash as well.

13846

13847

* JavaScriptCore.xcodeproj/project.pbxproj:

13848

* kjs/nodes.cpp:

13849

(ElementNode::evaluate):

13850

(PropertyListNode::evaluate):

13851

(ArgumentListNode::evaluateList):

13852

(StatListNode::StatListNode):

13853

(StatListNode::execute):

13854

(StatListNode::processVarDecls):

13855

(VarDeclListNode::evaluate):

13856

(VarDeclListNode::processVarDecls):

13857

(VarStatementNode::execute):

13858

(VarStatementNode::processVarDecls):

13859

(BlockNode::BlockNode):

13860

(CaseClauseNode::evalStatements):

13861

(CaseClauseNode::processVarDecls):

13862

(ClauseListNode::processVarDecls):

13863

(CaseBlockNode::CaseBlockNode):

13864

(CaseBlockNode::evalBlock):

13865

(SourceElementsNode::SourceElementsNode):

13866

(SourceElementsNode::execute):

13867

(SourceElementsNode::processFuncDecl):

13868

(SourceElementsNode::processVarDecls):

13869

* kjs/nodes.h:

13870

(KJS::ElementNode::ElementNode):

13871

(KJS::ElementNode::releaseNext):

13872

(KJS::ArrayNode::ArrayNode):

13873

(KJS::PropertyListNode::PropertyListNode):

13874

(KJS::PropertyListNode::releaseNext):

13875

(KJS::ObjectLiteralNode::ObjectLiteralNode):

13876

(KJS::ArgumentListNode::ArgumentListNode):

13877

(KJS::ArgumentListNode::releaseNext):

13878

(KJS::ArgumentsNode::ArgumentsNode):

13879

(KJS::StatListNode::releaseNext):

13880

(KJS::VarDeclListNode::VarDeclListNode):

13881

(KJS::VarDeclListNode::releaseNext):

13882

(KJS::VarStatementNode::VarStatementNode):

13883

(KJS::ForNode::ForNode):

13884

(KJS::CaseClauseNode::CaseClauseNode):

13885

(KJS::ClauseListNode::ClauseListNode):

13886

(KJS::ClauseListNode::getClause):

13887

(KJS::ClauseListNode::getNext):

13888

(KJS::ClauseListNode::releaseNext):

13889

(KJS::ParameterNode::ParameterNode):

13890

(KJS::ParameterNode::releaseNext):

13891

(KJS::SourceElementsNode::releaseNext):

13892

* kjs/nodes2string.cpp:

13893

(ElementNode::streamTo):

13894

(PropertyListNode::streamTo):

13895

(ArgumentListNode::streamTo):

13896

(StatListNode::streamTo):

13897

(VarDeclListNode::streamTo):

13898

(VarStatementNode::streamTo):

13899

(CaseClauseNode::streamTo):

13900

(ClauseListNode::streamTo):

13901

(CaseBlockNode::streamTo):

13902

(SourceElementsNode::streamTo):

13903

* kxmlcore/ListRefPtr.h: Added.

13904

(KXMLCore::ListRefPtr::ListRefPtr):

13905

(KXMLCore::ListRefPtr::~ListRefPtr):

13906

(KXMLCore::ListRefPtr::operator=):

13907

* kxmlcore/RefPtr.h:

13908

(KXMLCore::RefPtr::release):

13909

13910

2005-12-29 Geoffrey Garen <ggaren@apple.com>

13911

13912

Reviewed by mjs.

13913

13914

- Fixed http://bugs.webkit.org/show_bug.cgi?id=4026

13915

Math.random() not seeded.

13916

13917

Added call to sranddev() -- it executes the first time a process

13918

calls Math.random().

13919

13920

* kjs/math_object.cpp:

13921

(MathFuncImp::callAsFunction):

13922

13923

2005-12-29 Geoffrey Garen <ggaren@apple.com>

13924

13925

Reviewed by darin.

13926

13927

- Fixed http://bugs.webkit.org/show_bug.cgi?id=6265

13928

Name change regression: Java doesn't know what JavaJSObject is

13929

13930

Changed strings passed to Java back to original "JSObject".

13931

13932

* bindings/jni/jni_jsobject.cpp:

13933

(JavaJSObject::convertValueToJObject):

13934

(JavaJSObject::convertJObjectToValue):

13935

13936

2005-12-28 Anders Carlsson <andersca@mac.com>

13937

13938

Reviewed by Maciej.

13939

13940

- The JSC part of http://bugs.webkit.org/show_bug.cgi?id=6268

13941

Add undetectable document.all

13942

13943

* kjs/operations.cpp:

13944

(KJS::equal):

13945

When comparing an object with null or undefined, call toPrimitive with

13946

NullType as the preferred type.

13947

13948

2005-12-27 Anders Carlsson <andersca@mac.com>

13949

13950

Reviewed by Darin.

13951

13952

* kjs/array_object.cpp:

13953

(ArrayProtoFunc::callAsFunction):

13954

Implement filter and map. Also, make the existing

13955

array iteration functions not invoke the callback for

13956

non-existing properties, just as Mozilla does now.

13957

13958

* kjs/array_object.h:

13959

(KJS::ArrayProtoFunc::):

13960

Add filter and map.

13961

13962

* tests/mozilla/expected.html:

13963

Update, two 1.6 tests now pass.

13964

13965

2005-12-27 Maciej Stachowiak <mjs@apple.com>

13966

13967

- updated test results for new JS 1.6 tests

13968

13969

* tests/mozilla/expected.html:

13970

13971

2005-12-27 Anders Carlsson <andersca@mac.com>

13972

13973

Reviewed by Maciej.

13974

13975

Add Mozilla JS 1.6 tests.

13976

13977

* tests/mozilla/js1_6/Array/browser.js: Added.

13978

* tests/mozilla/js1_6/Array/regress-290592.js: Added.

13979

* tests/mozilla/js1_6/Array/regress-304828.js: Added.

13980

* tests/mozilla/js1_6/Array/regress-305002.js: Added.

13981

* tests/mozilla/js1_6/Array/regress-310425-01.js: Added.

13982

* tests/mozilla/js1_6/Array/regress-310425-02.js: Added.

13983

* tests/mozilla/js1_6/Array/regress-320887.js: Added.

13984

* tests/mozilla/js1_6/Array/shell.js: Added.

13985

* tests/mozilla/js1_6/README: Added.

13986

* tests/mozilla/js1_6/Regress/browser.js: Added.

13987

* tests/mozilla/js1_6/Regress/regress-301574.js: Added.

13988

* tests/mozilla/js1_6/Regress/regress-309242.js: Added.

13989

* tests/mozilla/js1_6/Regress/regress-311157-01.js: Added.

13990

* tests/mozilla/js1_6/Regress/regress-311157-02.js: Added.

13991

* tests/mozilla/js1_6/Regress/regress-314887.js: Added.

13992

* tests/mozilla/js1_6/Regress/regress-320172.js: Added.

13993

* tests/mozilla/js1_6/Regress/shell.js: Added.

13994

* tests/mozilla/js1_6/String/browser.js: Added.

13995

* tests/mozilla/js1_6/String/regress-306591.js: Added.

13996

* tests/mozilla/js1_6/String/shell.js: Added.

13997

* tests/mozilla/js1_6/browser.js: Added.

13998

* tests/mozilla/js1_6/shell.js: Added.

13999

* tests/mozilla/js1_6/template.js: Added.

14000

14001

2005-12-27 Maks Orlovich <maksim@kde.org>

14002

14003

Reviewed and landed by Maciej.

14004

14005

- fixed 6234: Can delete array index property incorrectly.

14006

http://bugs.webkit.org/show_bug.cgi?id=6234

14007

14008

* kjs/array_object.cpp:

14009

(ArrayInstance::deleteProperty): use toArrayIndex instead of toUInt32 when

14010

looking for array properties.

14011

14012

2005-12-27 Anders Carlsson <andersca@mac.com>

14013

14014

Reviewed by Maciej.

14015

14016

* kjs/object.cpp:

14017

(KJS::JSObject::defineSetter):

14018

Remove duplicate call to putDirect.

14019

14020

2005-12-26 Maciej Stachowiak <mjs@apple.com>

14021

14022

Reviewed by Darin and Geoff.

14023

14024

Changes by me and Anders.

14025

14026

- mostly fixed REGRESSION: 5-10% performance regression on JS iBench from getter/setter change

14027

http://bugs.webkit.org/show_bug.cgi?id=6083

14028

14029

- also fixed some warnings reported by -Winline

14030

14031

* JavaScriptCorePrefix.h: Move new and delete definitions higher so there

14032

aren't conflicts with use in standard C++ headers

14033

* kjs/object.cpp:

14034

(KJS::throwSetterError): Moved this piece of put into a seprate function

14035

to avoid the PIC branch.

14036

(KJS::JSObject::put): Use hasGetterSetterProperties to avoid expensive stuff

14037

when not needed. Also use GetterSetter properties attribute.

14038

(KJS::JSObject::deleteProperty): Recompute whether any properties are getter/setter

14039

properties any more, if this one was one.

14040

(KJS::JSObject::defineGetter): Let the PropertyMap know that it has getter/setter

14041

properties now (and use the new attribute).

14042

(KJS::JSObject::defineSetter): Ditto.

14043

(KJS::JSObject::fillGetterPropertySlot): Out-of-line helper for getOwnPropertySlot,

14044

to avoid global variable access in the hot code path.

14045

* kjs/object.h:

14046

(KJS::): Added GetterSetter attribute.

14047

(KJS::JSCell::isObject): Moved lower to be after inline methods it uses.

14048

(KJS::JSValue::isObject): ditto

14049

(KJS::JSObject::getOwnPropertySlot): try to avoid impact of getters and setters

14050

as much as possible in the case where they are not being used

14051

* kjs/property_map.cpp:

14052

(KJS::PropertyMap::containsGettersOrSetters): New method to help with this

14053

* kjs/property_map.h:

14054

(KJS::PropertyMap::hasGetterSetterProperties): Ditto

14055

(KJS::PropertyMap::setHasGetterSetterProperties): Ditto

14056

(KJS::PropertyMap::PropertyMap): Added a crazy hack to store the

14057

global "has getter/setter properties" flag in the property map

14058

single entry, to avoid making objects any bigger.

14059

* kjs/value.h: Moved some things to object.h to make -Winline happier

14060

14061

2005-12-24 Maciej Stachowiak <mjs@apple.com>

14062

14063

Reviewed by Eric and Dave Hyatt.

14064

14065

- make even const PassRefPtrs give transfer of ownership semantics

14066

http://bugs.webkit.org/show_bug.cgi?id=6238

14067

14068

This is a somewhat cheesy change. Having to use PassRefPtr_Ref creates ambiguities

14069

in assignment and copy construction. And this makes life way easier and removes

14070

the need for pass(). It is not really correct, but we pretty much never need a real

14071

const PassRefPtr, and this takes care of things for PassRefPtr temporaries.

14072

14073

* kjs/identifier.cpp:

14074

(KJS::Identifier::add): No more need for pass()

14075

* kjs/property_map.cpp:

14076

(KJS::PropertyMap::addSparseArrayPropertiesToReferenceList): No more need for pass()

14077

* kjs/ustring.cpp:

14078

(KJS::UString::Rep::create): Use adoptRef

14079

(KJS::UString::UString): No more need for pass

14080

(KJS::UString::append): No more need for pass

14081

(KJS::UString::substr): No more need for pass

14082

* kxmlcore/PassRefPtr.h: made m_ptr mutable (ugh)

14083

(KXMLCore::PassRefPtr::PassRefPtr): Take a const PassRefPtr reference

14084

(KXMLCore::PassRefPtr::release): Made this a const method (ugh)

14085

(KXMLCore::PassRefPtr::operator=): clean up appropriately

14086

(KXMLCore::adoptRef): Added this to use instead of PassRefPtr<T>::adopt, I think

14087

it makes the behavior more clear and it is less verbose.

14088

(KXMLCore::static_pointer_cast): use adoptRef

14089

(KXMLCore::const_pointer_cast): use adoptRef

14090

* kxmlcore/RefPtr.h:

14091

(KXMLCore::RefPtr::RefPtr): take const PassRefPtr&

14092

(KXMLCore::PassRefPtr::operator=): take const PassRefPtr&

14093

14094

2005-12-25 Eric Seidel <eseidel@apple.com>

14095

14096

Reviewed by mjs.

14097

14098

Unbreak HashTableConstIterator++ by returning const_iterator

14099

14100

* kxmlcore/HashTable.h:

14101

(KXMLCore::HashTableConstIterator::operator++): use const_iterator

14102

14103

2005-12-25 Eric Seidel <eseidel@apple.com>

14104

14105

Reviewed by mjs.

14106

14107

Un-break HashTable copy constructor.

14108

14109

* kxmlcore/HashTable.h:

14110

(KXMLCore::::HashTable): use const_iterator instead

14111

14112

2005-12-23 Maciej Stachowiak <mjs@apple.com>

14113

14114

Reviewed by Eric.

14115

14116

- fixed "HashMap does not work with const pointer keys or values"

14117

http://bugs.webkit.org/show_bug.cgi?id=6222

14118

14119

* kxmlcore/HashMapPtrSpec.h:

14120

(KXMLCore::HashMap): In all methods, explicitly cast all pointers

14121

to void * before passing to internal implementation. Use C-style

14122

casts instead of new-style casts, because the real solution would

14123

require a combo of reinterpret_cast anc const_cast.

14124

14125

14126

2005-12-23 Maciej Stachowiak <mjs@apple.com>

14127

14128

- this time for sure

14129

14130

* kxmlcore/RefPtr.h:

14131

(KXMLCore::::swap):

14132

14133

2005-12-22 Maciej Stachowiak <mjs@apple.com>

14134

14135

- fix build problem from last commit.

14136

14137

* kxmlcore/RefPtr.h:

14138

(KXMLCore::::swap):

14139

14140

2005-12-21 Maciej Stachowiak <mjs@apple.com>

14141

14142

Reviewed by Darin.

14143

14144

- Make HashMap/HashSet support non-POD types

14145

http://bugs.webkit.org/show_bug.cgi?id=5332

14146

14147

The changes for support are relatively simple, but I also made extensive changes to

14148

avoid copying, so that there isn't refcount thrash when you put RefPtrs into a HashMap.

14149

14150

* kxmlcore/HashTable.h:

14151

(KXMLCore::swap): specialize swap for pairs, to swap elements individually,

14152

so that excess copies can be avoided.

14153

(KXMLCore::Mover::move): Template function to either copy or swap, used

14154

when transferring elements from old table to new.

14155

(KXMLCore::IdentityHashTranslator::hash): The old "converting lookup" templates

14156

that took two or three function parameters now take a class parameter, this is

14157

the class used to do a normal lookup.

14158

(KXMLCore::IdentityHashTranslator::equal): Ditto.

14159

(KXMLCore::IdentityHashTranslator::translate): Ditto. Translate now takes a reference

14160

to write into instead of returning a value to avoid redundant copies.

14161

(KXMLCore::HashTable::~HashTable): Use deallocateTable instead of freeing directly.

14162

(KXMLCore::HashTable::insert): Based on HashTranslator now instead of separate

14163

functions. Added a FIXME about a remaining rare excess copy.

14164

(KXMLCore::HashTable::isEmptyBucket): Use KeyTraits directly instead of unwrapping

14165

the key from Traits, to avoid creating and destroying pair, which copies.

14166

(KXMLCore::HashTable::isDeletedBucket): ditto

14167

(KXMLCore::HashTable::lookup): Use HashTranslator now instead of separate functions.

14168

(KXMLCore::HashTable::initializeBucket): Renamed from emptyBucket. Use placement new to

14169

work right for non-POD types.

14170

(KXMLCore::HashTable::deleteBucket): Use assignDeleted to avoid excess copies.

14171

(KXMLCore::HashTable::reinsert): use Mover template to copy or swap as appropriate

14172

(KXMLCore::HashTable::allocateTable): Initialize every bucket if calloc won't do.

14173

(KXMLCore::HashTable::deallocateTable): Destruct every bucket if needed.

14174

(KXMLCore::HashTable::rehash): Avoid copy before reinserting, so that swap can do its magic.

14175

(KXMLCore::HashTable::clear): use deallocateTable instead of freeing directly.

14176

(KXMLCore::HashTable::HashTable): be more dumb when copying to ensure that non-POD types

14177

work right

14178

* kxmlcore/HashFunctions.h:

14179

(KXMLCore::PointerHash): Specialize PointerHash for RefPtr

14180

* kxmlcore/HashMap.h:

14181

(KXMLCore::extractFirst): Return a reference not a full object to avoid

14182

copies.

14183

(KXMLCore::HashMapTranslator::hash): Use a special translator for insertion

14184

to defer making the pair as long as possible, thus avoiding needless copies.

14185

(KXMLCore::HashMapTranslator::equal): ditto

14186

(KXMLCore::HashMapTranslator::translate): ditto

14187

(KXMLCore::::inlineAdd): Shared by set and add to insert using HashMapTranslator

14188

(KXMLCore::::set): Use inlineAdd

14189

(KXMLCore::::add): Use inlineAdd

14190

* kxmlcore/HashMapPtrSpec.h:

14191

(KXMLCore::): Pass KeyTraits along

14192

* kxmlcore/HashSet.h:

14193

(KXMLCore::identityExtract): Return a reference not a full object to avoid copies.

14194

(KXMLCore::HashSetTranslatorAdapter::hash): Redo adapter stuff to work with

14195

the new HashTranslator approach.

14196

(KXMLCore::HashSetTranslatorAdapter::equal): ditto

14197

(KXMLCore::HashSetTranslatorAdapter::translate): ditto

14198

(KXMLCore::::insert): ditto

14199

* kxmlcore/HashTraits.h:

14200

(KXMLCore::GenericHashTraits): This is intended be used as a base class for

14201

customized traits: sensible defaults.

14202

(KXMLCore::): Use it a bunch

14203

(KXMLCore::assignDeleted): template function to allow pairs to be assigned the

14204

deleted value w/o excess copies.

14205

(KXMLCore::PairHashTraits::emptyValue): Updated

14206

(KXMLCore::PairHashTraits::deletedValue): Updated

14207

(KXMLCore::PairHashTraits::assignDeletedValue): part of assignDeleted hack

14208

(KXMLCore::DeletedValueAssigner::assignDeletedValue): Use template magic

14209

to either use use deletedValue or assignDeletedValue for the cases where we care.

14210

* kxmlcore/RefPtr.h:

14211

(KXMLCore::RefPtr::swap): Added swap method.

14212

(KXMLCore::swap): Added swap free function.

14213

* kjs/identifier.cpp:

14214

(KJS::CStringTranslator::hash): Use new HashTranslator class approach to

14215

alternate type based insertion.

14216

(KJS::CStringTranslator::equal): ditto

14217

(KJS::CStringTranslator::translate): ditto

14218

(KJS::Identifier::add): ditto

14219

(KJS::UCharBufferTranslator::hash): ditto

14220

(KJS::UCharBufferTranslator::equal): ditto

14221

(KJS::UCharBufferTranslator::translate): ditto

14222

14223

- irrelevant change:

14224

14225

* kjs/array_object.cpp:

14226

(ArrayProtoFunc::callAsFunction): Removed a stray space.

14227

14228

2005-12-22 Anders Carlsson <andersca@mac.com>

14229

14230

Reviewed by Eric and Darin.

14231

14232

- fix http://bugs.webkit.org/show_bug.cgi?id=6196

14233

Would like to be able to define prototypes in headers

14234

14235

* kjs/lookup.h:

14236

Move ClassName from KJS_DECLARE_PROTOTYPE to KJS_IMPLEMENT_PROTOTYPE.

14237

Also, namespace all macros by prefixing them with KJS_.

14238

14239

2005-12-22 Darin Adler <darin@apple.com>

14240

14241

Reviewed by Maciej.

14242

14243

- fix http://bugs.webkit.org/show_bug.cgi?id=6191

14244

RefPtr/PassRefPtr have a leak issue, operator== issues

14245

14246

* kxmlcore/PassRefPtr.h:

14247

(KXMLCore::PassRefPtr::PassRefPtr): Remove non-template constructor that takes RefPtr

14248

since the constructor template that takes RefPtr should be sufficient. Add a constructor

14249

template that takes PassRefPtr&.

14250

(KXMLCore::PassRefPtr::adopt): Use PassRefPtr_Ref to avoid setting pointer first to

14251

0 and then to the pointer.

14252

(KXMLCore::PassRefPtr::operator=): Added template versions that take PassRefPtr& and

14253

RefPtr parameters.

14254

(KXMLCore::PassRefPtr::operator PassRefPtr<U>): Changed to fix leak -- old version

14255

would release and then ref.

14256

(KXMLCore::operator==): Make templates have two parameters so you can mix types.

14257

Also remove unneeded const in raw pointer versions.

14258

(KXMLCore::operator!=): Ditto.

14259

14260

* kxmlcore/RefPtr.h:

14261

(KXMLCore::RefPtr::RefPtr): Add constructor template that takes PassRefPtr.

14262

(KXMLCore::RefPtr::operator=): Add assignment operator templates that take

14263

RefPtr and PassRefPtr.

14264

(KXMLCore::operator==): Make templates have two parameters so you can mix types.

14265

Also remove unneeded const in raw pointer versions.

14266

(KXMLCore::operator!=): Ditto.

14267

14268

2005-12-21 Timothy Hatcher <timothy@apple.com>

14269

14270

* JavaScriptCore.xcodeproj/project.pbxproj:

14271

Set tab width to 8, indent width to 4 and uses tabs to false per file.

14272

14273

2005-12-21 Geoffrey Garen <ggaren@apple.com>

14274

14275

Reviewed by Darin.

14276

14277

Removed evil hack for determining if a type is an integer, replaced

14278

with template metaprogramming.

14279

14280

* JavaScriptCore.xcodeproj/project.pbxproj: Set tab size to 2 for

14281

testkjs.cpp

14282

* kjs/testkjs.cpp:

14283

(main): Inserted asserts to test IsInteger. FIXME: Move these to

14284

KXMLCore unit tests directory when we create one.

14285

* kxmlcore/HashTraits.h:

14286

(KXMLCore::): Added IsInteger class for querying types.

14287

14288

2005-12-20 Maciej Stachowiak <mjs@apple.com>

14289

14290

Reviewed by Darin.

14291

14292

- made ALWAYS_INLINE declare things inline as well as __attribute__((always_inline))

14293

http://bugs.webkit.org/show_bug.cgi?id=6159

14294

14295

* kxmlcore/AlwaysInline.h:

14296

14297

2005-12-19 Maciej Stachowiak <mjs@apple.com>

14298

14299

Reviewed by Darin.

14300

14301

- fixed a leak in the assignment operator from PassRefPtr to RefPtr

14302

http://bugs.webkit.org/show_bug.cgi?id=6158

14303

14304

* kxmlcore/RefPtr.h:

14305

(KXMLCore::RefPtr::operator=):

14306

14307

- fix problem with PassRefPtr that darin spotted - it lacked a copy constructor

14308

and therefore was using the default one, which can lead to excess derefs

14309

14310

I fixed this by adding a copy constructor from non-const

14311

reference, and by adding a template pass() function that you have

14312

to use when raw pointer or RefPtr are passed where PassRefPtr is

14313

expected.

14314

14315

* kjs/identifier.cpp:

14316

(KJS::Identifier::add): Changed to have PassRefPtr return type and

14317

pass() the results.

14318

* kjs/identifier.h:

14319

* kjs/property_map.cpp:

14320

(KJS::PropertyMap::addSparseArrayPropertiesToReferenceList): Use pass()

14321

where required.

14322

* kjs/ustring.cpp:

14323

(KJS::UString::UString): Use pass() as needed.

14324

(KJS::UString::append): ditto

14325

(KJS::UString::substr): ditto

14326

* kjs/ustring.h:

14327

(KJS::UString::UString): Use initializer instead of assignment

14328

* kxmlcore/PassRefPtr.h:

14329

(KXMLCore::PassRefPtr::PassRefPtr): Added copy constructor

14330

(KXMLCore::pass): new template function to make it convenient to pass

14331

a PassRefPtr

14332

14333

2005-12-19 Geoffrey Garen <ggaren@apple.com>

14334

14335

Reviewed by Maciej.

14336

14337

Fixed <rdar://problem/4370397> Missing return statement in

14338

JSMethodNameToObjcMethodName.

14339

14340

JSMethodNameToObjcMethodName had a check for a name being too long, but

14341

the check was missing a return statement.

14342

14343

A lot of this code was confusing and some of it was wrong, so I fixed

14344

it up, added some asserts to catch this type of bug in the future,

14345

changed some comments, and renamed some variables.

14346

14347

The two advantages of the new algorithm are (1) It makes writing past

14348

the end of the buffer virtually impossible because the test on the main

14349

loop is "while (not past end of buffer)" and (2) It's twice as fast

14350

because it doesn't call strlen. (There's no need to call strlen when

14351

we're walking the string ourselves.)

14352

14353

methodsNamed also supports arbitrary-length method names now. Just in

14354

case the AppKit folks start getting REALLY verbose...

14355

14356

* bindings/objc/objc_class.mm:

14357

(KJS::Bindings::ObjcClass::methodsNamed):

14358

* bindings/objc/objc_utility.h:

14359

* bindings/objc/objc_utility.mm:

14360

(KJS::Bindings::JSMethodNameToObjcMethodName):

14361

14362

2005-12-19 Darin Adler <darin@apple.com>

14363

14364

Originally done by both George Staikos and Alexey Proskuryakov.

14365

14366

- fix http://bugs.webkit.org/show_bug.cgi?id=5706

14367

Sharedptr dependency can be removed

14368

14369

Our coding guidelines say "use 0 instead of NULL" and both RefPtr and

14370

PassRefPtr were using NULL, which required including a header that

14371

defines NULL.

14372

14373

* kxmlcore/PassRefPtr.h:

14374

(KXMLCore::PassRefPtr::PassRefPtr): Use 0 instead of NULL.

14375

(KXMLCore::PassRefPtr::operator!): Use ! instead of == NULL.

14376

* kxmlcore/RefPtr.h:

14377

(KXMLCore::RefPtr::RefPtr): Use 0 instead of NULL.

14378

(KXMLCore::RefPtr::operator!): Use ! instead of == NULL.

14379

Also did some reformatting.

14380

14381

2005-12-19 Darin Adler <darin@apple.com>

14382

14383

Reviewed by Geoff Garen and Eric Seidel.

14384

14385

- fix http://bugs.webkit.org/show_bug.cgi?id=4923

14386

stop using <ostream> in WebCore, eliminating the <cmath> troubles it causes

14387

14388

* kjs/simple_number.h: Removed many unnecessary includes, including

14389

the <cmath> one to work around GCC library header bugs. We may have to

14390

add some includes elsewhere for platforms other than OS X, since our

14391

prefix header takes care of some things.

14392

14393

* kxmlcore/AlwaysInline.h: Added. Now clients that don't include

14394

simple_number.h can still get the ALWAYS_INLINE macro.

14395

* JavaScriptCore.xcodeproj/project.pbxproj: Added AlwaysInline.h.

14396

14397

* bindings/NP_jsobject.h: Removed a lot of unnecessary includes

14398

and removed C-specific stuff from this C++-only header.

14399

* bindings/jni/jni_jsobject.h: Removed a lot of unnecessary includes

14400

and did some reformatting.

14401

* bindings/objc/objc_runtime.h: Removed an unnecessary include.

14402

* bindings/runtime.h: Removed some unneeded includes. Reformatted.

14403

* bindings/runtime.cpp: Updated to compile with header changes,

14404

including a lot of reformatting.

14405

* bindings/runtime_object.h: Removed an unnecessary include.

14406

14407

2005-12-13 Maciej Stachowiak <mjs@apple.com>

14408

14409

Reviewed by Geoff and Adele

14410

14411

- replaced custom Identifier hashtable with HashSet

14412

14413

* kjs/identifier.cpp:

14414

(KXMLCore::):

14415

(KJS::identifierTable):

14416

(KJS::Identifier::equal):

14417

(KJS::hash):

14418

(KJS::equal):

14419

(KJS::convert):

14420

(KJS::Identifier::add):

14421

(KJS::Identifier::remove):

14422

* kjs/identifier.h:

14423

* kjs/internal.cpp:

14424

(KJS::InterpreterImp::initGlobalObject):

14425

14426

2005-12-18 Justin Haygood <justin@xiondigital.net>

14427

14428

Reviewed, tweaked, and landed by Darin.

14429

14430

- fix http://bugs.webkit.org/show_bug.cgi?id=5227

14431

Array indexOf() extension for JavaScript 1.5 Core

14432

14433

* kjs/array_object.h:

14434

* kjs/array_object.cpp: (ArrayProtoFunc::callAsFunction): Added implementation of indexOf.

14435

14436

2005-12-18 Anders Carlsson <andersca@mac.com>

14437

14438

Reviewed by Darin and Geoffrey.

14439

14440

- fix for <http://bugs.webkit.org/show_bug.cgi?id=4000>

14441

Object.prototype is missing isPrototypeOf

14442

14443

* kjs/object_object.cpp:

14444

(ObjectPrototype::ObjectPrototype):

14445

Add isPrototypeOf to object prototype.

14446

14447

(ObjectProtoFunc::callAsFunction):

14448

Implement isPrototypeOf

14449

14450

* kjs/object_object.h:

14451

(KJS::ObjectProtoFunc::):

14452

Add id for isPrototypeOf.

14453

14454

2005-12-17 Geoffrey Garen <ggaren@apple.com>

14455

14456

Reviewed by Darin.

14457

14458

Fixed http://bugs.webkit.org/show_bug.cgi?id=6119

14459

split() function ignores case insensitive modifier.

14460

14461

Glossary:

14462

RegExpImp: The C++ object you get when JavaScript executes

14463

"new RegExp()".

14464

RegExp: A C++ wrapper object that performs regular expression

14465

matching on behalf of a RegExpImp.

14466

14467

Instead of unnecessarily constructing a RegExp which (wrongly) lacks

14468

any modifiers, String.split() now uses the RegExp built in to the

14469

RegExpImp passed to it, which has the right modifiers already.

14470

14471

I also cleaned up other bits of the string code to standardized how

14472

we handle RegExpImp arguments.

14473

14474

* ChangeLog:

14475

* kjs/string_object.cpp:

14476

(replace):

14477

(StringProtoFunc::callAsFunction):

14478

14479

2005-12-16 David Hyatt <hyatt@apple.com>

14480

14481

Remove unused RefPtr constructors that can create an ambiguity in ustring on some platforms.

14482

14483

Reviewed by mjs

14484

14485

* kxmlcore/RefPtr.h:

14486

(KXMLCore::RefPtr::RefPtr):

14487

14488

2005-12-15 Darin Adler <darin@apple.com>

14489

14490

Reviewed by Maciej.

14491

14492

- fix http://bugs.webkit.org/show_bug.cgi?id=5688

14493

speed up JavaScript parsing by not creating a UString just to parse

14494

14495

* kjs/internal.h:

14496

* kjs/internal.cpp: (KJS::InterpreterImp::evaluate): Change to take a character pointer

14497

and length rather than a UString.

14498

14499

* kjs/interpreter.h:

14500

* kjs/interpreter.cpp: (Interpreter::evaluate): Ditto.

14501

14502

* kjs/protect.h: Remove uneeded "convert to bool" operator since we already have a

14503

"convert to raw pointer" operator in this class.

14504

14505

=== Safari-521~5 ===

14506

14507

2005-12-13 Geoffrey Garen <ggaren@apple.com>

14508

14509

Updated test results to match Anders's last fix.

14510

14511

* tests/mozilla/expected.html:

14512

14513

2005-12-13 Anders Carlsson <andersca@mac.com>

14514

14515

* ChangeLog: Add titles for my bugzilla bugs.

14516

14517

2005-12-13 Anders Carlsson <andersca@mac.com>

14518

14519

Reviewed by Darin.

14520

14521

- Fixes <http://bugs.webkit.org/show_bug.cgi?id=6041>

14522

Support property getters and setters.

14523

14524

* bindings/runtime_array.cpp:

14525

(RuntimeArray::lengthGetter):

14526

(RuntimeArray::indexGetter):

14527

* bindings/runtime_array.h:

14528

* bindings/runtime_method.cpp:

14529

(RuntimeMethod::lengthGetter):

14530

* bindings/runtime_method.h:

14531

* bindings/runtime_object.cpp:

14532

(RuntimeObjectImp::fallbackObjectGetter):

14533

(RuntimeObjectImp::fieldGetter):

14534

(RuntimeObjectImp::methodGetter):

14535

* bindings/runtime_object.h:

14536

* kjs/array_instance.h:

14537

* kjs/array_object.cpp:

14538

(ArrayInstance::lengthGetter):

14539

(getProperty):

14540

Update for changes to PropertySlot::getValue and

14541

PropertySlot::GetValueFunc.

14542

14543

* kjs/collector.cpp:

14544

(KJS::className):

14545

Handle GetterSetterType.

14546

14547

* kjs/function.cpp:

14548

(KJS::FunctionImp::argumentsGetter):

14549

(KJS::FunctionImp::lengthGetter):

14550

(KJS::Arguments::mappedIndexGetter):

14551

(KJS::ActivationImp::argumentsGetter):

14552

* kjs/function.h:

14553

Update for changes to PropertySlot::getValue and

14554

PropertySlot::GetValueFunc.

14555

14556

* kjs/grammar.y:

14557

Rework grammar parts for get set declarations directly

14558

in the object literal.

14559

14560

* kjs/internal.cpp:

14561

(KJS::GetterSetterImp::mark):

14562

(KJS::GetterSetterImp::toPrimitive):

14563

(KJS::GetterSetterImp::toBoolean):

14564

(KJS::GetterSetterImp::toNumber):

14565

(KJS::GetterSetterImp::toString):

14566

(KJS::GetterSetterImp::toObject):

14567

Add type conversion functions. These aren't meant to be called.

14568

14569

(KJS::printInfo):

14570

Handle GetterSetterType.

14571

14572

* kjs/lookup.h:

14573

(KJS::staticFunctionGetter):

14574

(KJS::staticValueGetter):

14575

Update for changes to PropertySlot::GetValueFunc.

14576

14577

* kjs/nodes.cpp:

14578

Refactor they way properties nodes are implemented.

14579

We now have a PropertyListNode which is a list of PropertyNodes.

14580

Each PropertyNode has a name (which is a PropertyNameNode) and an associated

14581

value node. PropertyNodes can be of different types. The Constant type is the

14582

old constant declaration and the Getter and Setter types are for property getters

14583

and setters.

14584

(ResolveNode::evaluate):

14585

Update for changes to PropertySlot::getValue.

14586

14587

(PropertyListNode::evaluate):

14588

Go through all property nodes and set them on the newly created object. If the

14589

property nodes are of type Getter or Setter, define getters and setters. Otherwise,

14590

just add the properties like before.

14591

14592

(PropertyNode::evaluate):

14593

This should never be called directly.

14594

14595

(PropertyNameNode::evaluate):

14596

Rename from PropertyNode::evaluate.

14597

14598

(FunctionCallResolveNode::evaluate):

14599

(FunctionCallBracketNode::evaluate):

14600

(FunctionCallDotNode::evaluate):

14601

(PostfixResolveNode::evaluate):

14602

(PostfixBracketNode::evaluate):

14603

(PostfixDotNode::evaluate):

14604

(TypeOfResolveNode::evaluate):

14605

(PrefixResolveNode::evaluate):

14606

(PrefixBracketNode::evaluate):

14607

(PrefixDotNode::evaluate):

14608

(AssignResolveNode::evaluate):

14609

(AssignDotNode::evaluate):

14610

(AssignBracketNode::evaluate):

14611

Update for changes to PropertySlot::getValue.

14612

14613

* kjs/nodes.h:

14614

(KJS::PropertyNameNode::PropertyNameNode):

14615

Rename from PropertyNode.

14616

14617

(KJS::PropertyNode::):

14618

(KJS::PropertyNode::PropertyNode):

14619

New class, representing a single property.

14620

14621

(KJS::PropertyListNode::PropertyListNode):

14622

Rename from PropertyValueNode.

14623

14624

(KJS::FuncExprNode::FuncExprNode):

14625

Put ParameterNode parameter last, and make it optional.

14626

14627

(KJS::ObjectLiteralNode::ObjectLiteralNode):

14628

Use a PropertyListNode here now.

14629

14630

* kjs/nodes2string.cpp:

14631

(PropertyListNode::streamTo):

14632

Iterate through all property nodes.

14633

14634

(PropertyNode::streamTo):

14635

Print out the name and value. Doesn't handle getters and setters currently.

14636

14637

(PropertyNameNode::streamTo):

14638

Rename from PropertyNode::streamTo.

14639

14640

* kjs/object.cpp:

14641

(KJS::JSObject::get):

14642

Update for changes to PropertySlot::getValue.

14643

14644

(KJS::JSObject::put):

14645

If the property already exists and has a Setter, invoke

14646

the setter function instead of setting the property directly.

14647

14648

(KJS::JSObject::defineGetter):

14649

(KJS::JSObject::defineSetter):

14650

New functions for defining property getters and setters on the object.

14651

14652

* kjs/object.h:

14653

(KJS::GetterSetterImp::type):

14654

(KJS::GetterSetterImp::GetterSetterImp):

14655

(KJS::GetterSetterImp::getGetter):

14656

(KJS::GetterSetterImp::setGetter):

14657

(KJS::GetterSetterImp::getSetter):

14658

(KJS::GetterSetterImp::setSetter):

14659

New class for properties which have getters and setters defined.

14660

This class is only used internally and should never be seen from the outside.

14661

14662

(KJS::JSObject::getOwnPropertySlot):

14663

If the property is a getter, call setGetterSlot on the property slot.

14664

14665

* kjs/object_object.cpp:

14666

(ObjectPrototype::ObjectPrototype):

14667

Add __defineGetter__, __defineSetter, __lookupGetter__, __lookupSetter__

14668

to prototype.

14669

14670

(ObjectProtoFunc::callAsFunction):

14671

Implement handlers for new functions.

14672

14673

* kjs/object_object.h:

14674

(KJS::ObjectProtoFunc::):

14675

Add ids for new functions.

14676

14677

* kjs/property_slot.cpp:

14678

(KJS::PropertySlot::undefinedGetter):

14679

Update for changes to PropertySlot::GetValueFunc.

14680

14681

(KJS::PropertySlot::functionGetter):

14682

Call the function getter object and return its value.

14683

14684

* kjs/property_slot.h:

14685

(KJS::PropertySlot::getValue):

14686

Add a new argument which is the original object that

14687

getPropertySlot was called on.

14688

14689

(KJS::PropertySlot::setGetterSlot):

14690

(KJS::PropertySlot::):

14691

New function which sets a getter slot. When getValue is called on a

14692

getter slot, the getter function object is invoked.

14693

14694

* kjs/string_object.cpp:

14695

(StringInstance::lengthGetter):

14696

(StringInstance::indexGetter):

14697

* kjs/string_object.h:

14698

Update for changes to PropertySlot::GetValueFunc.

14699

14700

* kjs/value.h:

14701

(KJS::):

14702

Add GetterSetterType and make GetterSetterImp a friend class of JSCell.

14703

14704

2005-12-12 Maciej Stachowiak <mjs@apple.com>

14705

14706

Reviewed by Eric.

14707

14708

- added a new HashCountedSet class for the common pattern of mapping items to counts that can change

14709

14710

* kxmlcore/HashCountedSet.h: Added.

14711

(KXMLCore::HashCountedSet::*): Implemented, on top of HashMap.

14712

* kxmlcore/HashMap.h:

14713

(KXMLCore::HashMap::add): New method - does not replace existing value if key already present

14714

but otherwise like set().

14715

(KXMLCore::HashMap::set): Improved comments.

14716

* kxmlcore/HashMapPtrSpec.h:

14717

(KXMLCore::HashMap::add): Added to specializations too.

14718

* JavaScriptCore.xcodeproj/project.pbxproj: Add new file.

14719

* kxmlcore/HashFunctions.h: Added include of stdint.h

14720

14721

- replaced the custom hashtable for values protected from GC with HashCountedSet

14722

14723

* kjs/collector.cpp:

14724

(KJS::Collector::protect): Moved code here from ProtectedValues::increaseProtectCount

14725

since the code is so simple now.

14726

(KJS::Collector::unprotect): Ditto for ProtectedValues::decreaseProtectCount.

14727

(KJS::Collector::markProtectedObjects): Updated for new way of doing things, now

14728

simpler and safer.

14729

(KJS::Collector::numReferencedObjects): ditto

14730

(KJS::Collector::rootObjectClasses): ditto

14731

* kjs/collector.h: Added protect and unprotect static methods

14732

* kjs/protect.h:

14733

(KJS::gcProtect): Updated for removal of ProtectedValues class

14734

(KJS::gcUnprotect): likewise

14735

* kjs/protected_values.cpp: Removed.

14736

* kjs/protected_values.h: Removed.

14737

14738

2005-12-10 Darin Adler <darin@apple.com>

14739

14740

Rubber stamped by Maciej.

14741

14742

- did long-promised KJS renaming:

14743

14744

ValueImp -> JSValue

14745

ObjectImp -> JSObject

14746

AllocatedValueImp -> JSCell

14747

14748

A renaming to get a class out of the way

14749

14750

KJS::Bindings::JSObject -> JavaJSObject

14751

14752

and some other "imp-reduction" renaming

14753

14754

*InstanceImp -> *Instance

14755

*ProtoFuncImp -> *ProtoFunc

14756

*PrototypeImp -> *Prototype

14757

ArgumentsImp -> Arguments

14758

RuntimeArrayImp -> RuntimeArray

14759

RuntimeMethodImp -> RuntimeMethod

14760

14761

* most files and functions

14762

14763

2005-12-10 Darin Adler <darin@apple.com>

14764

14765

Reviewed by Maciej.

14766

14767

- eliminated the old Undefined(), Null(), Boolean(), Number(), and String()

14768

14769

Code now uses jsUndefined(), jsNull(), jsBoolean(), jsNumber(), and jsString().

14770

14771

* bindings/NP_jsobject.cpp:

14772

(_NPN_Evaluate):

14773

* bindings/c/c_instance.cpp:

14774

(KJS::Bindings::CInstance::invokeMethod):

14775

(KJS::Bindings::CInstance::invokeDefaultMethod):

14776

* bindings/c/c_runtime.cpp:

14777

(CField::valueFromInstance):

14778

* bindings/c/c_utility.cpp:

14779

(KJS::Bindings::convertNPVariantToValue):

14780

* bindings/jni/jni_instance.cpp:

14781

(JavaInstance::invokeMethod):

14782

(JavaInstance::invokeDefaultMethod):

14783

* bindings/jni/jni_jsobject.cpp:

14784

(JSObject::eval):

14785

(JSObject::convertJObjectToValue):

14786

* bindings/jni/jni_runtime.cpp:

14787

(JavaArray::convertJObjectToArray):

14788

(JavaField::valueFromInstance):

14789

(JavaArray::valueAt):

14790

* bindings/objc/WebScriptObject.mm:

14791

(-[WebScriptObject callWebScriptMethod:withArguments:]):

14792

(-[WebScriptObject evaluateWebScript:]):

14793

(-[WebScriptObject valueForKey:]):

14794

(-[WebScriptObject webScriptValueAtIndex:]):

14795

* bindings/objc/objc_instance.mm:

14796

(ObjcInstance::invokeMethod):

14797

(ObjcInstance::invokeDefaultMethod):

14798

(ObjcInstance::getValueOfUndefinedField):

14799

* bindings/objc/objc_runtime.mm:

14800

(ObjcField::valueFromInstance):

14801

(ObjcFallbackObjectImp::callAsFunction):

14802

* bindings/objc/objc_utility.mm:

14803

(KJS::Bindings::convertNSStringToString):

14804

(KJS::Bindings::convertObjcValueToValue):

14805

* bindings/runtime.h:

14806

(KJS::Bindings::Class::fallbackObject):

14807

(KJS::Bindings::Instance::getValueOfUndefinedField):

14808

(KJS::Bindings::Instance::valueOf):

14809

* bindings/runtime_array.cpp:

14810

(RuntimeArrayImp::lengthGetter):

14811

* bindings/runtime_method.cpp:

14812

(RuntimeMethodImp::lengthGetter):

14813

(RuntimeMethodImp::callAsFunction):

14814

(RuntimeMethodImp::execute):

14815

* kjs/array_object.cpp:

14816

(ArrayInstanceImp::lengthGetter):

14817

(CompareWithCompareFunctionArguments::CompareWithCompareFunctionArguments):

14818

(ArrayPrototypeImp::ArrayPrototypeImp):

14819

(ArrayProtoFuncImp::ArrayProtoFuncImp):

14820

(ArrayProtoFuncImp::callAsFunction):

14821

(ArrayObjectImp::ArrayObjectImp):

14822

* kjs/bool_object.cpp:

14823

(BooleanPrototypeImp::BooleanPrototypeImp):

14824

(BooleanProtoFuncImp::callAsFunction):

14825

(BooleanObjectImp::BooleanObjectImp):

14826

(BooleanObjectImp::callAsFunction):

14827

* kjs/error_object.cpp:

14828

(ErrorPrototypeImp::ErrorPrototypeImp):

14829

(ErrorProtoFuncImp::ErrorProtoFuncImp):

14830

(ErrorProtoFuncImp::callAsFunction):

14831

(ErrorObjectImp::ErrorObjectImp):

14832

(NativeErrorImp::NativeErrorImp):

14833

* kjs/function.cpp:

14834

(KJS::FunctionImp::callAsFunction):

14835

(KJS::FunctionImp::processParameters):

14836

(KJS::FunctionImp::argumentsGetter):

14837

(KJS::FunctionImp::lengthGetter):

14838

(KJS::DeclaredFunctionImp::execute):

14839

(KJS::encode):

14840

(KJS::decode):

14841

(KJS::GlobalFuncImp::callAsFunction):

14842

* kjs/function_object.cpp:

14843

(FunctionPrototypeImp::FunctionPrototypeImp):

14844

(FunctionPrototypeImp::callAsFunction):

14845

(FunctionProtoFuncImp::callAsFunction):

14846

(FunctionObjectImp::FunctionObjectImp):

14847

* kjs/internal.cpp:

14848

(KJS::InterpreterImp::initGlobalObject):

14849

* kjs/interpreter.h:

14850

* kjs/lookup.h:

14851

* kjs/math_object.cpp:

14852

(MathObjectImp::getValueProperty):

14853

(MathFuncImp::callAsFunction):

14854

* kjs/nodes.cpp:

14855

(Node::setExceptionDetailsIfNeeded):

14856

(NullNode::evaluate):

14857

(PropertyNode::evaluate):

14858

(FunctionCallBracketNode::evaluate):

14859

(FunctionCallDotNode::evaluate):

14860

(PostfixBracketNode::evaluate):

14861

(PostfixDotNode::evaluate):

14862

(VoidNode::evaluate):

14863

(PrefixBracketNode::evaluate):

14864

(PrefixDotNode::evaluate):

14865

(ShiftNode::evaluate):

14866

(valueForReadModifyAssignment):

14867

(AssignDotNode::evaluate):

14868

(AssignBracketNode::evaluate):

14869

(VarDeclNode::evaluate):

14870

(VarDeclNode::processVarDecls):

14871

(VarDeclListNode::evaluate):

14872

(ReturnNode::execute):

14873

(CaseClauseNode::evalStatements):

14874

(ParameterNode::evaluate):

14875

(FuncDeclNode::processFuncDecl):

14876

* kjs/nodes.h:

14877

(KJS::StatementNode::evaluate):

14878

* kjs/number_object.cpp:

14879

(NumberPrototypeImp::NumberPrototypeImp):

14880

(NumberProtoFuncImp::callAsFunction):

14881

(NumberObjectImp::NumberObjectImp):

14882

(NumberObjectImp::getValueProperty):

14883

(NumberObjectImp::callAsFunction):

14884

* kjs/object.cpp:

14885

(KJS::ObjectImp::get):

14886

(KJS::Error::create):

14887

* kjs/object_object.cpp:

14888

(ObjectPrototypeImp::ObjectPrototypeImp):

14889

(ObjectProtoFuncImp::callAsFunction):

14890

(ObjectObjectImp::ObjectObjectImp):

14891

* kjs/property_slot.cpp:

14892

(KJS::PropertySlot::undefinedGetter):

14893

* kjs/regexp_object.cpp:

14894

(RegExpPrototypeImp::RegExpPrototypeImp):

14895

(RegExpProtoFuncImp::callAsFunction):

14896

(RegExpObjectImp::RegExpObjectImp):

14897

(RegExpObjectImp::arrayOfMatches):

14898

(RegExpObjectImp::getBackref):

14899

(RegExpObjectImp::getLastMatch):

14900

(RegExpObjectImp::getLastParen):

14901

(RegExpObjectImp::getLeftContext):

14902

(RegExpObjectImp::getRightContext):

14903

(RegExpObjectImp::getValueProperty):

14904

(RegExpObjectImp::construct):

14905

* kjs/string_object.cpp:

14906

(StringInstanceImp::StringInstanceImp):

14907

(StringPrototypeImp::StringPrototypeImp):

14908

(replace):

14909

(StringProtoFuncImp::callAsFunction):

14910

(StringObjectImp::StringObjectImp):

14911

(StringObjectImp::callAsFunction):

14912

(StringObjectFuncImp::StringObjectFuncImp):

14913

(StringObjectFuncImp::callAsFunction):

14914

* kjs/testkjs.cpp:

14915

(TestFunctionImp::callAsFunction):

14916

(VersionFunctionImp::callAsFunction):

14917

* kjs/value.h:

14918

14919

2005-12-10 Oliver Hunt <ojh16@student.canterbury.ac.nz>

14920

14921

Reviewed by Maciej, landed by Darin.

14922

14923

- fix http://bugs.webkit.org/show_bug.cgi?id=3539

14924

Array join and toString methods do not support circular references

14925

14926

* kjs/array_object.cpp: (ArrayProtoFuncImp::callAsFunction):

14927

Added set of visited objects -- don't recurse if item is already in the set.

14928

14929

2005-12-08 Maciej Stachowiak <mjs@apple.com>

14930

14931

Reviewed by John.

14932

14933

- fix major memory leak and resultant slowdown on JavaScript iBench from

14934

my PassRefPtr changes

14935

14936

* kjs/ustring.cpp:

14937

(KJS::UString::Rep::create): I forgot to change one of the two overloads to create

14938

with a refcount of 0 instead of 1 (the smart pointer then bumps it. But instead of

14939

changing it, I changed both to start with a refcounter of 1 and use PassRefPtr::adopt

14940

to adopt the initial refcount, this may be a hair more efficient.

14941

14942

- made the assignment operators for smart pointers inline because Shark said so

14943

14944

* kxmlcore/PassRefPtr.h:

14945

(KXMLCore::::operator=):

14946

* kxmlcore/RefPtr.h:

14947

(KXMLCore::::operator=):

14948

14949

2005-12-06 Anders Carlsson <andersca@mac.com>

14950

14951

Reviewed by Darin.

14952

14953

- fix build when using gcc 4

14954

14955

* kjs/ustring.h:

14956

Make Rep public.

14957

14958

* kxmlcore/PassRefPtr.h:

14959

(KXMLCore::::operator):

14960

Fix a typo.

14961

14962

2005-12-05 Maciej Stachowiak <mjs@apple.com>

14963

14964

Reviewed by Eric.

14965

14966

- add PassRefPtr, a smart pointer class that works in conjunction

14967

with RefPtr but has transfer-of-ownership semantics

14968

- apply RefPtr and PassRefPtr to UString

14969

- cleaned up UString a little so that it doesn't need to have so many friend classes

14970

14971

* JavaScriptCore.xcodeproj/project.pbxproj:

14972

* kjs/identifier.cpp:

14973

(KJS::Identifier::add):

14974

* kjs/identifier.h:

14975

(KJS::Identifier::Identifier):

14976

(KJS::Identifier::equal):

14977

* kjs/property_map.cpp:

14978

(KJS::PropertyMap::get):

14979

(KJS::PropertyMap::getLocation):

14980

(KJS::PropertyMap::put):

14981

(KJS::PropertyMap::remove):

14982

* kjs/ustring.cpp:

14983

(KJS::UCharReference::operator=):

14984

(KJS::UCharReference::ref):

14985

(KJS::UString::Rep::createCopying):

14986

(KJS::UString::Rep::create):

14987

(KJS::UString::usedCapacity):

14988

(KJS::UString::usedPreCapacity):

14989

(KJS::UString::expandCapacity):

14990

(KJS::UString::expandPreCapacity):

14991

(KJS::UString::UString):

14992

(KJS::UString::spliceSubstringsWithSeparators):

14993

(KJS::UString::append):

14994

(KJS::UString::operator=):

14995

(KJS::UString::toStrictUInt32):

14996

(KJS::UString::substr):

14997

(KJS::UString::copyForWriting):

14998

(KJS::operator==):

14999

* kjs/ustring.h:

15000

(KJS::UString::UString):

15001

(KJS::UString::~UString):

15002

(KJS::UString::data):

15003

(KJS::UString::isNull):

15004

(KJS::UString::isEmpty):

15005

(KJS::UString::size):

15006

(KJS::UString::rep):

15007

* kxmlcore/RefPtr.h:

15008

(KXMLCore::RefPtr::RefPtr):

15009

(KXMLCore::RefPtr::operator*):

15010

(KXMLCore::::operator):

15011

(KXMLCore::operator==):

15012

(KXMLCore::operator!=):

15013

(KXMLCore::static_pointer_cast):

15014

(KXMLCore::const_pointer_cast):

15015

15016

2005-12-04 Geoffrey Garen <ggaren@apple.com>

15017

15018

Update test results to match Anders's last checkin.

15019

15020

* tests/mozilla/expected.html:

15021

15022

2005-12-04 Anders Carlsson <andersca@mac.com>

15023

15024

Reviewed by Geoffrey.

15025

15026

- Fixes <http://bugs.webkit.org/show_bug.cgi?id=3999>

15027

Object.prototype is missing propertyIsEnumerable

15028

15029

* kjs/object.cpp:

15030

(KJS::ObjectImp::canPut):

15031

Refactor to use getPropertyAttributes.

15032

15033

(KJS::ObjectImp::propertyIsEnumerable):

15034

New function which checks if a property is enumerable.

15035

15036

(KJS::ObjectImp::getPropertyAttributes):

15037

* kjs/object.h:

15038

Add getPropertyAttributes and propertyIsEnumerable.

15039

15040

* kjs/object_object.cpp:

15041

(ObjectPrototypeImp::ObjectPrototypeImp):

15042

(ObjectProtoFuncImp::callAsFunction):

15043

* kjs/object_object.h:

15044

(KJS::ObjectProtoFuncImp::):

15045

Add propertyIsEnumerable to the Object prototype.

15046

15047

2005-12-01 Maciej Stachowiak <mjs@apple.com>

15048

15049

Reviewed by Tim Hatcher.

15050

15051

- removed deprecated reset, isNull and nonNull methods

15052

15053

* kxmlcore/RefPtr.h:

15054

15055

2005-12-01 Anders Carlsson <andersca@mac.com>

15056

15057

Reviewed by Darin.

15058

15059

- Fixes <http://bugs.webkit.org/show_bug.cgi?id=3382>

15060

nodes2strings.cpp fails to print left expression of ForInNode when 'var' is not used

15061

15062

Patch by Mark Rowe.

15063

15064

* kjs/nodes2string.cpp:

15065

(ForInNode::streamTo):

15066

Add lexpr if there's no varDecl.

15067

15068

2005-12-01 Maciej Stachowiak <mjs@apple.com>

15069

15070

Rubber stamped by Eric.

15071

15072

- renamed SharedPtr to RefPtr via script

15073

15074

* JavaScriptCore.xcodeproj/project.pbxproj:

15075

* kjs/function.cpp:

15076

(KJS::GlobalFuncImp::callAsFunction):

15077

* kjs/function.h:

15078

* kjs/function_object.cpp:

15079

(FunctionObjectImp::construct):

15080

* kjs/internal.cpp:

15081

(KJS::Parser::parse):

15082

(KJS::InterpreterImp::checkSyntax):

15083

(KJS::InterpreterImp::evaluate):

15084

* kjs/internal.h:

15085

* kjs/nodes.h:

15086

* kjs/nodes2string.cpp:

15087

(KJS::SourceStream::operator<<):

15088

* kjs/protect.h:

15089

* kxmlcore/RefPtr.h: Added.

15090

(KXMLCore::RefPtr::RefPtr):

15091

(KXMLCore::RefPtr::~RefPtr):

15092

(KXMLCore::RefPtr::isNull):

15093

(KXMLCore::RefPtr::notNull):

15094

(KXMLCore::RefPtr::reset):

15095

(KXMLCore::RefPtr::get):

15096

(KXMLCore::RefPtr::operator*):

15097

(KXMLCore::RefPtr::operator->):

15098

(KXMLCore::RefPtr::operator!):

15099

(KXMLCore::RefPtr::operator UnspecifiedBoolType):

15100

(KXMLCore::::operator):

15101

(KXMLCore::operator==):

15102

(KXMLCore::operator!=):

15103

(KXMLCore::static_pointer_cast):

15104

(KXMLCore::const_pointer_cast):

15105

* kxmlcore/SharedPtr.h: Removed.

15106

15107

2005-11-30 Maciej Stachowiak <mjs@apple.com>

15108

15109

Reviewed by Dave Hyatt.

15110

15111

- change idiom used for implicit bool conversion of smart pointers, because the old one gives weird error messages sometimes

15112

15113

* kjs/protect.h:

15114

(KJS::ProtectedPtr::operator UnspecifiedBoolType):

15115

* kxmlcore/SharedPtr.h:

15116

(KXMLCore::SharedPtr::operator UnspecifiedBoolType):

15117

15118

2005-11-29 Mitz Pettel <opendarwin.org@mitzpettel.com>

15119

15120

Reviewed by ggaren. Committed by eseidel.

15121

15122

Date conversion to local time gets the DST flag wrong sometimes

15123

http://bugs.webkit.org/show_bug.cgi?id=5514

15124

15125

* kjs/date_object.cpp:

15126

(KJS::isTime_tSigned):

15127

(KJS::DateProtoFuncImp::callAsFunction):

15128

15129

2005-11-26 Maciej Stachowiak <mjs@apple.com>

15130

15131

Rubber stamped by Eric.

15132

15133

- renamed InterpreterLock to JSLock

15134

15135

* bindings/NP_jsobject.cpp:

15136

(_NPN_Invoke):

15137

(_NPN_Evaluate):

15138

(_NPN_GetProperty):

15139

(_NPN_SetProperty):

15140

(_NPN_RemoveProperty):

15141

(_NPN_HasProperty):

15142

(_NPN_HasMethod):

15143

(_NPN_SetException):

15144

* bindings/jni/jni_jsobject.cpp:

15145

(JSObject::call):

15146

(JSObject::eval):

15147

(JSObject::getMember):

15148

(JSObject::setMember):

15149

(JSObject::removeMember):

15150

(JSObject::getSlot):

15151

(JSObject::setSlot):

15152

(JSObject::toString):

15153

(JSObject::convertJObjectToValue):

15154

* bindings/objc/WebScriptObject.mm:

15155

(-[WebScriptObject callWebScriptMethod:withArguments:]):

15156

(-[WebScriptObject evaluateWebScript:]):

15157

(-[WebScriptObject setValue:forKey:]):

15158

(-[WebScriptObject valueForKey:]):

15159

(-[WebScriptObject removeWebScriptKey:]):

15160

(-[WebScriptObject stringRepresentation]):

15161

(-[WebScriptObject webScriptValueAtIndex:]):

15162

(-[WebScriptObject setWebScriptValueAtIndex:value:]):

15163

(+[WebScriptObject _convertValueToObjcValue:originExecutionContext:executionContext:]):

15164

* bindings/runtime.cpp:

15165

(Instance::createRuntimeObject):

15166

* bindings/runtime_root.cpp:

15167

(KJS::Bindings::addNativeReference):

15168

(KJS::Bindings::removeNativeReference):

15169

(RootObject::removeAllNativeReferences):

15170

* bindings/runtime_root.h:

15171

(KJS::Bindings::RootObject::~RootObject):

15172

(KJS::Bindings::RootObject::setRootObjectImp):

15173

* bindings/testbindings.cpp:

15174

(main):

15175

* bindings/testbindings.mm:

15176

(main):

15177

* kjs/JSLock.cpp:

15178

(KJS::initializeJSLock):

15179

(KJS::JSLock::lock):

15180

(KJS::JSLock::unlock):

15181

(KJS::JSLock::lockCount):

15182

(KJS::JSLock::DropAllLocks::DropAllLocks):

15183

(KJS::JSLock::DropAllLocks::~DropAllLocks):

15184

* kjs/JSLock.h:

15185

(KJS::JSLock::JSLock):

15186

(KJS::JSLock::~JSLock):

15187

* kjs/collector.cpp:

15188

(KJS::Collector::allocate):

15189

(KJS::Collector::collect):

15190

* kjs/internal.cpp:

15191

(KJS::InterpreterImp::InterpreterImp):

15192

(KJS::InterpreterImp::clear):

15193

(KJS::InterpreterImp::checkSyntax):

15194

(KJS::InterpreterImp::evaluate):

15195

* kjs/interpreter.cpp:

15196

(Interpreter::evaluate):

15197

* kjs/protect.h:

15198

(KJS::::ProtectedPtr):

15199

(KJS::::~ProtectedPtr):

15200

(KJS::::operator):

15201

* kjs/protected_reference.h:

15202

(KJS::ProtectedReference::ProtectedReference):

15203

(KJS::ProtectedReference::~ProtectedReference):

15204

(KJS::ProtectedReference::operator=):

15205

* kjs/protected_values.cpp:

15206

(KJS::ProtectedValues::getProtectCount):

15207

(KJS::ProtectedValues::increaseProtectCount):

15208

(KJS::ProtectedValues::decreaseProtectCount):

15209

* kjs/testkjs.cpp:

15210

(TestFunctionImp::callAsFunction):

15211

(main):

15212

15213

2005-11-26 Darin Adler <darin@apple.com>

15214

15215

Reviewed by eseidel. Committed by eseidel.

15216

15217

Inline ScopeChain functions for speed.

15218

http://bugs.webkit.org/show_bug.cgi?id=5687

15219

15220

* kjs/object.h:

15221

(KJS::ScopeChain::mark):

15222

* kjs/scope_chain.cpp:

15223

* kjs/scope_chain.h:

15224

(KJS::ScopeChain::ref):

15225

(KJS::ScopeChain::operator=):

15226

(KJS::ScopeChain::bottom):

15227

(KJS::ScopeChain::push):

15228

(KJS::ScopeChain::pop):

15229

15230

2005-11-21 Maciej Stachowiak <mjs@apple.com>

15231

15232

Reviewed by Geoff.

15233

15234

<rdar://problem/4139620> Seed: WebKit: hang when sending XMLHttpRequest if automatic proxy config is used

15235

15236

Also factored locking code completely into a separate class, and

15237

added a convenient packaged way to temporarily drop locks.

15238

15239

* JavaScriptCore.xcodeproj/project.pbxproj:

15240

* kjs/JSLock.cpp: Added.

15241

(KJS::initializeInterpreterLock):

15242

(KJS::InterpreterLock::lock):

15243

(KJS::InterpreterLock::unlock):

15244

(KJS::InterpreterLock::lockCount):

15245

(KJS::InterpreterLock::DropAllLocks::DropAllLocks):

15246

(KJS::InterpreterLock::DropAllLocks::~DropAllLocks):

15247

* kjs/JSLock.h: Added.

15248

(KJS::InterpreterLock::InterpreterLock):

15249

(KJS::InterpreterLock::~InterpreterLock):

15250

* kjs/internal.cpp:

15251

* kjs/internal.h:

15252

* kjs/interpreter.cpp:

15253

* kjs/interpreter.h:

15254

* kjs/protect.h:

15255

* kjs/testkjs.cpp:

15256

(TestFunctionImp::callAsFunction):

15257

15258

2005-11-21 Eric Seidel <eseidel@apple.com>

15259

15260

Rubber-stamped by hyatt.

15261

15262

Removed JavaScriptCore+SVG target.

15263

15264

* JavaScriptCore.xcodeproj/project.pbxproj:

15265

15266

2005-11-15 Geoffrey Garen <ggaren@apple.com>

15267

15268

Reviewed by mjs.

15269

15270

- Fixed <rdar://problem/4342216> Installer crash in

15271

KJS::ValueImp::marked() when garbage collector runs inside call to

15272

ConstantValues::init()

15273

15274

I took responsibility for initializing and marking ConstantValues away

15275

from InterpreterImp, since it's possible to reference such a value

15276

before any interpreter has been created and after the last interpreter

15277

has been destroyed.

15278

15279

InterpreterImp::lock now initializes ConstantValues. It's a good

15280

place for the initialization because you have to call it before

15281

creating any objects. Since ::lock can be called more than once,

15282

I added a check in ConstantValues::init to ensure that it executes

15283

only once.

15284

15285

Collector:collect is now responsible for marking ConstantValues.

15286

15287

We no longer clear the ConstantValues since we can't guarantee that no

15288

one has a reference to them.

15289

15290

FIXME: This is hackery. The long-term plan is to make ConstantValues

15291

use immediate values that require no initialization.

15292

15293

* ChangeLog:

15294

* kjs/collector.cpp:

15295

(KJS::Collector::collect):

15296

* kjs/internal.cpp:

15297

(KJS::InterpreterImp::InterpreterImp):

15298

(KJS::InterpreterImp::lock):

15299

(KJS::InterpreterImp::clear):

15300

(KJS::InterpreterImp::mark):

15301

* kjs/internal.h:

15302

* kjs/value.cpp:

15303

(KJS::ConstantValues::initIfNeeded):

15304

* kjs/value.h:

15305

15306

2005-11-08 Geoffrey Garen <ggaren@apple.com>

15307

15308

Reviewed by Darin.

15309

15310

This patch fixes some naughty naughty code -- 5 crashes and 2

15311

may-go-haywire-in-the-futures.

15312

15313

One such crash is <rdar://problem/4247330> 8C46 Crash with with

15314

incomplete parameter list to webScript object function.

15315

15316

I replaced early returns from within NS_DURINGs with calls to

15317

NS_VALUERETURN because the doc says, "You cannot use goto or

15318

return to exit an exception handling domain -- errors will result."

15319

15320

I replaced hard-coded analyses of -[NSMethodSignature

15321

methodReturnType] with more abstracted alternatives, since

15322

the documentation says "This encoding is implementation-specific,

15323

so applications should use it with caution," and then emits an

15324

evil cackle.

15325

15326

I removed the early return in the case where a JavaScript caller

15327

supplies an insufficient number of arguments, because the right

15328

thing to do in such a case is to use JavaScript's defined behavior

15329

of supplying "undefined" for any missing arguments.

15330

15331

I also changed ObjcInstance::invokeMethod so that it no longer

15332

deletes the method passed to it. It doesn't create the method,

15333

so it shouldn't delete it. A friend of mine named

15334

KERNEL_PROTECTION_FAILURE agrees with me on this point.

15335

15336

Finally, I changed an assert(true) to assert(false) because

15337

all the other asserts were making fun of it.

15338

15339

* bindings/objc/objc_instance.mm:

15340

(ObjcInstance::invokeMethod):

15341

(ObjcInstance::invokeDefaultMethod):

15342

15343

2005-11-06 Geoffrey Garen <ggaren@apple.com>

15344

15345

Reviewed by Darin.

15346

15347

- Fixed http://bugs.webkit.org/show_bug.cgi?id=5571

15348

REGRESSION (412.5-TOT): duplicated words/sentences at

15349

shakespeer.sourceforge.net

15350

15351

Our UTF16-modified PCRE didn't work with extended character classes

15352

(classes involving characters > 255) because it used the GETCHARINC

15353

macro to read through them. In UTF16 mode, GETCHARINC expects UTF16

15354

input, but PCRE encodes character classes in UTF8 regardless of the

15355

input mode of the subject string.

15356

15357

The fix is to explicitly define GETUTF8CHARINC, and to use it,

15358

rather than GETCHARINC, when reading extended character classes.

15359

15360

In UTF8 mode, we simply define GETCHARINC to be GETUTF8CHARINC.

15361

15362

* pcre/pcre_internal.h:

15363

* pcre/pcre_xclass.c:

15364

(_pcre_xclass):

15365

15366

2005-11-05 Geoffrey Garen <ggaren@apple.com>

15367

15368

Patch by Mitz Pettel, reviewed by Maciej.

15369

15370

- Fixed http://bugs.webkit.org/show_bug.cgi?id=5357

15371

REGRESSION: Scriptable plugin hides properties of OBJECT element

15372

15373

* bindings/objc/objc_class.mm:

15374

(KJS::Bindings::ObjcClass::fallbackObject):

15375

15376

2005-11-05 Geoffrey Garen <ggaren@apple.com>

15377

15378

Reviewed by Darin.

15379

15380

- Fixed http://bugs.webkit.org/show_bug.cgi?id=5409

15381

slice() testcase doesn't pass

15382

15383

Modified String.slice to deal with funky values.

15384

Updated test results. We now pass <js1_2/String/slice.js>.

15385

15386

* kjs/string_object.cpp:

15387

(StringProtoFuncImp::callAsFunction):

15388

* tests/mozilla/expected.html:

15389

15390

2005-11-04 Darin Adler <darin@apple.com>

15391

15392

Reviewed by Tim Hatcher.

15393

15394

* kxmlcore/HashSet.h: Fixed case of "hashfunctions.h" -- needs to be "HashFunctions.h".

15395

15396

2005-11-03 Timothy Hatcher <timothy@apple.com>

15397

15398

Reviewed by Darin and Vicki.

15399

15400

* JavaScriptCore.xcodeproj/project.pbxproj:

15401

Change to use $(SYSTEM_LIBRARY_DIR) consistently and place

15402

$(NEXT_ROOT) in a few spots to make build-root work.

15403

15404

2005-11-03 Geoffrey Garen <ggaren@apple.com>

15405

15406

- Updated JavaScriptCore test results to reflect recent fixes.

15407

15408

* tests/mozilla/expected.html:

15409

15410

2005-11-03 Geoffrey Garen <ggaren@apple.com>

15411

15412

Reviewed by darin.

15413

15414

- Fixed http://bugs.webkit.org/show_bug.cgi?id=5602

15415

REGRESSION: RegExp("[^\\s$]+", "g") returns extra matches

15416

15417

We now update lastIndex relative to the start of the last match,

15418

rather than the start of the last search. We used to assume that

15419

the two were equal, but that is not the case when a pattern

15420

matches at a character after the first in the string.

15421

15422

* kjs/regexp_object.cpp:

15423

(RegExpProtoFuncImp::callAsFunction):

15424

15425

2005-10-24 John Sullivan <sullivan@apple.com>

15426

15427

Reviewed by Darin Adler. Code changes by Alexey Proskuryakov.

15428

15429

- fixed http://bugs.webkit.org/show_bug.cgi?id=4931

15430

Unicode format characters (Cf) should be removed from JavaScript source

15431

15432

* kjs/lexer.cpp:

15433

include <unicode/uchar.h>

15434

(Lexer::Lexer):

15435

use KJS::UChar instead of UChar to avoid ambiguity caused by new include

15436

(Lexer::setCode):

15437

ditto; also, use shift(4) to skip first 4 chars to take advantage of new

15438

logic there.

15439

(Lexer::shift):

15440

skip chars of type U_FORMAT_CHAR

15441

(Lexer::convertUnicode):

15442

use KJS::UChar instead of UChar to avoid ambiguity caused by new include

15443

(Lexer::record16):

15444

ditto

15445

(Lexer::makeIdentifier):

15446

ditto

15447

(Lexer::makeUString):

15448

ditto

15449

15450

* tests/mozilla/ecma/Array/15.4.5.1-1.js:

15451

updated to skip soft hyphens

15452

15453

2005-10-24 John Sullivan <sullivan@apple.com>

15454

15455

Reviewed by Darin Adler. Code changes by George Staikos/Geoff Garen.

15456

15457

- fixed http://bugs.webkit.org/show_bug.cgi?id=4142

15458

Date object does not always adjust daylight savings correctly

15459

15460

* kjs/date_object.cpp:

15461

(KJS::makeTime):

15462

Fix the case where a time change crosses the daylight savings start/end dates.

15463

15464

2005-10-17 Maciej Stachowiak <mjs@apple.com>

15465

15466

Reviewed by Geoff. Code changes by Darin.

15467

15468

- some micro-optimizations to FastMalloc to reduce math and branches.

15469

15470

* kxmlcore/FastMalloc.cpp:

15471

(KXMLCore::TCMalloc_Central_FreeList::Populate):

15472

(KXMLCore::fastMallocRegisterThread):

15473

(KXMLCore::TCMalloc_ThreadCache::GetCache):

15474

(KXMLCore::TCMalloc_ThreadCache::GetCacheIfPresent):

15475

15476

2005-10-15 Maciej Stachowiak <mjs@apple.com>

15477

15478

Reverted fix for this bug, because it was part of a time range that caused a performance

15479

regression:

15480

15481

<rdar://problem/4260481> Remove Reference type from JavaScriptCore

15482

15483

2005-10-15 Darin Adler <darin@apple.com>

15484

15485

* kxmlcore/HashTable.cpp: Fixed build failure (said hashtable.h instead of HashTable.h).

15486

15487

2005-10-14 Geoffrey Garen <ggaren@apple.com>

15488

15489

Style changes recommended by Darin.

15490

15491

Changed to camelCase, changed ValueImp* to ValueImp *.

15492

15493

* kjs/simple_number.h:

15494

(KJS::SimpleNumber::make):

15495

(KJS::SimpleNumber::value):

15496

15497

2005-10-11 Geoffrey Garen <ggaren@apple.com>

15498

15499

Added regexp_object.lut.h build phase from JavaScriptCore

15500

to JavaScriptCore+SVG.

15501

15502

Reviewed by mitz.

15503

15504

* JavaScriptCore.xcodeproj/project.pbxproj:

15505

15506

2005-10-11 Geoffrey Garen <ggaren@apple.com>

15507

15508

Fixed build bustage from last checkin (stray characters

15509

in the project file).

15510

15511

Reviewed by mitz.

15512

15513

* JavaScriptCore.xcodeproj/project.pbxproj:

15514

15515

2005-10-11 Geoffrey Garen <ggaren@apple.com>

15516

15517

New JavaScriptCore test results to reflect the last change.

15518

15519

* tests/mozilla/expected.html:

15520

15521

2005-10-10 Geoffrey Garen <ggaren@apple.com>

15522

15523

- Implemented caching of match state inside the global RegExp object

15524

(lastParen, leftContext, rightContext, lastMatch, input).

15525

15526

exec(), test(), match(), search(), and replace() now dipatch regular

15527

expression matching through the RegExp object's performMatch function,

15528

to facilitate caching. This replaces registerRegexp and

15529

setSubPatterns.

15530

15531

- Implemented the special '$' aliases (e.g. RegExp.input aliases to

15532

RegExp.$_).

15533

15534

- Moved support for backreferences into the new static hash table

15535

used for other special RegExp properties. Truncated backreferences

15536

at $9 to match IE, FF, and the "What's New in Netscape 1.2?" doc.

15537

(String.replace still supports double-digit backreferences.)

15538

15539

- Tweaked RegExp.prototype.exec to handle ginormous values in lastIndex.

15540

15541

Fixes 11 -- count em, 11 -- JavaScriptCore tests.

15542

15543

* fast/js/regexp-caching-expected.txt: Added.

15544

* fast/js/regexp-caching.html: Added.

15545

15546

Reviewed by mjs.

15547

15548

* JavaScriptCore.xcodeproj/project.pbxproj: Added regexp_object.lut.h

15549

* kjs/create_hash_table: Tweaked to allow for more exotic characters.

15550

We now rely on the compiler to catch illegal

15551

identifiers.

15552

* kjs/regexp.cpp:

15553

(KJS::RegExp::RegExp):

15554

* kjs/regexp_object.cpp:

15555

(RegExpProtoFuncImp::callAsFunction):

15556

(RegExpObjectImp::RegExpObjectImp):

15557

(RegExpObjectImp::performMatch):

15558

(RegExpObjectImp::arrayOfMatches):

15559

(RegExpObjectImp::backrefGetter):

15560

(RegExpObjectImp::getLastMatch):

15561

(RegExpObjectImp::getLastParen):

15562

(RegExpObjectImp::getLeftContext):

15563

(RegExpObjectImp::getRightContext):

15564

(RegExpObjectImp::getOwnPropertySlot):

15565

(RegExpObjectImp::getValueProperty):

15566

(RegExpObjectImp::put):

15567

(RegExpObjectImp::putValueProperty):

15568

* kjs/regexp_object.h:

15569

(KJS::RegExpObjectImp::):

15570

* kjs/string_object.cpp:

15571

(substituteBackreferences):

15572

(replace):

15573

(StringProtoFuncImp::callAsFunction):

15574

15575

2005-10-09 Darin Adler <darin@apple.com>

15576

15577

Reviewed by Maciej; some changes done after review.

15578

15579

- fixed <rdar://problem/4092064> hanging loading page; rte.ie (works in IE and Firefox)

15580

- fixed http://bugs.webkit.org/show_bug.cgi?id=5280

15581

Date.setMonth fails with negative values

15582

- fixed http://bugs.webkit.org/show_bug.cgi?id=5154

15583

JSC should switch to _r variants of unix time/date functions

15584

- fixed a few possible overflow cases

15585

15586

Retested all tests to be sure nothing broke; added layout test for bug 5280.

15587

15588

* kjs/config.h: Removed TIME_WITH_SYS_TIME define. Also set HAVE_SYS_TIMEB_H

15589

for the __APPLE__ case (the latter is accurate but irrelevant).

15590

15591

* kjs/date_object.h: Reformatted. Removed unnecessary include of "function_object.h".

15592

Moved declarations of helper classes and functions into the cpp file.

15593

15594

* kjs/date_object.cpp: Removed code at top to define macros to use CoreFoundation instead of

15595

POSIX date functions.

15596

(KJS::styleFromArgString): Tweaked to return early instead of using a variable.

15597

(KJS::formatLocaleDate): Tweaked to check for undefined rather than checking argument count.

15598

(KJS::formatDate): Made parameter const.

15599

(KJS::formatDateUTCVariant): Ditto.

15600

(KJS::formatTime): Ditto.

15601

(KJS::DateProtoFuncImp::callAsFunction): Use gmtime_r and localtime_r instead of gmtime and

15602

localtime.

15603

(KJS::DateObjectImp::callAsFunction): Use localtime_r instead of localtime.

15604

(KJS::ymdhmsToSeconds): Renamed from ymdhms_to_seconds. Changed computation to avoid possible

15605

overflow if year is an extremely large or small number.

15606

(KJS::makeTime): Removed code to move large month numbers from tm_mon to tm_year; this was

15607

to accomodate CFGregorianDate, which is no longer used (and didn't handle negative values).

15608

(KJS::parseDate): Renamed from KRFCDate_parseDate; changed to return a value in milliseconds

15609

rather than in seconds. Reformatted the code. Changed to use UTF8String() instead of ascii(),

15610

since ascii() is not thread safe. Changed some variables back from int to long to avoid

15611

trouble if the result of strtol does not fit in an int (64-bit issue only).

15612

15613

2005-10-08 Mitz Pettel <opendarwin.org@mitzpettel.com>

15614

15615

Reviewed by Geoff.

15616

Tweaked and landed by Darin.

15617

15618

- fixed http://bugs.webkit.org/show_bug.cgi?id=5266

15619

Support parenthesized comments in Date.parse()

15620

15621

* kjs/date_object.cpp:

15622

(KJS::skipSpacesAndComments): Take a pointer, and advance it past spaces,

15623

and also past anything enclosed in parentheses.

15624

(KJS::KRFCDate_parseDate): Use skipSpacesAndComments wherever we formerly had

15625

code to skip spaces.

15626

15627

2005-10-08 Justin Haygood <justin@xiondigital.net>

15628

15629

Reviewed, tweaked, and landed by Darin.

15630

15631

- fixed http://bugs.webkit.org/show_bug.cgi?id=5189

15632

pcre_exec.c fails to compile using MSVC

15633

- fixed http://bugs.webkit.org/show_bug.cgi?id=5190

15634

KJS config.h adjustment for Win32

15635

15636

* kjs/config.h: Make sure HAVE_MMAP and HAVE_SBRK are off for Win32.

15637

Turn HAVE_ERRNO_H on for Mac OS X. Sort defines so they are easy to compare

15638

with each other. Remove #undef of DEBUG_COLLECTOR.

15639

* pcre/pcre_exec.c: (match): Work around strange MSVC complaint by splitting

15640

the definition of a local variable into a separate declaration and

15641

initialization.

15642

15643

2005-10-05 Geoffrey Garen <ggaren@apple.com>

15644

15645

- Darin and I rewrote our implementation of the SimpleNumber class

15646

to store number bit patterns in their floating point formats.

15647

15648

My tweaks reviewed by Darin.

15649

15650

~1% speedup on JS iBench.

15651

15652

* kjs/internal.h: removed obsolete jsNumber declarations.

15653

* kjs/math_object.cpp:

15654

(MathFuncImp::callAsFunction): changed KJS::isNaN to isNaN

15655

* kjs/nodes.cpp:

15656

(PostfixResolveNode::evaluate): removed obsolete knownToBeInteger

15657

(PostfixBracketNode::evaluate): ditto

15658

(PostfixDotNode::evaluate): ditto

15659

(PrefixResolveNode::evaluate): ditto

15660

(PrefixBracketNode::evaluate): ditto

15661

(PrefixDotNode::evaluate): ditto

15662

(NegateNode::evaluate): ditto

15663

(valueForReadModifyAssignment): ditto

15664

* kjs/number_object.cpp: removed obsolete comment

15665

* kjs/operations.cpp:

15666

(KJS::equal): removed unnecessary isNaN checks

15667

(KJS::strictEqual): ditto

15668

(KJS::add): removed obsolete knownToBeInteger

15669

(KJS::mult): ditto

15670

* kjs/operations.h: removed include of "value.h" to prevent circular reference

15671

* kjs/simple_number.h: removed unnecessary #includes

15672

(KJS::SimpleNumber::make): see above

15673

(KJS::SimpleNumber::is): ditto

15674

(KJS::SimpleNumber::value): ditto

15675

* kjs/string_object.cpp:

15676

(StringProtoFuncImp::callAsFunction): changed KJS::isNaN to isNaN

15677

* kjs/ustring.cpp: removed unnecessary isNaN check

15678

(KJS::UString::toUInt32): ditto

15679

* kjs/value.cpp:

15680

(KJS::jsNumber): removed obsolete jsNumber definitions

15681

(KJS::ConstantValues::init): NaN is no longer a ConstantValue

15682

(KJS::ConstantValues::clear): ditto

15683

(KJS::ConstantValues::mark): ditto

15684

* kjs/value.h: removed obsolete knownToBeInteger

15685

(KJS::jsNaN): now returns a SimpleNumber

15686

(KJS::ValueImp::getUInt32): changed to account for NaN being a SimpleNumber

15687

(KJS::ValueImp::toBoolean): ditto

15688

(KJS::ValueImp::toString): changed to account for +/- 0.0

15689

(KJS::jsZero): changed to reflect that SimpleNumber::make takes a double

15690

(KJS::jsOne): ditto

15691

(KJS::jsTwo): ditto

15692

(KJS::Number): removed obsolete non-double constructor declarations

15693

15694

2005-10-05 Maciej Stachowiak <mjs@apple.com>

15695

15696

Reviewed by Eric.

15697

15698

- fixed <rdar://problem/4260481> Remove Reference type from JavaScriptCore

15699

15700

Also fixed some bugs with for..in enumeration while I was at it. object

15701

properties now come before prototype properties and duplicates

15702

between object and prototype are listed only once.

15703

15704

* JavaScriptCore.xcodeproj/project.pbxproj:

15705

* kjs/IdentifierSequencedSet.cpp: Added.

15706

(KJS::IdentifierSequencedSet::IdentifierSequencedSet):

15707

(KJS::IdentifierSequencedSet::deallocateVector):

15708

(KJS::IdentifierSequencedSet::~IdentifierSequencedSet):

15709

(KJS::IdentifierSequencedSet::insert):

15710

* kjs/IdentifierSequencedSet.h: Added.

15711

(KJS::IdentifierSequencedSetIterator::IdentifierSequencedSetIterator):

15712

(KJS::IdentifierSequencedSetIterator::operator*):

15713

(KJS::IdentifierSequencedSetIterator::operator->):

15714

(KJS::IdentifierSequencedSetIterator::operator++):

15715

(KJS::IdentifierSequencedSetIterator::operator==):

15716

(KJS::IdentifierSequencedSetIterator::operator!=):

15717

(KJS::IdentifierSequencedSet::begin):

15718

(KJS::IdentifierSequencedSet::end):

15719

(KJS::IdentifierSequencedSet::size):

15720

* kjs/array_instance.h:

15721

* kjs/array_object.cpp:

15722

(ArrayInstanceImp::getPropertyNames):

15723

(ArrayInstanceImp::setLength):

15724

(ArrayInstanceImp::pushUndefinedObjectsToEnd):

15725

* kjs/nodes.cpp:

15726

(ForInNode::execute):

15727

* kjs/nodes.h:

15728

* kjs/object.cpp:

15729

(KJS::ObjectImp::getPropertyNames):

15730

* kjs/object.h:

15731

* kjs/property_map.cpp:

15732

(KJS::PropertyMap::getEnumerablePropertyNames):

15733

(KJS::PropertyMap::getSparseArrayPropertyNames):

15734

* kjs/property_map.h:

15735

* kjs/protect.h:

15736

* kjs/protected_reference.h: Removed.

15737

* kjs/reference.cpp: Removed.

15738

* kjs/reference.h: Removed.

15739

* kjs/reference_list.cpp: Removed.

15740

* kjs/reference_list.h: Removed.

15741

* kjs/ustring.h:

15742

(KJS::UString::impl):

15743

* kxmlcore/HashSet.h:

15744

15745

2005-10-04 Eric Seidel <eseidel@apple.com>

15746

15747

Reviewed by mjs.

15748

15749

Code cleanup, which resulted in a small win on iBench.

15750

15751

* kjs/object.cpp:

15752

(KJS::tryGetAndCallProperty): new static inline

15753

(KJS::ObjectImp::defaultValue): code cleanup

15754

15755

2005-10-03 Maciej Stachowiak <mjs@apple.com>

15756

15757

Patch from George Staikos <staikos@kde.org>, reviewed and tweaked a bit by me.

15758

15759

- more Linux build fixes

15760

15761

* kjs/operations.cpp:

15762

* kxmlcore/FastMalloc.h:

15763

* kxmlcore/TCSystemAlloc.cpp:

15764

(TCMalloc_SystemAlloc):

15765

15766

2005-10-03 Maciej Stachowiak <mjs@apple.com>

15767

15768

Patch from George Staikos <staikos@kde.org>, reviewed and tweaked a bit by me.

15769

15770

http://bugs.webkit.org/show_bug.cgi?id=5174

15771

Add support for compiling on Linux (likely to help for other POSIX systems too)

15772

15773

* kjs/collector.cpp:

15774

(KJS::Collector::markCurrentThreadConservatively):

15775

(KJS::Collector::markOtherThreadConservatively):

15776

* kjs/config.h:

15777

* kjs/date_object.cpp:

15778

(KJS::formatDate):

15779

(KJS::formatDateUTCVariant):

15780

(KJS::formatTime):

15781

(KJS::timeZoneOffset):

15782

(KJS::DateProtoFuncImp::callAsFunction):

15783

(KJS::DateObjectImp::construct):

15784

(KJS::DateObjectImp::callAsFunction):

15785

(KJS::makeTime):

15786

* kjs/identifier.cpp:

15787

* kjs/internal.cpp:

15788

(KJS::initializeInterpreterLock):

15789

(KJS::lockInterpreter):

15790

(KJS::unlockInterpreter):

15791

(KJS::UndefinedImp::toPrimitive):

15792

(KJS::UndefinedImp::toBoolean):

15793

(KJS::UndefinedImp::toNumber):

15794

(KJS::UndefinedImp::toString):

15795

(KJS::NullImp::toPrimitive):

15796

(KJS::NullImp::toBoolean):

15797

(KJS::NullImp::toNumber):

15798

(KJS::NullImp::toString):

15799

(KJS::BooleanImp::toPrimitive):

15800

(KJS::BooleanImp::toBoolean):

15801

(KJS::BooleanImp::toNumber):

15802

(KJS::BooleanImp::toString):

15803

(KJS::StringImp::toPrimitive):

15804

(KJS::StringImp::toBoolean):

15805

(KJS::StringImp::toNumber):

15806

(KJS::StringImp::toString):

15807

* kjs/internal.h:

15808

* kjs/protected_values.cpp:

15809

15810

2005-10-03 Maciej Stachowiak <mjs@apple.com>

15811

15812

- fix Development build after last checkin

15813

15814

* kxmlcore/FastMalloc.cpp:

15815

(KXMLCore::fastMallocRegisterThread):

15816

15817

2005-10-02 Maciej Stachowiak <mjs@apple.com>

15818

15819

Reviewed by Darin.

15820

15821

<rdar://problem/4283967> REGRESSION: 3% regression on PLT from new FastMalloc

15822

http://bugs.webkit.org/show_bug.cgi?id=5243

15823

15824

A number of optimizations to the new threadsafe malloc that make it actually as fast

15825

as dlmalloc (I measured wrong before) and as memory-efficient as the system malloc.

15826

15827

- use fastMalloc for everything - it now gets applied to all new/delete allocations

15828

via a private inline operator new that is now included into every file via config.h.

15829

15830

- tweaked some of the numeric parameters for size classes and amount of wasted memory

15831

allowed per allocation - this saves on memory use and consequently improves speed.

15832

15833

- so long as the allocator is not being used on background threads, get the per-thread

15834

cache from a global variable instead of from pthread_getspecific, since the latter is slow.

15835

15836

- inline more functions, and force the ones GCC refuses to inline with

15837

attribute(always_inline), nearly all of these have one call site so inlining them has

15838

to be a win.

15839

15840

- use some tricks to calculate allocation size more efficiently and fewer times for small

15841

allocations, to avoid hitting the huge size table array.

15842

15843

- avoid hitting the per-thread cache on code paths that don't need it.

15844

15845

- implement inline assembly version of spinlock for PowerPC (was already done for x86)

15846

15847

* bindings/NP_jsobject.cpp:

15848

* bindings/c/c_class.cpp:

15849

* bindings/c/c_instance.cpp:

15850

* bindings/c/c_runtime.cpp:

15851

* bindings/c/c_utility.cpp:

15852

* bindings/jni/jni_class.cpp:

15853

* bindings/jni/jni_instance.cpp:

15854

* bindings/jni/jni_jsobject.cpp:

15855

* bindings/jni/jni_objc.mm:

15856

* bindings/jni/jni_runtime.cpp:

15857

* bindings/jni/jni_utility.cpp:

15858

* bindings/npruntime.cpp:

15859

* bindings/objc/WebScriptObject.mm:

15860

* bindings/objc/objc_class.mm:

15861

* bindings/objc/objc_instance.mm:

15862

* bindings/objc/objc_runtime.mm:

15863

* bindings/objc/objc_utility.mm:

15864

* bindings/runtime.cpp:

15865

* bindings/runtime_array.cpp:

15866

* bindings/runtime_method.cpp:

15867

* bindings/runtime_object.cpp:

15868

* bindings/runtime_root.cpp:

15869

* bindings/testbindings.cpp:

15870

* bindings/testbindings.mm:

15871

* kjs/array_object.cpp:

15872

(ArrayInstanceImp::ArrayInstanceImp):

15873

(ArrayInstanceImp::~ArrayInstanceImp):

15874

(ArrayInstanceImp::resizeStorage):

15875

* kjs/bool_object.cpp:

15876

* kjs/collector.cpp:

15877

(KJS::Collector::registerThread):

15878

* kjs/config.h:

15879

* kjs/debugger.cpp:

15880

* kjs/error_object.cpp:

15881

* kjs/function.cpp:

15882

* kjs/function_object.cpp:

15883

* kjs/identifier.cpp:

15884

(KJS::Identifier::rehash):

15885

* kjs/internal.cpp:

15886

(KJS::Parser::saveNewNode):

15887

(KJS::clearNewNodes):

15888

* kjs/interpreter.cpp:

15889

* kjs/lexer.cpp:

15890

(Lexer::doneParsing):

15891

(Lexer::makeIdentifier):

15892

(Lexer::makeUString):

15893

* kjs/list.cpp:

15894

* kjs/math_object.cpp:

15895

* kjs/nodes.cpp:

15896

* kjs/nodes.h:

15897

* kjs/nodes2string.cpp:

15898

* kjs/number_object.cpp:

15899

(integer_part_noexp):

15900

(char_sequence):

15901

* kjs/object.cpp:

15902

* kjs/object_object.cpp:

15903

* kjs/property_map.cpp:

15904

* kjs/property_slot.cpp:

15905

* kjs/protected_values.cpp:

15906

(KJS::ProtectedValues::rehash):

15907

* kjs/reference.cpp:

15908

* kjs/reference_list.cpp:

15909

* kjs/regexp.cpp:

15910

* kjs/regexp_object.cpp:

15911

* kjs/scope_chain.cpp:

15912

* kjs/scope_chain.h:

15913

* kjs/string_object.cpp:

15914

* kjs/testkjs.cpp:

15915

* kjs/ustring.h:

15916

* kjs/value.cpp:

15917

* kxmlcore/Assertions.mm:

15918

* kxmlcore/FastMalloc.cpp:

15919

(KXMLCore::InitSizeClasses):

15920

(KXMLCore::DLL_IsEmpty):

15921

(KXMLCore::DLL_Prepend):

15922

(KXMLCore::TCMalloc_Central_FreeList::Insert):

15923

(KXMLCore::TCMalloc_Central_FreeList::Remove):

15924

(KXMLCore::TCMalloc_Central_FreeList::Populate):

15925

(KXMLCore::TCMalloc_ThreadCache::Allocate):

15926

(KXMLCore::TCMalloc_ThreadCache::FetchFromCentralCache):

15927

(KXMLCore::fastMallocRegisterThread):

15928

(KXMLCore::TCMalloc_ThreadCache::GetCache):

15929

(KXMLCore::TCMalloc_ThreadCache::GetCacheIfPresent):

15930

(KXMLCore::TCMalloc_ThreadCache::CreateCacheIfNecessary):

15931

(KXMLCore::do_malloc):

15932

(KXMLCore::do_free):

15933

(KXMLCore::realloc):

15934

* kxmlcore/FastMalloc.h:

15935

(operator new):

15936

(operator delete):

15937

(operator new[]):

15938

(operator delete[]):

15939

* kxmlcore/HashTable.cpp:

15940

* kxmlcore/TCSpinLock.h:

15941

(TCMalloc_SpinLock::Lock):

15942

(TCMalloc_SpinLock::Unlock):

15943

(TCMalloc_SlowLock):

15944

* kxmlcore/TCSystemAlloc.cpp:

15945

15946

2005-09-30 Geoffrey Garen <ggaren@apple.com>

15947

15948

- Second cut at fixing <rdar://problem/4275206> Denver Regression: Seed:

15949

Past Editions of Opinions display "NAN/Undefined" for www.washingtonpost.com

15950

15951

Reviewed by john.

15952

15953

* kjs/date_object.cpp:

15954

(KJS::KRFCDate_parseDate): Intead of creating a timezone when one isn't specified,

15955

just rely on the fallback logic, which will do it for you. Also, return invalidDate

15956

if the date includes trailing garbage. (Somewhat accidentally, the timezone logic

15957

used to catch trailing garbage.)

15958

15959

Added test case to fast/js/date-parse-test.html.

15960

15961

2005-09-29 Eric Seidel <eseidel@apple.com>

15962

Fix from Mitz Pettel <opendarwin.org@mitzpettel.com>

15963

15964

Reviewed by darin.

15965

15966

Fix JSC memory smasher in TOT.

15967

http://bugs.webkit.org/show_bug.cgi?id=5176

15968

15969

* pcre/pcre_exec.c:

15970

(match):

15971

15972

2005-09-29 Eric Seidel <eseidel@apple.com>

15973

Fix from Mitz Pettel <opendarwin.org@mitzpettel.com>

15974

15975

Reviewed by mjs.

15976

15977

* JavaScriptCore.xcodeproj/project.pbxproj:

15978

Build fix for JSC+SVG after 5161.

15979

http://bugs.webkit.org/show_bug.cgi?id=5179

15980

15981

2005-09-28 Geoffrey Garen <ggaren@apple.com>

15982

15983

- Fixed <rdar://problem/4275206> Denver Regression: Seed: Past Editions of Opinions display

15984

"NAN/Undefined" for www.washingtonpost.com

15985

15986

Reviewed by darin.

15987

15988

* kjs/date_object.cpp:

15989

(KJS::KRFCDate_parseDate): If the timezone isn't specified, rather than returning

15990

invalidDate, substitute the local timezone. This matches the behavior of FF/IE.

15991

15992

2005-09-28 Maciej Stachowiak <mjs@apple.com>

15993

15994

Patch from George Staikos, reviewed by me.

15995

15996

- fixed some compile issues on Linux

15997

15998

* kjs/property_slot.h:

15999

* kjs/simple_number.h:

16000

16001

2005-09-27 Maciej Stachowiak <mjs@apple.com>

16002

16003

Reviewed by Eric.

16004

16005

- move HashMap/HashSet code down to JavaScriptCore

16006

http://bugs.webkit.org/show_bug.cgi?id=5161

16007

16008

* JavaScriptCore.xcodeproj/project.pbxproj:

16009

* kjs/internal.cpp:

16010

(KJS::interpreterMap): Function that fetches the interpreter map on demand.

16011

(KJS::InterpreterImp::InterpreterImp): Replace use of InterpreterMap

16012

class with an appropriate HashMap.

16013

(KJS::InterpreterImp::clear): ditto

16014

(KJS::InterpreterImp::interpreterWithGlobalObject): ditto

16015

* kjs/interpreter_map.cpp: Removed.

16016

* kjs/interpreter_map.h: Removed.

16017

16018

The HashMap/HashSet code (copied and slightly tweaked from WebCore)

16019

16020

* kxmlcore/HashFunctions.h: Added.

16021

(KXMLCore::4):

16022

(KXMLCore::8):

16023

(KXMLCore::):

16024

(KXMLCore::PointerHash::hash):

16025

(KXMLCore::PointerHash::equal):

16026

* kxmlcore/HashMap.h: Added.

16027

(KXMLCore::extractFirst):

16028

(KXMLCore::HashMap::HashMap):

16029

(KXMLCore::::size):

16030

(KXMLCore::::capacity):

16031

(KXMLCore::::isEmpty):

16032

(KXMLCore::::begin):

16033

(KXMLCore::::end):

16034

(KXMLCore::::find):

16035

(KXMLCore::::contains):

16036

(KXMLCore::::set):

16037

(KXMLCore::::get):

16038

(KXMLCore::::remove):

16039

(KXMLCore::::clear):

16040

(KXMLCore::deleteAllValues):

16041

* kxmlcore/HashMapPtrSpec.h: Added.

16042

(KXMLCore::PointerHashIteratorAdapter::PointerHashIteratorAdapter):

16043

(KXMLCore::PointerHashIteratorAdapter::operator*):

16044

(KXMLCore::PointerHashIteratorAdapter::operator->):

16045

(KXMLCore::PointerHashIteratorAdapter::operator++):

16046

(KXMLCore::PointerHashIteratorAdapter::operator==):

16047

(KXMLCore::PointerHashIteratorAdapter::operator!=):

16048

(KXMLCore::PointerHashConstIteratorAdapter::PointerHashConstIteratorAdapter):

16049

(KXMLCore::PointerHashConstIteratorAdapter::operator*):

16050

(KXMLCore::PointerHashConstIteratorAdapter::operator->):

16051

(KXMLCore::PointerHashConstIteratorAdapter::operator++):

16052

(KXMLCore::PointerHashConstIteratorAdapter::operator==):

16053

(KXMLCore::PointerHashConstIteratorAdapter::operator!=):

16054

(KXMLCore::):

16055

* kxmlcore/HashSet.h: Added.

16056

(KXMLCore::identityExtract):

16057

(KXMLCore::convertAdapter):

16058

(KXMLCore::HashSet::HashSet):

16059

(KXMLCore::::size):

16060

(KXMLCore::::capacity):

16061

(KXMLCore::::isEmpty):

16062

(KXMLCore::::begin):

16063

(KXMLCore::::end):

16064

(KXMLCore::::find):

16065

(KXMLCore::::contains):

16066

(KXMLCore::::insert):

16067

(KXMLCore::::remove):

16068

(KXMLCore::::clear):

16069

* kxmlcore/HashTable.cpp: Added.

16070

(KXMLCore::HashTableStats::~HashTableStats):

16071

(KXMLCore::HashTableStats::recordCollisionAtCount):

16072

* kxmlcore/HashTable.h: Added.

16073

(KXMLCore::HashTableIterator::skipEmptyBuckets):

16074

(KXMLCore::HashTableIterator::HashTableIterator):

16075

(KXMLCore::HashTableIterator::operator*):

16076

(KXMLCore::HashTableIterator::operator->):

16077

(KXMLCore::HashTableIterator::operator++):

16078

(KXMLCore::HashTableIterator::operator==):

16079

(KXMLCore::HashTableIterator::operator!=):

16080

(KXMLCore::HashTableConstIterator::HashTableConstIterator):

16081

(KXMLCore::HashTableConstIterator::operator*):

16082

(KXMLCore::HashTableConstIterator::operator->):

16083

(KXMLCore::HashTableConstIterator::skipEmptyBuckets):

16084

(KXMLCore::HashTableConstIterator::operator++):

16085

(KXMLCore::HashTableConstIterator::operator==):

16086

(KXMLCore::HashTableConstIterator::operator!=):

16087

(KXMLCore::HashTable::HashTable):

16088

(KXMLCore::HashTable::~HashTable):

16089

(KXMLCore::HashTable::begin):

16090

(KXMLCore::HashTable::end):

16091

(KXMLCore::HashTable::size):

16092

(KXMLCore::HashTable::capacity):

16093

(KXMLCore::HashTable::insert):

16094

(KXMLCore::HashTable::isEmptyBucket):

16095

(KXMLCore::HashTable::isDeletedBucket):

16096

(KXMLCore::HashTable::isEmptyOrDeletedBucket):

16097

(KXMLCore::HashTable::hash):

16098

(KXMLCore::HashTable::equal):

16099

(KXMLCore::HashTable::identityConvert):

16100

(KXMLCore::HashTable::extractKey):

16101

(KXMLCore::HashTable::lookup):

16102

(KXMLCore::HashTable::shouldExpand):

16103

(KXMLCore::HashTable::mustRehashInPlace):

16104

(KXMLCore::HashTable::shouldShrink):

16105

(KXMLCore::HashTable::shrink):

16106

(KXMLCore::HashTable::clearBucket):

16107

(KXMLCore::HashTable::deleteBucket):

16108

(KXMLCore::HashTable::makeLookupResult):

16109

(KXMLCore::HashTable::makeIterator):

16110

(KXMLCore::HashTable::makeConstIterator):

16111

(KXMLCore::::lookup):

16112

(KXMLCore::::insert):

16113

(KXMLCore::::reinsert):

16114

(KXMLCore::::find):

16115

(KXMLCore::::contains):

16116

(KXMLCore::::remove):

16117

(KXMLCore::::allocateTable):

16118

(KXMLCore::::expand):

16119

(KXMLCore::::rehash):

16120

(KXMLCore::::clear):

16121

(KXMLCore::::HashTable):

16122

(KXMLCore::::swap):

16123

(KXMLCore::::operator):

16124

(KXMLCore::::checkTableConsistency):

16125

(KXMLCore::::checkTableConsistencyExceptSize):

16126

* kxmlcore/HashTraits.h: Added.

16127

(KXMLCore::HashTraits::emptyValue):

16128

(KXMLCore::):

16129

(KXMLCore::PairHashTraits::emptyValue):

16130

(KXMLCore::PairHashTraits::deletedValue):

16131

16132

2005-09-27 Darin Adler <darin@apple.com>

16133

16134

Reviewed by Maciej.

16135

16136

- update grammar to fix conflicts; fixes one of our test cases

16137

because it resolves the relationship between function expressions

16138

and declarations in the way required by the ECMA specification

16139

16140

* kjs/grammar.y: Added lots of new grammar rules so we have no conflicts.

16141

A new set of rules for "no bracket or function at start of expression" and

16142

another set of rules for "no in anywhere in expression". Also simplified the

16143

handling of try to use only a single node and used operator precedence to

16144

get rid of the conflict in handling of if and else. Also used a macro to

16145

streamline the handling of automatic semicolons and changed parenthesis

16146

handling to use a virtual function.

16147

16148

* kjs/nodes.h: Added nodeInsideAllParens, removed unused abortStatement.

16149

(KJS::TryNode::TryNode): Updated to hold catch and finally blocks directly instead

16150

of using a special node for each.

16151

* kjs/nodes.cpp:

16152

(Node::createErrorCompletion): Added. Used instead of throwError when creating errors

16153

that should not be in a completion rather than an ExecState.

16154

(Node::throwUndefinedVariableError): Added. Sets source location unlike the call it

16155

replaces.

16156

(Node::nodeInsideAllParens): Added.

16157

(GroupNode::nodeInsideAllParens): Added.

16158

(StatListNode::execute): Removed code to move exceptions into completion objects;

16159

that's now done solely by the KJS_CHECKEXCEPTION macro.

16160

(TryNode::execute): Include execution of catch and finally here rather than using

16161

separate nodes.

16162

(FuncDeclNode::execute): Moved here, no longer inline.

16163

* kjs/nodes2string.cpp:

16164

(TryNode::streamTo): Updated for change.

16165

(FuncDeclNode::streamTo): Ditto.

16166

(FuncExprNode::streamTo): Ditto.

16167

16168

* kjs/kjs-test: Removed. Was part of "make check".

16169

* kjs/kjs-test.chk: Ditto.

16170

* kjs/test.js: Ditto.

16171

16172

* tests/mozilla/expected.html: Updated because one more test succeeds.

16173

16174

2005-09-27 Adele Peterson <adele@apple.com>

16175

16176

Reviewed by Maciej.

16177

16178

Changed ints to size_t where appropriate.

16179

16180

* kjs/collector.cpp:

16181

(KJS::Collector::allocate):

16182

(KJS::Collector::markStackObjectsConservatively):

16183

(KJS::Collector::collect):

16184

(KJS::Collector::size):

16185

(KJS::Collector::numInterpreters):

16186

(KJS::Collector::numGCNotAllowedObjects):

16187

(KJS::Collector::numReferencedObjects):

16188

* kjs/collector.h:

16189

16190

2005-09-27 Eric Seidel <eseidel@apple.com>

16191

16192

Reviewed by kevin.

16193

16194

* JavaScriptCore.xcodeproj/project.pbxproj: fix after malloc changes.

16195

16196

2005-09-27 Eric Seidel <eseidel@apple.com>

16197

16198

Reviewed by mjs.

16199

16200

* kjs/nodes.cpp:

16201

(FuncExprNode::evaluate): Now sets .constructor properly.

16202

Test cases added to WebCore/layout-tests.

16203

http://bugs.webkit.org/show_bug.cgi?id=3537

16204

16205

2005-09-26 Maciej Stachowiak <mjs@apple.com>

16206

16207

Reviewed by John.

16208

16209

- replace dlmalloc with tcmalloc

16210

http://bugs.webkit.org/show_bug.cgi?id=5145

16211

16212

I also moved SharedPtr and the assertion code from WebCore into a

16213

new kxmlcore directory.

16214

16215

* JavaScriptCore.xcodeproj/project.pbxproj:

16216

* kjs/collector.cpp:

16217

(KJS::Collector::allocate):

16218

(KJS::Collector::collect):

16219

* kjs/config.h:

16220

* kjs/fast_malloc.cpp: Removed.

16221

* kjs/fast_malloc.h: Removed.

16222

* kjs/function.cpp:

16223

* kjs/function.h:

16224

* kjs/function_object.cpp:

16225

* kjs/identifier.cpp:

16226

(KJS::Identifier::add):

16227

* kjs/internal.cpp:

16228

* kjs/internal.h:

16229

* kjs/nodes.h:

16230

* kjs/nodes2string.cpp:

16231

* kjs/property_map.cpp:

16232

(KJS::PropertyMap::~PropertyMap):

16233

(KJS::PropertyMap::rehash):

16234

* kjs/scope_chain.h:

16235

* kjs/shared_ptr.h: Removed.

16236

* kjs/string_object.cpp:

16237

(StringObjectFuncImp::callAsFunction):

16238

* kjs/ustring.cpp:

16239

(KJS::UString::Rep::createCopying):

16240

(KJS::UString::Rep::destroy):

16241

(KJS::UString::expandCapacity):

16242

(KJS::UString::expandPreCapacity):

16243

(KJS::UString::UString):

16244

(KJS::UString::spliceSubstringsWithSeparators):

16245

(KJS::UString::append):

16246

(KJS::UString::operator=):

16247

(KJS::UString::detach):

16248

* kjs/ustring.h:

16249

* kxmlcore/Assertions.h: Added.

16250

* kxmlcore/Assertions.mm: Added.

16251

* kxmlcore/FastMalloc.cpp: Added.

16252

(KXMLCore::LgFloor):

16253

(KXMLCore::SizeClass):

16254

(KXMLCore::ByteSizeForClass):

16255

(KXMLCore::InitSizeClasses):

16256

(KXMLCore::MetaDataAlloc):

16257

(KXMLCore::PageHeapAllocator::Init):

16258

(KXMLCore::PageHeapAllocator::New):

16259

(KXMLCore::PageHeapAllocator::Delete):

16260

(KXMLCore::PageHeapAllocator::inuse):

16261

(KXMLCore::pages):

16262

(KXMLCore::AllocationSize):

16263

(KXMLCore::Event):

16264

(KXMLCore::NewSpan):

16265

(KXMLCore::DeleteSpan):

16266

(KXMLCore::DLL_Init):

16267

(KXMLCore::DLL_Remove):

16268

(KXMLCore::DLL_IsEmpty):

16269

(KXMLCore::DLL_Length):

16270

(KXMLCore::DLL_Print):

16271

(KXMLCore::DLL_Prepend):

16272

(KXMLCore::DLL_InsertOrdered):

16273

(KXMLCore::):

16274

(KXMLCore::TCMalloc_PageHeap::GetDescriptor):

16275

(KXMLCore::TCMalloc_PageHeap::SystemBytes):

16276

(KXMLCore::TCMalloc_PageHeap::FreeBytes):

16277

(KXMLCore::TCMalloc_PageHeap::RecordSpan):

16278

(KXMLCore::TCMalloc_PageHeap::TCMalloc_PageHeap):

16279

(KXMLCore::TCMalloc_PageHeap::New):

16280

(KXMLCore::TCMalloc_PageHeap::Split):

16281

(KXMLCore::TCMalloc_PageHeap::Carve):

16282

(KXMLCore::TCMalloc_PageHeap::Delete):

16283

(KXMLCore::TCMalloc_PageHeap::RegisterSizeClass):

16284

(KXMLCore::TCMalloc_PageHeap::Dump):

16285

(KXMLCore::TCMalloc_PageHeap::GrowHeap):

16286

(KXMLCore::TCMalloc_PageHeap::Check):

16287

(KXMLCore::TCMalloc_PageHeap::CheckList):

16288

(KXMLCore::TCMalloc_ThreadCache_FreeList::Init):

16289

(KXMLCore::TCMalloc_ThreadCache_FreeList::length):

16290

(KXMLCore::TCMalloc_ThreadCache_FreeList::empty):

16291

(KXMLCore::TCMalloc_ThreadCache_FreeList::lowwatermark):

16292

(KXMLCore::TCMalloc_ThreadCache_FreeList::clear_lowwatermark):

16293

(KXMLCore::TCMalloc_ThreadCache_FreeList::Push):

16294

(KXMLCore::TCMalloc_ThreadCache_FreeList::Pop):

16295

(KXMLCore::TCMalloc_ThreadCache::freelist_length):

16296

(KXMLCore::TCMalloc_ThreadCache::Size):

16297

(KXMLCore::TCMalloc_Central_FreeList::length):

16298

(KXMLCore::TCMalloc_Central_FreeList::Init):

16299

(KXMLCore::TCMalloc_Central_FreeList::Insert):

16300

(KXMLCore::TCMalloc_Central_FreeList::Remove):

16301

(KXMLCore::TCMalloc_Central_FreeList::Populate):

16302

(KXMLCore::TCMalloc_ThreadCache::SampleAllocation):

16303

(KXMLCore::TCMalloc_ThreadCache::Init):

16304

(KXMLCore::TCMalloc_ThreadCache::Cleanup):

16305

(KXMLCore::TCMalloc_ThreadCache::Allocate):

16306

(KXMLCore::TCMalloc_ThreadCache::Deallocate):

16307

(KXMLCore::TCMalloc_ThreadCache::FetchFromCentralCache):

16308

(KXMLCore::TCMalloc_ThreadCache::ReleaseToCentralCache):

16309

(KXMLCore::TCMalloc_ThreadCache::Scavenge):

16310

(KXMLCore::TCMalloc_ThreadCache::GetCache):

16311

(KXMLCore::TCMalloc_ThreadCache::GetCacheIfPresent):

16312

(KXMLCore::TCMalloc_ThreadCache::PickNextSample):

16313

(KXMLCore::TCMalloc_ThreadCache::InitModule):

16314

(KXMLCore::TCMalloc_ThreadCache::InitTSD):

16315

(KXMLCore::TCMalloc_ThreadCache::CreateCacheIfNecessary):

16316

(KXMLCore::TCMalloc_ThreadCache::DeleteCache):

16317

(KXMLCore::TCMalloc_ThreadCache::RecomputeThreadCacheSize):

16318

(KXMLCore::TCMalloc_ThreadCache::Print):

16319

(KXMLCore::ExtractStats):

16320

(KXMLCore::DumpStats):

16321

(KXMLCore::PrintStats):

16322

(KXMLCore::DumpStackTraces):

16323

(KXMLCore::TCMallocImplementation::GetStats):

16324

(KXMLCore::TCMallocImplementation::ReadStackTraces):

16325

(KXMLCore::TCMallocImplementation::GetNumericProperty):

16326

(KXMLCore::TCMallocImplementation::SetNumericProperty):

16327

(KXMLCore::DoSampledAllocation):

16328

(KXMLCore::do_malloc):

16329

(KXMLCore::do_free):

16330

(KXMLCore::do_memalign):

16331

(KXMLCore::TCMallocGuard::TCMallocGuard):

16332

(KXMLCore::TCMallocGuard::~TCMallocGuard):

16333

(KXMLCore::malloc):

16334

(KXMLCore::free):

16335

(KXMLCore::calloc):

16336

(KXMLCore::cfree):

16337

(KXMLCore::realloc):

16338

(KXMLCore::memalign):

16339

(KXMLCore::posix_memalign):

16340

(KXMLCore::valloc):

16341

(KXMLCore::pvalloc):

16342

(KXMLCore::malloc_stats):

16343

(KXMLCore::mallopt):

16344

(KXMLCore::mallinfo):

16345

* kxmlcore/FastMalloc.h: Added.

16346

(KXMLCore::FastAllocated::operator new):

16347

(KXMLCore::FastAllocated::operator delete):

16348

(KXMLCore::FastAllocated::operator new[]):

16349

(KXMLCore::FastAllocated::operator delete[]):

16350

* kxmlcore/SharedPtr.h: Added.

16351

(KXMLCore::SharedPtr::SharedPtr):

16352

(KXMLCore::SharedPtr::~SharedPtr):

16353

(KXMLCore::SharedPtr::isNull):

16354

(KXMLCore::SharedPtr::notNull):

16355

(KXMLCore::SharedPtr::reset):

16356

(KXMLCore::SharedPtr::get):

16357

(KXMLCore::SharedPtr::operator*):

16358

(KXMLCore::SharedPtr::operator->):

16359

(KXMLCore::SharedPtr::operator!):

16360

(KXMLCore::SharedPtr::operator bool):

16361

(KXMLCore::::operator):

16362

(KXMLCore::operator==):

16363

(KXMLCore::operator!=):

16364

(KXMLCore::static_pointer_cast):

16365

(KXMLCore::const_pointer_cast):

16366

* kxmlcore/TCPageMap.h: Added.

16367

(TCMalloc_PageMap1::TCMalloc_PageMap1):

16368

(TCMalloc_PageMap1::Ensure):

16369

(TCMalloc_PageMap1::get):

16370

(TCMalloc_PageMap1::set):

16371

(TCMalloc_PageMap2::TCMalloc_PageMap2):

16372

(TCMalloc_PageMap2::get):

16373

(TCMalloc_PageMap2::set):

16374

(TCMalloc_PageMap2::Ensure):

16375

(TCMalloc_PageMap3::NewNode):

16376

(TCMalloc_PageMap3::TCMalloc_PageMap3):

16377

(TCMalloc_PageMap3::get):

16378

(TCMalloc_PageMap3::set):

16379

(TCMalloc_PageMap3::Ensure):

16380

* kxmlcore/TCSpinLock.h: Added.

16381

(TCMalloc_SpinLock::Init):

16382

(TCMalloc_SpinLock::Finalize):

16383

(TCMalloc_SpinLock::Lock):

16384

(TCMalloc_SpinLock::Unlock):

16385

(TCMalloc_SlowLock):

16386

(TCMalloc_SpinLockHolder::TCMalloc_SpinLockHolder):

16387

(TCMalloc_SpinLockHolder::~TCMalloc_SpinLockHolder):

16388

* kxmlcore/TCSystemAlloc.cpp: Added.

16389

(TrySbrk):

16390

(TryMmap):

16391

(TryDevMem):

16392

(TCMalloc_SystemAlloc):

16393

* kxmlcore/TCSystemAlloc.h: Added.

16394

16395

2005-09-23 Maciej Stachowiak <mjs@apple.com>

16396

16397

Reviewed by Darin.

16398

16399

<rdar://problem/4260479> Finish deploying PropertySlot in the interpreter

16400

http://bugs.webkit.org/show_bug.cgi?id=5112

16401

16402

Convert postfix, prefix, delete, prefix, and for..in expressions to use

16403

PropertySlot-based lookup instead of evaluateReference.

16404

16405

3% speedup on JS iBench.

16406

16407

Fixed two of the JS tests:

16408

* tests/mozilla/expected.html:

16409

16410

* kjs/grammar.y:

16411

* kjs/nodes.cpp:

16412

(PostfixResolveNode::evaluate):

16413

(PostfixBracketNode::evaluate):

16414

(PostfixDotNode::evaluate):

16415

(DeleteResolveNode::evaluate):

16416

(DeleteBracketNode::evaluate):

16417

(DeleteDotNode::evaluate):

16418

(DeleteValueNode::evaluate):

16419

(typeStringForValue):

16420

(TypeOfResolveNode::evaluate):

16421

(TypeOfValueNode::evaluate):

16422

(PrefixResolveNode::evaluate):

16423

(PrefixBracketNode::evaluate):

16424

(PrefixDotNode::evaluate):

16425

(ForInNode::execute):

16426

* kjs/nodes.h:

16427

(KJS::PostfixResolveNode::PostfixResolveNode):

16428

(KJS::PostfixBracketNode::PostfixBracketNode):

16429

(KJS::PostfixDotNode::PostfixDotNode):

16430

(KJS::DeleteResolveNode::DeleteResolveNode):

16431

(KJS::DeleteBracketNode::DeleteBracketNode):

16432

(KJS::DeleteDotNode::DeleteDotNode):

16433

(KJS::DeleteValueNode::DeleteValueNode):

16434

(KJS::TypeOfResolveNode::TypeOfResolveNode):

16435

(KJS::TypeOfValueNode::TypeOfValueNode):

16436

(KJS::PrefixResolveNode::PrefixResolveNode):

16437

(KJS::PrefixBracketNode::PrefixBracketNode):

16438

(KJS::PrefixDotNode::PrefixDotNode):

16439

* kjs/nodes2string.cpp:

16440

(PostfixResolveNode::streamTo):

16441

(PostfixBracketNode::streamTo):

16442

(PostfixDotNode::streamTo):

16443

(DeleteResolveNode::streamTo):

16444

(DeleteBracketNode::streamTo):

16445

(DeleteDotNode::streamTo):

16446

(DeleteValueNode::streamTo):

16447

(TypeOfValueNode::streamTo):

16448

(TypeOfResolveNode::streamTo):

16449

(PrefixResolveNode::streamTo):

16450

(PrefixBracketNode::streamTo):

16451

(PrefixDotNode::streamTo):

16452

* kjs/reference.cpp:

16453

(KJS::Reference::Reference):

16454

(KJS::Reference::getPropertyName):

16455

(KJS::Reference::getValue):

16456

(KJS::Reference::deleteValue):

16457

* kjs/reference.h:

16458

16459

2005-09-23 Krzysztof Kowalczyk <kkowalczyk@gmail.com>

16460

16461

Reviewed and landed by Darin.

16462

16463

- a Windows-specific file

16464

16465

* os-win32/stdint.h: Added. We plan to remove dependency on the <stdint.h> types,

16466

and if we do so, we will remove this file.

16467

16468

2005-09-22 Geoffrey Garen <ggaren@apple.com>

16469

16470

- Fixed http://bugs.webkit.org/show_bug.cgi?id=5053

16471

Need to restore int/long changes to simple_number.h

16472

16473

Reviewed by darin and mjs.

16474

16475

* kjs/simple_number.h: changed enums to indenpendent constants to clarify types

16476

(KJS::isNegativeZero): changed to static function - no reason to export

16477

(KJS::SimpleNumber::rightShiftSignExtended): new function for clarity

16478

(KJS::SimpleNumber::make): specified cast as reinterpret_cast

16479

(KJS::SimpleNumber::is): changed to use uintptr_t for portability

16480

(KJS::SimpleNumber::value): changed to use uintptr_t and rightShiftSignExtended

16481

(KJS::SimpleNumber::fits): inverted tests - probably only a performance win for double

16482

(KJS::SimpleNumber::integerFits): ditto

16483

16484

2005-09-20 Maciej Stachowiak <mjs@apple.com>

16485

16486

Reviewed by Geoff and partly by Darin.

16487

16488

- fixed http://bugs.webkit.org/post_bug.cgi

16489

(Reduce conflicts in JavaScriptCore grammar)

16490

16491

This change gets us down from over 200 shift/reduce and 45 reduce/reduce to

16492

9 shift/reduce and 45 reduce/reduce.

16493

16494

* kjs/grammar.y:

16495

* kjs/grammar_types.h: Removed.

16496

* kjs/lexer.cpp:

16497

* kjs/nodes.h:

16498

(KJS::Node::isGroupNode):

16499

(KJS::Node::isLocation):

16500

(KJS::Node::isResolveNode):

16501

(KJS::Node::isBracketAccessorNode):

16502

(KJS::Node::isDotAccessorNode):

16503

(KJS::ResolveNode::isLocation):

16504

(KJS::ResolveNode::isResolveNode):

16505

(KJS::ResolveNode::identifier):

16506

(KJS::GroupNode::isGroupNode):

16507

(KJS::GroupNode::leafNode):

16508

(KJS::BracketAccessorNode::isLocation):

16509

(KJS::BracketAccessorNode::isBracketAccessorNode):

16510

(KJS::BracketAccessorNode::base):

16511

(KJS::BracketAccessorNode::subscript):

16512

(KJS::DotAccessorNode::isLocation):

16513

(KJS::DotAccessorNode::isDotAccessorNode):

16514

(KJS::DotAccessorNode::base):

16515

(KJS::DotAccessorNode::identifier):

16516

(KJS::FuncExprNode::FuncExprNode):

16517

(KJS::FuncExprNode::identifier):

16518

(KJS::FuncDeclNode::FuncDeclNode):

16519

(KJS::FuncDeclNode::execute):

16520

16521

2005-09-20 Geoffrey Garen <ggaren@apple.com>

16522

16523

- Oops. The 4263434 change was only appropriate on the branch. Rolling out.

16524

16525

Reviewed by eric.

16526

16527

* kjs/internal.cpp:

16528

(KJS::InterpreterImp::mark):

16529

16530

2005-09-20 Geoffrey Garen <ggaren@apple.com>

16531

16532

- More changes needed to fix <rdar://problem/4214783> 8F29 REGRESSION(Denver/Chardonnay):

16533

kjs_fast_malloc crash due to lack of locking on multiple threads (seen selecting volumes in

16534

the installer)

16535

16536

Added InterpreterLocks in some places in the bindings we missed before.

16537

16538

Reviewed by john.

16539

16540

* bindings/runtime_root.cpp:

16541

(KJS::Bindings::addNativeReference):

16542

(KJS::Bindings::removeNativeReference):

16543

(RootObject::removeAllNativeReferences):

16544

* bindings/runtime_root.h:

16545

(KJS::Bindings::RootObject::~RootObject):

16546

(KJS::Bindings::RootObject::setRootObjectImp):

16547

16548

2005-09-20 Geoffrey Garen <ggaren@apple.com>

16549

16550

- Fixed <rdar://problem/4263434> <rdar://problem/4263434> Denver 8F29 Regression:

16551

KJS::InterpreterImp::mark() crash

16552

16553

Fix by mjs, review by me.

16554

16555

* kjs/internal.cpp:

16556

(KJS::InterpreterImp::mark): Added a null check on globExec in case a

16557

garbage collection occurs inside InterpreterImp::globalInit (called

16558

from InterpreterImp::InterpreterImp), at which point globExec has not yet been initialized.

16559

16560

2005-09-20 Geoffrey Garen <ggaren@apple.com>

16561

16562

- Rolled in fix for http://bugs.webkit.org/show_bug.cgi?id=4892

16563

Date constructor has problems with months larger than 11

16564

16565

Test cases added:

16566

16567

* layout-tests/fast/js/date-big-constructor-expected.txt: Added.

16568

* layout-tests/fast/js/date-big-constructor.html: Added.

16569

16570

Reviewed by darin.

16571

16572

* kjs/date_object.cpp:

16573

(KJS::fillStructuresUsingDateArgs):

16574

(KJS::makeTime):

16575

16576

2005-09-19 Geoffrey Garen <ggaren@apple.com>

16577

16578

- Fixed http://bugs.webkit.org/show_bug.cgi?id=5028

16579

9 layout tests fail following the change from long to int

16580

16581

- Rolled out changes to simple_number.h, and added fits(long long)

16582

and SimpleNumber::fits(unsigned long long) to the old system.

16583

16584

Reviewed by mjs.

16585

16586

* kjs/simple_number.h:

16587

(KJS::SimpleNumber::):

16588

(KJS::SimpleNumber::value):

16589

(KJS::SimpleNumber::fits):

16590

(KJS::SimpleNumber::integerFits):

16591

(KJS::SimpleNumber::make):

16592

16593

2005-09-14 Maciej Stachowiak <mjs@apple.com>

16594

16595

Reviewed by Geoff.

16596

16597

- fixed <rdar://problem/4214783> REGRESSION: kjs_fast_malloc crash due to lack of locking on multiple threads (seen selecting volumes in the installer)

16598

16599

Make sure to lock using the InterpreterLock class in all places that need it

16600

(including anything that uses the collector, the parser, the protect count hash table,

16601

and anything that allocates via fast_malloc).

16602

16603

Also added assertions to ensure that the locking rules are followed for the relevant

16604

resources.

16605

16606

* Makefile.am:

16607

* bindings/NP_jsobject.cpp:

16608

(identifierFromNPIdentifier):

16609

(_NPN_Invoke):

16610

(_NPN_Evaluate):

16611

(_NPN_GetProperty):

16612

(_NPN_SetProperty):

16613

(_NPN_RemoveProperty):

16614

(_NPN_HasProperty):

16615

(_NPN_HasMethod):

16616

(_NPN_SetException):

16617

* bindings/jni/jni_jsobject.cpp:

16618

(JSObject::call):

16619

(JSObject::eval):

16620

(JSObject::getMember):

16621

(JSObject::setMember):

16622

(JSObject::removeMember):

16623

(JSObject::getSlot):

16624

(JSObject::setSlot):

16625

(JSObject::toString):

16626

(JSObject::convertJObjectToValue):

16627

* bindings/objc/WebScriptObject.mm:

16628

(-[WebScriptObject callWebScriptMethod:withArguments:]):

16629

(-[WebScriptObject evaluateWebScript:]):

16630

(-[WebScriptObject setValue:forKey:]):

16631

(-[WebScriptObject valueForKey:]):

16632

(-[WebScriptObject removeWebScriptKey:]):

16633

(-[WebScriptObject stringRepresentation]):

16634

(-[WebScriptObject webScriptValueAtIndex:]):

16635

(-[WebScriptObject setWebScriptValueAtIndex:value:]):

16636

(+[WebScriptObject _convertValueToObjcValue:KJS::originExecutionContext:Bindings::executionContext:Bindings::]):

16637

* bindings/runtime.cpp:

16638

(Instance::createRuntimeObject):

16639

* bindings/runtime_root.h:

16640

* bindings/testbindings.cpp:

16641

(main):

16642

* bindings/testbindings.mm:

16643

(main):

16644

* kjs/fast_malloc.cpp:

16645

(KJS::kjs_fast_malloc):

16646

(KJS::kjs_fast_calloc):

16647

(KJS::kjs_fast_free):

16648

(KJS::kjs_fast_realloc):

16649

* kjs/fast_malloc.h:

16650

* kjs/identifier.h:

16651

* kjs/internal.cpp:

16652

(InterpreterImp::InterpreterImp):

16653

(InterpreterImp::clear):

16654

(InterpreterImp::mark):

16655

(InterpreterImp::checkSyntax):

16656

(InterpreterImp::evaluate):

16657

* kjs/internal.h:

16658

(KJS::InterpreterImp::globalObject):

16659

* kjs/interpreter.cpp:

16660

(Interpreter::evaluate):

16661

* kjs/interpreter.h:

16662

(KJS::InterpreterLock::InterpreterLock):

16663

(KJS::InterpreterLock::~InterpreterLock):

16664

* kjs/nodes.h:

16665

* kjs/protect.h:

16666

(KJS::ProtectedValue::ProtectedValue):

16667

(KJS::ProtectedValue::~ProtectedValue):

16668

(KJS::ProtectedValue::operator=):

16669

(KJS::ProtectedObject::ProtectedObject):

16670

(KJS::ProtectedObject::~ProtectedObject):

16671

(KJS::ProtectedObject::operator=):

16672

(KJS::ProtectedReference::ProtectedReference):

16673

(KJS::ProtectedReference::~ProtectedReference):

16674

(KJS::ProtectedReference::operator=):

16675

* kjs/protected_object.h:

16676

* kjs/protected_values.cpp:

16677

(KJS::ProtectedValues::getProtectCount):

16678

(KJS::ProtectedValues::increaseProtectCount):

16679

(KJS::ProtectedValues::decreaseProtectCount):

16680

* kjs/string_object.cpp:

16681

(StringObjectImp::StringObjectImp):

16682

* kjs/testkjs.cpp:

16683

(main):

16684

16685

2005-09-16 Adele Peterson <adele@apple.com>

16686

16687

Change by Darin, reviewed by me and Maciej.

16688

16689

Fixes http://bugs.webkit.org/show_bug.cgi?id=4547

16690

use int instead of long for 32-bit (to prepare for LP64 compiling)

16691

16692

* bindings/c/c_class.h:

16693

(KJS::Bindings::CClass::constructorAt):

16694

(KJS::Bindings::CClass::numConstructors):

16695

* bindings/c/c_runtime.h:

16696

(KJS::Bindings::CMethod::numParameters):

16697

* bindings/jni/jni_class.cpp:

16698

(JavaClass::JavaClass):

16699

* bindings/jni/jni_class.h:

16700

(KJS::Bindings::JavaClass::constructorAt):

16701

(KJS::Bindings::JavaClass::numConstructors):

16702

* bindings/jni/jni_instance.cpp:

16703

(JavaInstance::invokeMethod):

16704

* bindings/jni/jni_jsobject.cpp:

16705

(JSObject::convertJObjectToValue):

16706

(JSObject::listFromJArray):

16707

* bindings/jni/jni_runtime.cpp:

16708

(JavaMethod::JavaMethod):

16709

* bindings/jni/jni_runtime.h:

16710

(KJS::Bindings::JavaConstructor::_commonCopy):

16711

(KJS::Bindings::JavaConstructor::parameterAt):

16712

(KJS::Bindings::JavaConstructor::numParameters):

16713

(KJS::Bindings::JavaMethod::_commonCopy):

16714

(KJS::Bindings::JavaMethod::parameterAt):

16715

(KJS::Bindings::JavaMethod::numParameters):

16716

* bindings/npapi.h:

16717

* bindings/objc/WebScriptObject.mm:

16718

(listFromNSArray):

16719

* bindings/objc/objc_class.h:

16720

(KJS::Bindings::ObjcClass::constructorAt):

16721

(KJS::Bindings::ObjcClass::numConstructors):

16722

* bindings/objc/objc_instance.h:

16723

* bindings/objc/objc_runtime.h:

16724

* bindings/objc/objc_runtime.mm:

16725

(ObjcMethod::numParameters):

16726

* bindings/runtime.h:

16727

* kjs/identifier.h:

16728

* kjs/internal.h:

16729

* kjs/property_slot.h:

16730

(KJS::PropertySlot::setCustomIndex):

16731

(KJS::PropertySlot::index):

16732

(KJS::PropertySlot::):

16733

* kjs/regexp_object.cpp:

16734

(RegExpObjectImp::backrefGetter):

16735

(RegExpObjectImp::getOwnPropertySlot):

16736

* kjs/simple_number.h:

16737

(KJS::SimpleNumber::):

16738

(KJS::SimpleNumber::value):

16739

(KJS::SimpleNumber::fits):

16740

(KJS::SimpleNumber::integerFits):

16741

(KJS::SimpleNumber::make):

16742

* kjs/string_object.cpp:

16743

(substituteBackreferences):

16744

* kjs/ustring.cpp:

16745

(KJS::UString::from):

16746

(KJS::UString::toUInt32):

16747

(KJS::UString::find):

16748

(KJS::UString::rfind):

16749

* kjs/ustring.h:

16750

* kjs/value.cpp:

16751

(KJS::jsNumber):

16752

* kjs/value.h:

16753

16754

2005-09-11 Eric Seidel <eseidel@apple.com>

16755

16756

No review requested, build fix affects only SVG.

16757

16758

* JavaScriptCore.xcodeproj/project.pbxproj: Fixed JSC+SVG

16759

Fixed JavaScriptCore+SVG after PCRE 6.1 merger.

16760

http://bugs.webkit.org/show_bug.cgi?id=4932

16761

16762

2005-09-10 Krzysztof Kowalczyk <kkowalczyk@gmail.com>

16763

16764

Reviewed and landed by Darin.

16765

16766

* Makefile.vc: Added.

16767

* README-Win32.txt: Added.

16768

16769

2005-09-10 Darin Adler <darin@apple.com>

16770

16771

- fixed compilation for WebCore (another try)

16772

16773

* kjs/simple_number.h: Added more "using" lines.

16774

16775

2005-09-10 Darin Adler <darin@apple.com>

16776

16777

- fixed compilation for WebCore

16778

16779

* kjs/simple_number.h: Have to include <cmath> here to work around a bug in the GCC

16780

standard C++ library headers.

16781

16782

2005-09-10 Darin Adler <darin@apple.com>

16783

16784

Windows changes by Krzysztof Kowalczyk <kkowalczyk@gmail.com>.

16785

16786

- fixed http://bugs.webkit.org/show_bug.cgi?id=4870

16787

win portability: fix IS_NEGATIVE_ZERO macro in simple_number.h

16788

16789

* kjs/simple_number.h:

16790

(KJS::isNegativeZero): Added. Inline function. Has a case for Windows that

16791

uses _fpclass and a case for other platforms that uses signbit.

16792

(KJS::SimpleNumber::fits): Use inline isNegativeZero instead of macro IS_NEGATIVE_ZERO.

16793

16794

* kjs/internal.cpp: Remove definition of now-unneeded negZero global.

16795

16796

* kjs/value.cpp: Touched the file because Xcode didn't know it needed to

16797

recompile it.

16798

16799

- improved test engine

16800

16801

* tests/mozilla/jsDriver.pl: Sort tests in numeric order instead of using

16802

a plain-ASCII sort; now test 33 will be after test 5 in any given set of

16803

numbered tests.

16804

16805

2005-09-08 Darin Adler <darin@apple.com>

16806

16807

- fixed overloaded versions of throwError so that they substitute *all*

16808

expected parameters into the message string -- some versions used to

16809

skip parameters, resulting in "%s" being printed in the error message.

16810

16811

Reviewed by Geoff.

16812

16813

* kjs/nodes.h: Updated declarations to use "const &" and not to name parameters

16814

* kjs/nodes.cpp: (Node::throwError): Updated to match above and add one missing

16815

call to substitute.

16816

16817

2005-09-08 Darin Adler <darin@apple.com>

16818

16819

Reviewed by Geoff.

16820

16821

- updated to PCRE 6.1

16822

16823

The original PCRE 6.1 sources are checked into the tree with the tag

16824

"pcre-6-1" for reference. What we're checking in right now is the original

16825

plus our changes to make it support UTF-16 and at least one other tweak

16826

(vertical tab considered whitespace). Our work to get our changes was

16827

done on "pcre-6-1-branch", with an anchor at "pcre-6-1-anchor" so you can

16828

see the evolution of the UTF-16 changes.

16829

16830

Note also that there was one small change made here that's not on the branch

16831

in pcre_compile.c.

16832

16833

* Info.plist: Updated the part of the copyright message that's about PCRE.

16834

16835

* JavaScriptCore.xcodeproj/project.pbxproj: Added new PCRE source files,

16836

removed obsolete ones.

16837

16838

* pcre/AUTHORS: Updated to PCRE 6.1. Includes credits for Apple's UTF-16

16839

changes, but not the credits for Google's C++ wrapper, since we don't include that.

16840

* pcre/COPYING: Updated to PCRE 6.1.

16841

* pcre/LICENCE: Ditto.

16842

* pcre/dftables.c: Ditto.

16843

* pcre/pcre-config.h: Ditto.

16844

* pcre/pcre.h: Ditto.

16845

16846

* pcre/pcre_compile.c: Added for PCRE 6.1.

16847

* pcre/pcre_config.c: Ditto.

16848

* pcre/pcre_exec.c: Ditto.

16849

* pcre/pcre_fullinfo.c: Ditto.

16850

* pcre/pcre_get.c: Ditto.

16851

* pcre/pcre_globals.c: Ditto.

16852

* pcre/pcre_info.c: Ditto.

16853

* pcre/pcre_internal.h: Ditto.

16854

* pcre/pcre_maketables.c: Ditto.

16855

* pcre/pcre_ord2utf8.c: Ditto.

16856

* pcre/pcre_printint.c: Ditto.

16857

* pcre/pcre_refcount.c: Ditto.

16858

* pcre/pcre_study.c: Ditto.

16859

* pcre/pcre_tables.c: Ditto.

16860

* pcre/pcre_try_flipped.c: Ditto.

16861

* pcre/pcre_ucp_findchar.c: Ditto.

16862

* pcre/pcre_version.c: Ditto.

16863

* pcre/pcre_xclass.c: Ditto.

16864

* pcre/ucp.h: Ditto.

16865

* pcre/ucp_findchar.c: Ditto.

16866

* pcre/ucpinternal.h: Ditto.

16867

* pcre/ucptable.c: Ditto.

16868

16869

* pcre/get.c: Removed.

16870

* pcre/internal.h: Removed.

16871

* pcre/maketables.c: Removed.

16872

* pcre/pcre.c: Removed.

16873

* pcre/study.c: Removed.

16874

16875

2005-09-07 Geoffrey Garen <ggaren@apple.com>

16876

16877

-fixed http://bugs.webkit.org/show_bug.cgi?id=4781

16878

Date.setMonth fails with big values due to overflow

16879

16880

Reviewed by darin.

16881

16882

* kjs/date_object.cpp:

16883

(timetUsingCF): for consistency, changed return statement to invalidDate instead of LONG_MAX

16884

(KJS::fillStructuresUsingTimeArgs): modified for readability

16885

(KJS::fillStructuresUsingDateArgs): new function analogous to fillStructuresUsingTimeArgs

16886

(KJS::DateProtoFuncImp::callAsFunction): modified to use fillStructuresUsingDateArgs

16887

(KJS::DateObjectImp::construct): moved variable declaration to proper scope

16888

(KJS::DateObjectFuncImp::callAsFunction): moved variable declaration to proper scope

16889

16890

2005-09-07 Geoffrey Garen <ggaren@apple.com>

16891

-updated expected test results to reflect fix for

16892

http://bugs.webkit.org/show_bug.cgi?id=4698

16893

kjs does not allow named functions in function expressions

16894

16895

* tests/mozilla/expected.html:

16896

16897

2005-09-04 Darin Adler <darin@apple.com>

16898

16899

* kjs/identifier.cpp: Fix comment, add missing include.

16900

(Follow-on to changes from yesterday.)

16901

16902

2005-09-03 Krzysztof Kowalczyk <kkowalczyk@gmail.com>

16903

16904

Reviewed, tweaked and landed by Darin.

16905

16906

- another try at some of the Windows compilation fixes

16907

should fix these bugs: 4546, 4831, 4834, 4643, 4830, 4832, 4833, 4835

16908

16909

* kjs/collector.cpp: Add missing <setjmp.h> include.

16910

* kjs/date_object.cpp: Fix broken copysign macro.

16911

* kjs/dtoa.cpp: Move macro definitions down after all header includes.

16912

* kjs/fast_malloc.cpp: Add missing <assert.h> and <stddef.h> includes.

16913

* kjs/function.cpp: Remove broken isxdigit definition.

16914

* kjs/grammar.y: Add a missing semicolon (and remove an excess one).

16915

* kjs/identifier.cpp: Turn off AVOID_STATIC_CONSTRUCTORS because the placement new syntax

16916

doesn't seem to work in Visual C++ (I'm surprised to hear that, by the way).

16917

* kjs/value.h: Made ValueImp's destructor virtual because otherwise pointers to ValueImp

16918

on the stack aren't right for garbage collection on Windows (don't think it works that

16919

way with gcc's virtual table scheme, but it's a harmless change).

16920

16921

2005-09-03 Krzysztof Kowalczyk <kkowalczyk@gmail.com>

16922

16923

Reviewed, tweaked and landed by Darin.

16924

16925

- some Windows compilation fixes, hoping to fix the problems reported in these bugs:

16926

4627, 4629, 4630, 4631, 4632, 4633, 4634, 4635, 4636, 4637, 4639, 4640, 4641, 4644, 4645

16927

16928

* kjs/collector.cpp: Include <windows.h> on WIN32. Put thread-related code inside

16929

KJS_MULTIPLE_THREADS #if directives.

16930

(KJS::Collector::markCurrentThreadConservatively): Use NT_TIB to find the stack base on Win32.

16931

16932

* kjs/config.h: Define HAVE_SYS_TIMEB_H for Win32.

16933

16934

* kjs/date_object.cpp: Add include of <limits.h>. Add definitions of strncasecmp, isfinite, and

16935

copysign for Win32.

16936

(KJS::KRFCDate_parseDate): Move "errno = 0" line down closer to the first call to strol -- I believe

16937

that on Win32 there's some other call before that setting errno.

16938

16939

* kjs/date_object.h: Remove unneeded include of <sys/time.h>.

16940

16941

* kjs/dtoa.cpp: Add an undef of strtod, needed on Win32.

16942

16943

* kjs/fast_malloc.cpp: Put #if !WIN32 around some customization that's not appropriate on Win32.

16944

(KJS::region_list_append): Add a missing cast so this Win32-specific function compiles in C++.

16945

(KJS::sbrk): Change parameter type to match the declaration.

16946

16947

* kjs/function.cpp: (isxdigit): Define a locale-independent isxdigit on Win32.

16948

16949

* kjs/function.h: Remove unneeded friend class Function for FunctionImp.

16950

16951

* kjs/identifier.cpp: Took out the APPLE_CHANGES from around the AVOID_STATIC_CONSTRUCTORS

16952

define. We ultimately intend to phase out APPLE_CHANGES entirely. Also fix the

16953

non-AVOID_STATIC_CONSTRUCTORS code path.

16954

16955

* kjs/internal.cpp: Remove uneeded include of <strings.h>, which was confused with <string.h>!

16956

Add a Win32 implementation of copysign. Put the threads code inside KJS_MULTIPLE_THREADS.

16957

16958

* kjs/internal.h: Define a KJS_MULTIPLE_THREADS macro on non-Win32 only. Later we can make this

16959

specific to Mac OS X if we like.

16960

16961

* kjs/interpreter_map.cpp: Add missing include of <stdlib.h>.

16962

16963

* kjs/list.cpp:

16964

(KJS::ListImp::markValues): Use std::min instead of MIN.

16965

(KJS::List::copy): Ditto.

16966

(KJS::List::copyTail): Ditto.

16967

16968

* kjs/math_object.cpp: (signbit): Add a Win32 implementation of signbit.

16969

16970

* kjs/nodes.cpp: (Node::finalCheck): Use unsigned instead of uint.

16971

Put the use of always_inline inside __GNUC__.

16972

16973

* kjs/number_object.cpp: (NumberProtoFuncImp::callAsFunction): Use "10.0" instead of "10"

16974

inside all the calls to pow to avoid ambiguity caused by overloading of pow on Win32, seen

16975

when passing an int rather than a double or float.

16976

16977

* kjs/operations.cpp:

16978

(KJS::isInf): Add Win32 implementation.

16979

(KJS::isPosInf): Add Win32 implementation.

16980

(KJS::isNegInf): Add Win32 implementation.

16981

16982

* kjs/regexp.cpp: Use unsigned instead of uint.

16983

* kjs/regexp.h: Ditto.

16984

* kjs/regexp_object.cpp: Ditto.

16985

* kjs/regexp_object.h: Ditto.

16986

16987

2005-09-02 Beth Dakin <bdakin@apple.com>

16988

16989

Fix for <rdar://problem/4235531> Denver Regression: Safari crash in KWQStringData::makeUnicode

16990

The other half of the fix is in WebCore.

16991

16992

Fix written by Maciej and Darin.

16993

Reviewed by me/Maciej

16994

16995

As Maciej said in Radar: These problems was caused by a conflict between some of our custom

16996

allocators, causing them to return null. Symptom is typically a null pointer dereference in

16997

a place where it might be expected an allocation has just occurred.

16998

16999

* kjs/fast_malloc.cpp: Added #define for MORECORE_CONTIGUOUS, MORECORE_CANNOT_TRIM,

17000

and MALLOC_FAILURE_ACTION.

17001

17002

2005-08-31 Geoffrey Garen <ggaren@apple.com>

17003

17004

-rolled in fix for http://bugs.webkit.org/show_bug.cgi?id=4698

17005

kjs does not allow named functions in function expressions

17006

17007

Fix by Arthur Langereis.

17008

17009

Reviewed by darin.

17010

17011

* kjs/grammar.y:

17012

* kjs/nodes.cpp:

17013

(FuncExprNode::evaluate):

17014

* kjs/nodes.h:

17015

(KJS::FuncExprNode::FuncExprNode):

17016

17017

Test cases added:

17018

17019

* layout-tests/fast/js/named-function-expression-expected.txt: Added.

17020

* layout-tests/fast/js/named-function-expression.html: Added.

17021

17022

2005-08-31 Justin Haygood <justin@xiondigital.net>

17023

17024

Reviewed, tweaked, and landed by Darin.

17025

17026

- fixed http://bugs.webkit.org/show_bug.cgi?id=4085

17027

- fixed http://bugs.webkit.org/show_bug.cgi?id=4087

17028

- fixed http://bugs.webkit.org/show_bug.cgi?id=4096

17029

Some fixes for compiling on windows.

17030

17031

* kjs/config.h: Added a WIN32 case in here, with suitable defines.

17032

(To be tweaked as necessary.)

17033

* kjs/function.cpp: Took out APPLE_CHANGES around use of ICU.

17034

* kjs/operations.cpp: Removed some bogus code that always set HAVE_FLOAT_H.

17035

17036

2005-08-30 Darin Adler <darin@apple.com>

17037

17038

Reviewed by John Sullivan.

17039

17040

- fixed http://bugs.webkit.org/show_bug.cgi?id=4758

17041

unify SharedPtr in WebCore and JavaScriptCore

17042

17043

* kjs/shared_ptr.h: Updated namespace to KXMLCore instead of kxhmlcore.

17044

Made a few small improvements to use local variables a bit more and added

17045

an "operator int" to reduce the chance that we'll convert a SharedPtr to

17046

an int by accident. Also made the == operators normal functions rather than

17047

friend functions, added a couple of comemnts.

17048

17049

* kjs/function.h: Updated for namespace change.

17050

* kjs/function.cpp: Ditto.

17051

* kjs/function_object.cpp: Ditto.

17052

* kjs/internal.h: Ditto.

17053

* kjs/internal.cpp: Ditto.

17054

* kjs/nodes.h: Ditto.

17055

* kjs/nodes2string.cpp: Ditto.

17056

17057

2005-08-26 Maciej Stachowiak <mjs@apple.com>

17058

17059

Reviewed by John.

17060

17061

<rdar://problem/4224911> many many leaks in kjsyyparse with malformed Javascript

17062

17063

Record all nodes that are created during parsing, and delete any

17064

that are left floating with a refcount of 0.

17065

17066

* kjs/internal.cpp:

17067

(KJS::Parser::saveNewNode):

17068

(KJS::clearNewNodes):

17069

(KJS::Parser::parse):

17070

* kjs/internal.h:

17071

* kjs/nodes.cpp:

17072

(Node::Node):

17073

* kjs/nodes.h:

17074

(KJS::Node::refcount):

17075

17076

2005-08-26 Maciej Stachowiak <mjs@apple.com>

17077

17078

Reviewed by John.

17079

17080

- fixed <rdar://problem/4232452> many many leaks in kjsyyparse on some well-formed JavaScript (can repro on sony.com, webkit tests)

17081

17082

Fixed by changing the refcounting scheme for nodes. Instead of each node implementing a custom ref and

17083

deref for all its children (and being responsible for deleting them), nodes use a smart pointer to

17084

hold their children, and smart pointers are used outside the node tree as well. This change mostly

17085

removes code.

17086

17087

* JavaScriptCore.xcodeproj/project.pbxproj:

17088

* kjs/function.cpp:

17089

(KJS::DeclaredFunctionImp::DeclaredFunctionImp):

17090

(KJS::GlobalFuncImp::callAsFunction):

17091

* kjs/function.h:

17092

* kjs/function_object.cpp:

17093

(FunctionObjectImp::construct):

17094

* kjs/grammar.y:

17095

* kjs/internal.cpp:

17096

(KJS::Parser::parse):

17097

(KJS::Parser::accept):

17098

(KJS::InterpreterImp::checkSyntax):

17099

(KJS::InterpreterImp::evaluate):

17100

* kjs/internal.h:

17101

* kjs/nodes.cpp:

17102

(Node::Node):

17103

(Node::~Node):

17104

(ElementNode::evaluate):

17105

(PropertyValueNode::evaluate):

17106

(ArgumentListNode::evaluateList):

17107

(NewExprNode::evaluate):

17108

(FunctionCallValueNode::evaluate):

17109

(FunctionCallBracketNode::evaluate):

17110

(FunctionCallDotNode::evaluate):

17111

(RelationalNode::evaluate):

17112

(StatListNode::execute):

17113

(StatListNode::processVarDecls):

17114

(VarDeclListNode::evaluate):

17115

(VarDeclListNode::processVarDecls):

17116

(ForInNode::ForInNode):

17117

(ClauseListNode::processVarDecls):

17118

(CaseBlockNode::evalBlock):

17119

(FuncDeclNode::processFuncDecl):

17120

(FuncExprNode::evaluate):

17121

(SourceElementsNode::execute):

17122

(SourceElementsNode::processFuncDecl):

17123

(SourceElementsNode::processVarDecls):

17124

* kjs/nodes.h:

17125

(KJS::Node::ref):

17126

(KJS::Node::deref):

17127

(KJS::NumberNode::NumberNode):

17128

(KJS::GroupNode::GroupNode):

17129

(KJS::ElementNode::ElementNode):

17130

(KJS::ArrayNode::ArrayNode):

17131

(KJS::PropertyValueNode::PropertyValueNode):

17132

(KJS::ObjectLiteralNode::ObjectLiteralNode):

17133

(KJS::BracketAccessorNode::BracketAccessorNode):

17134

(KJS::DotAccessorNode::DotAccessorNode):

17135

(KJS::ArgumentListNode::ArgumentListNode):

17136

(KJS::ArgumentsNode::ArgumentsNode):

17137

(KJS::NewExprNode::NewExprNode):

17138

(KJS::FunctionCallValueNode::FunctionCallValueNode):

17139

(KJS::FunctionCallResolveNode::FunctionCallResolveNode):

17140

(KJS::FunctionCallBracketNode::FunctionCallBracketNode):

17141

(KJS::FunctionCallDotNode::FunctionCallDotNode):

17142

(KJS::PostfixNode::PostfixNode):

17143

(KJS::DeleteNode::DeleteNode):

17144

(KJS::VoidNode::VoidNode):

17145

(KJS::TypeOfNode::TypeOfNode):

17146

(KJS::PrefixNode::PrefixNode):

17147

(KJS::UnaryPlusNode::UnaryPlusNode):

17148

(KJS::NegateNode::NegateNode):

17149

(KJS::BitwiseNotNode::BitwiseNotNode):

17150

(KJS::LogicalNotNode::LogicalNotNode):

17151

(KJS::MultNode::MultNode):

17152

(KJS::AddNode::AddNode):

17153

(KJS::ShiftNode::ShiftNode):

17154

(KJS::RelationalNode::RelationalNode):

17155

(KJS::EqualNode::EqualNode):

17156

(KJS::BitOperNode::BitOperNode):

17157

(KJS::BinaryLogicalNode::BinaryLogicalNode):

17158

(KJS::ConditionalNode::ConditionalNode):

17159

(KJS::AssignResolveNode::AssignResolveNode):

17160

(KJS::AssignBracketNode::AssignBracketNode):

17161

(KJS::AssignDotNode::AssignDotNode):

17162

(KJS::CommaNode::CommaNode):

17163

(KJS::AssignExprNode::AssignExprNode):

17164

(KJS::VarDeclListNode::VarDeclListNode):

17165

(KJS::VarStatementNode::VarStatementNode):

17166

(KJS::ExprStatementNode::ExprStatementNode):

17167

(KJS::IfNode::IfNode):

17168

(KJS::DoWhileNode::DoWhileNode):

17169

(KJS::WhileNode::WhileNode):

17170

(KJS::ForNode::ForNode):

17171

(KJS::ReturnNode::ReturnNode):

17172

(KJS::WithNode::WithNode):

17173

(KJS::CaseClauseNode::CaseClauseNode):

17174

(KJS::ClauseListNode::ClauseListNode):

17175

(KJS::ClauseListNode::clause):

17176

(KJS::ClauseListNode::next):

17177

(KJS::SwitchNode::SwitchNode):

17178

(KJS::LabelNode::LabelNode):

17179

(KJS::ThrowNode::ThrowNode):

17180

(KJS::CatchNode::CatchNode):

17181

(KJS::FinallyNode::FinallyNode):

17182

(KJS::TryNode::TryNode):

17183

(KJS::ParameterNode::ParameterNode):

17184

(KJS::ParameterNode::nextParam):

17185

(KJS::FuncDeclNode::FuncDeclNode):

17186

(KJS::FuncExprNode::FuncExprNode):

17187

* kjs/nodes2string.cpp:

17188

(KJS::SourceStream::operator<<):

17189

(ElementNode::streamTo):

17190

(PropertyValueNode::streamTo):

17191

(ArgumentListNode::streamTo):

17192

(StatListNode::streamTo):

17193

(VarDeclListNode::streamTo):

17194

(CaseBlockNode::streamTo):

17195

(ParameterNode::streamTo):

17196

(SourceElementsNode::streamTo):

17197

* kjs/shared_ptr.h: Added.

17198

(kxmlcore::SharedPtr::SharedPtr):

17199

(kxmlcore::SharedPtr::~SharedPtr):

17200

(kxmlcore::SharedPtr::isNull):

17201

(kxmlcore::SharedPtr::notNull):

17202

(kxmlcore::SharedPtr::reset):

17203

(kxmlcore::SharedPtr::get):

17204

(kxmlcore::SharedPtr::operator*):

17205

(kxmlcore::SharedPtr::operator->):

17206

(kxmlcore::SharedPtr::operator!):

17207

(kxmlcore::SharedPtr::operator bool):

17208

(kxmlcore::SharedPtr::operator==):

17209

(kxmlcore::::operator):

17210

(kxmlcore::operator!=):

17211

(kxmlcore::static_pointer_cast):

17212

(kxmlcore::const_pointer_cast):

17213

17214

2005-08-26 Geoff Garen <ggaren@apple.com>

17215

17216

Reviewed by John.

17217

Landed by Darin.

17218

17219

- fixed http://bugs.webkit.org/show_bug.cgi?id=4664

17220

TOT Crash from backwards null check in WebScriptObject.mm

17221

17222

* bindings/objc/WebScriptObject.mm:

17223

(+[WebScriptObject _convertValueToObjcValue:originExecutionContext:executionContext:]):

17224

Remove bogus !.

17225

17226

2005-08-25 Darin Adler <darin@apple.com>

17227

17228

Reviewed by John Sullivan.

17229

17230

- rename KJS::UString::string() to KJS::UString::domString()

17231

- rename KJS::Identifier::string() to KJS::Identifier::domString()

17232

17233

* kjs/identifier.h: Renamed.

17234

* kjs/ustring.h: Ditto.

17235

17236

2005-08-19 Darin Adler <darin@apple.com>

17237

17238

Reviewed by Maciej.

17239

17240

- fixed http://bugs.webkit.org/show_bug.cgi?id=4435

17241

speed up JavaScript by tweaking the Identifier class

17242

17243

* kjs/identifier.h: Add a new global nullIdentifier and make Identifier::null a function

17244

that returns it.

17245

* kjs/identifier.cpp: (KJS::Identifier::init): Initialize a global for the null identifier

17246

as well as all the other globals for special identifiers.

17247

17248

* kjs/ustring.h: (KJS::UString::UString): Make this empty constructor inline.

17249

* kjs/ustring.cpp: Remove the old non-inline version.

17250

17251

2005-08-19 Mitz Pettel <opendarwin.org@mitzpettel.com>

17252

17253

Reviewed by Maciej.

17254

Revised and landed by Darin.

17255

17256

- fixed http://bugs.webkit.org/show_bug.cgi?id=4474

17257

REGRESSION: Crash when using in-place operator on uninitialized array element

17258

17259

* kjs/nodes.cpp:

17260

(AssignResolveNode::evaluate): Remove unneeded "isSet" assertion.

17261

(AssignBracketNode::evaluate): Replace code that tested "isSet" with code that

17262

tests the return value of getPropertySlot.

17263

17264

* kjs/property_slot.h: Removed unneeded "isSet" function. Property slots are

17265

either uninitialized or set. There's no "initialized and not set" state.

17266

17267

2005-08-18 Adele Peterson <adele@apple.com>

17268

17269

Checked "Inline Functions Hidden" box

17270

17271

* JavaScriptCore.xcodeproj/project.pbxproj:

17272

17273

2005-08-16 Darin Adler <darin@apple.com>

17274

17275

Reviewed by Geoff.

17276

17277

- fixed crash in one of the JavaScript tests (introduced by my throwError change)

17278

17279

* kjs/nodes.cpp: (Node::setExceptionDetailsIfNeeded): Check if the exception is an

17280

object before setting the file and line number properties on it. Something to think

17281

about in the future -- do we really want to do this on any object that's thrown?

17282

How about limiting it to error objects that were created by the JavaScript engine?

17283

17284

- changed kjs_fast_malloc so we don't have two conflicting versions of the same function

17285

17286

* kjs/fast_malloc.h: Took out all the ifdefs from this header.

17287

* kjs/fast_malloc.cpp: Added non-NDEBUG versions of the functions that just call

17288

the system malloc, and put the NDEBUG versions in an #else.

17289

17290

2005-08-16 Darin Adler <darin@apple.com>

17291

17292

Reviewed by Geoff.

17293

17294

- clean up exported symbols that are not in a "KJS" namespace

17295

17296

* bindings/NP_jsobject.cpp: (identiferFromNPIdentifier): Marked this function static

17297

so it no longer has external linkage.

17298

* bindings/c/c_utility.h: Put all this stuff inside the KJS namespace.

17299

* bindings/c/c_utility.cpp: Also marked some globals static so they don't have external

17300

linkage; not as important given the namespace.

17301

* bindings/npruntime.cpp: Marked functions static so they no longer have internal linkage.

17302

Also removed unused _NPN_SetExceptionWithUTF8 function (not in header, had C++ linkage!).

17303

17304

* bindings/jni/jni_utility.cpp: (KJS::Bindings::getJavaVM): Call KJS_GetCreatedJavaVMs

17305

using the soft linking header, instead of calling the JNI call. This allows processes

17306

to link both JavaScriptCore and JavaVM without a symbol conflict.

17307

* bindings/softlinking.c:

17308

(loadFramework): Marked this function static so it no longer has external linkage.

17309

(getFunctionPointer): Ditto.

17310

(KJS_GetCreatedJavaVMs): Renamed this so it has a KJS prefix.

17311

17312

* JavaScriptCore.xcodeproj/project.pbxproj: Added softlinking.h.

17313

* bindings/softlinking.h: Added.

17314

17315

* kjs/nodes2string.cpp: (streamAssignmentOperatorTo): Marked this function static so it

17316

no longer has external linkage.

17317

17318

2005-08-15 Darin Adler <darin@apple.com>

17319

17320

Reviewed by Geoff.

17321

17322

- fixed http://bugs.webkit.org/show_bug.cgi?id=4437

17323

clean up error creation with new throwError function

17324

17325

* bindings/NP_jsobject.cpp:

17326

(_NPN_SetException):

17327

* bindings/jni/jni_instance.cpp:

17328

(JavaInstance::invokeMethod):

17329

* bindings/jni/jni_runtime.cpp:

17330

(JavaField::dispatchValueFromInstance):

17331

(JavaField::dispatchSetValueToInstance):

17332

* bindings/objc/WebScriptObject.mm:

17333

(-[WebScriptObject _initializeWithObjectImp:originExecutionContext:executionContext:]):

17334

(-[WebScriptObject _initWithObjectImp:originExecutionContext:executionContext:]):

17335

(+[WebScriptObject throwException:]):

17336

(-[WebScriptObject setException:]):

17337

(+[WebScriptObject _convertValueToObjcValue:originExecutionContext:executionContext:]):

17338

* bindings/objc/objc_class.h:

17339

(KJS::Bindings::ObjcClass::~ObjcClass):

17340

(KJS::Bindings::ObjcClass::ObjcClass):

17341

(KJS::Bindings::ObjcClass::operator=):

17342

(KJS::Bindings::ObjcClass::constructorAt):

17343

(KJS::Bindings::ObjcClass::numConstructors):

17344

* bindings/objc/objc_header.h:

17345

* bindings/objc/objc_runtime.h:

17346

(KJS::Bindings::ObjcField::~ObjcField):

17347

(KJS::Bindings::ObjcField::ObjcField):

17348

(KJS::Bindings::ObjcField::operator=):

17349

(KJS::Bindings::ObjcMethod::ObjcMethod):

17350

(KJS::Bindings::ObjcMethod::~ObjcMethod):

17351

(KJS::Bindings::ObjcMethod::operator=):

17352

* bindings/objc/objc_runtime.mm:

17353

(ObjcField::valueFromInstance):

17354

(ObjcField::setValueToInstance):

17355

(ObjcArray::setValueAt):

17356

(ObjcArray::valueAt):

17357

* bindings/objc/objc_utility.h:

17358

* bindings/objc/objc_utility.mm:

17359

(KJS::Bindings::JSMethodNameToObjCMethodName):

17360

(KJS::Bindings::convertValueToObjcValue):

17361

(KJS::Bindings::convertNSStringToString):

17362

(KJS::Bindings::convertObjcValueToValue):

17363

(KJS::Bindings::objcValueTypeForType):

17364

(KJS::Bindings::createObjcInstanceForValue):

17365

(KJS::Bindings::throwError):

17366

* bindings/runtime.h:

17367

(KJS::Bindings::Parameter::~Parameter):

17368

(KJS::Bindings::Method::~Method):

17369

(KJS::Bindings::Instance::Instance):

17370

(KJS::Bindings::Instance::begin):

17371

(KJS::Bindings::Instance::end):

17372

(KJS::Bindings::Instance::getValueOfUndefinedField):

17373

(KJS::Bindings::Instance::supportsSetValueOfUndefinedField):

17374

(KJS::Bindings::Instance::setValueOfUndefinedField):

17375

(KJS::Bindings::Instance::valueOf):

17376

* bindings/runtime_array.cpp:

17377

(RuntimeArrayImp::put):

17378

* bindings/runtime_object.h:

17379

(KJS::RuntimeObjectImp::setInternalInstance):

17380

(KJS::RuntimeObjectImp::getInternalInstance):

17381

* kjs/array_object.cpp:

17382

(getProperty):

17383

(ArrayProtoFuncImp::callAsFunction):

17384

(ArrayObjectImp::construct):

17385

* kjs/bool_object.cpp:

17386

(BooleanProtoFuncImp::callAsFunction):

17387

* kjs/date_object.cpp:

17388

(KJS::DateProtoFuncImp::callAsFunction):

17389

* kjs/function.cpp:

17390

(KJS::decode):

17391

(KJS::GlobalFuncImp::callAsFunction):

17392

* kjs/function_object.cpp:

17393

(FunctionProtoFuncImp::callAsFunction):

17394

(FunctionObjectImp::construct):

17395

* kjs/internal.cpp:

17396

(KJS::UndefinedImp::toObject):

17397

(KJS::NullImp::toObject):

17398

(KJS::InterpreterImp::evaluate):

17399

(KJS::InternalFunctionImp::hasInstance):

17400

* kjs/nodes.cpp:

17401

(Node::throwError):

17402

(substitute):

17403

(Node::setExceptionDetailsIfNeeded):

17404

(undefinedVariableError):

17405

(ProgramNode::ProgramNode):

17406

* kjs/number_object.cpp:

17407

(NumberProtoFuncImp::callAsFunction):

17408

* kjs/object.cpp:

17409

(KJS::ObjectImp::call):

17410

(KJS::ObjectImp::defaultValue):

17411

(KJS::Error::create):

17412

(KJS::throwError):

17413

* kjs/object.h:

17414

(KJS::ObjectImp::clearProperties):

17415

(KJS::ObjectImp::getPropertySlot):

17416

(KJS::ObjectImp::getOwnPropertySlot):

17417

* kjs/object_object.cpp:

17418

(ObjectProtoFuncImp::callAsFunction):

17419

* kjs/reference.cpp:

17420

(KJS::Reference::getBase):

17421

(KJS::Reference::getValue):

17422

(KJS::Reference::putValue):

17423

(KJS::Reference::deleteValue):

17424

* kjs/regexp_object.cpp:

17425

(RegExpProtoFuncImp::callAsFunction):

17426

(RegExpObjectImp::construct):

17427

* kjs/string_object.cpp:

17428

(StringProtoFuncImp::callAsFunction):

17429

17430

2005-08-15 Anders Carlsson <andersca@mac.com>

17431

17432

Reviewed by Darin.

17433

17434

* tests/mozilla/ecma_3/Date/15.9.5.5.js:

17435

Remove the code which tests that Date.toLocaleString should be parsable

17436

by Date.parse. That is not true according to the spec.

17437

17438

2005-08-15 Darin Adler <darin@apple.com>

17439

17440

Reviewed by Geoff.

17441

17442

* kjs/collector.cpp: (KJS::Collector::allocate): Use a local instead of a global in one

17443

more place; slight speedup.

17444

17445

2005-08-14 Darin Adler <darin@apple.com>

17446

17447

Reviewed by Maciej.

17448

17449

- fixed crash observed on one of the Apple-only layout tests

17450

17451

* kjs/property_map.cpp: (KJS::PropertyMap::mark): Change code to understand that deleted

17452

entries have a value of NULL, so the deleted sentinel count doesn't need to be included

17453

in the count of things to mark since we're ignoring the keys.

17454

17455

2005-08-14 Darin Adler <darin@apple.com>

17456

17457

Reviewed by Maciej.

17458

17459

- fixed http://bugs.webkit.org/show_bug.cgi?id=4421

17460

speed up JavaScript by inlining some label stack functions

17461

17462

* kjs/internal.h: Removed the copy constructor and assignment operator for LabelStack.

17463

They were unused, and the implementations had bugs; I removed them rather than fixing them.

17464

Also removed the clear function, since that was only needed to help the assignment operator

17465

share code with the destructor, and was not efficient enough for the destructor.

17466

(KJS::LabelStack::~LabelStack): Made this inline. Also used an efficient implementation

17467

that's nice and fast when the stack is empty, better than the old clear() function which

17468

used to keep updating and refetching "tos" each time through the loop.

17469

(KJS::LabelStack::pop): Made this inline.

17470

17471

* kjs/internal.cpp: Deleted the now-inline functions and the obsolete functions. Also

17472

deleted a commented-out line of code.

17473

17474

2005-08-14 Darin Adler <darin@apple.com>

17475

17476

Reviewed by Maciej.

17477

17478

- fixed http://bugs.webkit.org/show_bug.cgi?id=4419

17479

speed up JavaScript by improving KJS::List

17480

17481

my measurements show an improvement of 1% on iBench JavaScript

17482

17483

* kjs/list.cpp: Rearrange list to make the values and free list share the same storage,

17484

which saves 4 bytes per list. Also remove the pointers used only on the heap from the

17485

lists that are in the pool, which saves 8 bytes per list. Moving the free list pointer

17486

closer to the start of the list object also speeds up access to the free list. New

17487

"HeapListImp" struct is used only for the lists on the heap.

17488

(KJS::List::markProtectedLists): Shadowed global variable in local and updated for the

17489

new terminology ("heap" instead of "outside pool").

17490

(KJS::allocateListImp): Updated for new terminology.

17491

(KJS::List::release): Moved the code from deallocateListImp in here -- it wasn't being

17492

inlined and didn't need to be in a separate function.

17493

17494

2005-08-14 Darin Adler <darin@apple.com>

17495

17496

Reviewed by Maciej.

17497

17498

- fixed http://bugs.webkit.org/show_bug.cgi?id=4417

17499

speed up JavaScript with some small changes to the property map code

17500

17501

my measurements show an improvement of 2% on iBench JavaScript

17502

17503

* kjs/property_map.h: (KJS::PropertyMap::PropertyMap): Made the default constructor inline.

17504

* kjs/property_map.cpp:

17505

(KJS::PropertyMap::~PropertyMap): Changed loop to exit early once we know we've processed

17506

all the hash table entries, based on the count.

17507

(KJS::PropertyMap::mark): Ditto.

17508

17509

* kjs/object.h: Made an arbitrary change here to force recompiling so we pick up changes to

17510

property_map.h. Works around what seems to be an Xcode header dependency bug.

17511

17512

2005-08-14 Darin Adler <darin@apple.com>

17513

17514

Reviewed by Maciej.

17515

17516

- fixed http://bugs.webkit.org/show_bug.cgi?id=4416

17517

speed up JavaScript with some improvements to the garbage collector

17518

17519

my measurements show an improvement of 2% on iBench JavaScript

17520

17521

* kjs/collector.cpp:

17522

(KJS::Collector::allocate): Use local variables to shadow globals instead of repeatedly

17523

going at global variables. Tighten up loop implementations to make the common case fast.

17524

(KJS::Collector::markStackObjectsConservatively): Use local variables to shadow globals.

17525

Used a goto to eliminate a boolean since it was showing up in the profile.

17526

(KJS::Collector::markProtectedObjects): Iterate through the table using pointer rather

17527

than an index since the profile showed that generating better code.

17528

(KJS::Collector::collect): Added a special case for blocks where all cells are used,

17529

Use local variables to shadow globals. Eliminated a boolean by computing it another

17530

way (checking to see if the number of live objects changed). Also used local variables

17531

to shadow fields in the current cell when sweeping.

17532

(KJS::Collector::numReferencedObjects): Use AllocatedValueImp instead of ValueImp

17533

in one place -- means we get faster versions of various functions that don't worry

17534

about SimpleNumber.

17535

(KJS::className): Ditto.

17536

(KJS::Collector::rootObjectClasses): Ditto.

17537

17538

2005-08-14 Darin Adler <darin@apple.com>

17539

17540

- fixed http://bugs.webkit.org/show_bug.cgi?id=4344

17541

REGRESSION: JavaScript crash when going back from viewing a thread (NULL protoype)

17542

17543

* kjs/error_object.cpp: (NativeErrorImp::NativeErrorImp): Set proto in a more

17544

straightforward way. The old code set the proto to 0 and then to the correct value.

17545

This showed up as a "false positive" when searching for places that set prototype

17546

to NULL/0 so I fixed it.

17547

17548

* kjs/function_object.cpp: (FunctionPrototypeImp::FunctionPrototypeImp): Change to

17549

not pass an explicit "0" to the base class (InternalFunctionImp) constructor.

17550

17551

* kjs/internal.h: Added a default constructor for InternalFunctionImp.

17552

* kjs/internal.cpp: (KJS::InternalFunctionImp::InternalFunctionImp): Added the

17553

default constructor (empty body, just calls base class's default constructor).

17554

17555

* kjs/object.h:

17556

(KJS::ObjectImp::ObjectImp): Add an assertion to catch NULL prototypes earlier

17557

in Development builds.

17558

(KJS::ObjectImp::setPrototype): Ditto.

17559

17560

2005-08-12 Maciej Stachowiak <mjs@apple.com>

17561

17562

Reviewed by John.

17563

17564

- two simple speed improvements for a 3% speed gain

17565

17566

* JavaScriptCore.xcodeproj/project.pbxproj: turn on -fstrict-aliasing

17567

17568

* kjs/scope_chain.h:

17569

(KJS::ScopeChainIterator::ScopeChainIterator): Add a scope chain iterator

17570

so you can walk a scope chain without having to make a copy that you then mutate.

17571

(KJS::ScopeChainIterator::operator*): standard iterator operation

17572

(KJS::ScopeChainIterator::operator->): ditto

17573

(KJS::ScopeChainIterator::operator++): ditto

17574

(KJS::ScopeChainIterator::operator==): ditto

17575

(KJS::ScopeChainIterator::operator!=): ditto

17576

(KJS::ScopeChain::begin): Iterator for the top of the scope chain

17577

(KJS::ScopeChain::end): Iterator for one past the bottom (i.e. null)

17578

* kjs/nodes.cpp:

17579

(ResolveNode::evaluate): Use scope chain iterator instead of copying

17580

a scope chain and then modifying the copy

17581

(ResolveNode::evaluateReference): ditto

17582

(FunctionCallResolveNode::evaluate): ditto

17583

(AssignResolveNode::evaluate): ditto

17584

17585

2005-08-12 Maciej Stachowiak <mjs@apple.com>

17586

17587

Patch from Anders Carlsson, reviewed by me.

17588

17589

* kjs/nodes.h: Fix build breakage.

17590

17591

2005-08-12 Maciej Stachowiak <mjs@apple.com>

17592

17593

Reviewed by hyatt.

17594

17595

- refactor function calls, 3% speedup on JS iBench.

17596

17597

* kjs/grammar.y:

17598

* kjs/nodes.cpp:

17599

(Node::throwError): Added new useful variants.

17600

(FunctionCallValueNode::evaluate): New node to handle calls on expressions

17601

that are strictly values, not references.

17602

(FunctionCallValueNode::ref): ditto

17603

(FunctionCallValueNode::deref): ditto

17604

(FunctionCallResolveNode::evaluate): New node to handle calls on identifier

17605

expressions, so that they are looked up in the scope chain.

17606

(FunctionCallResolveNode::ref): ditto

17607

(FunctionCallResolveNode::deref): ditto

17608

(FunctionCallBracketNode::evaluate): New node to handle calls on bracket

17609

dereferences, so that the expression before brackets is used as the this

17610

object.

17611

(FunctionCallBracketNode::ref): ditto

17612

(FunctionCallBracketNode::deref): ditto

17613

(FunctionCallDotNode::evaluate): New node to handle calls on dot

17614

dereferences, so that the expression before the dot is used as the this

17615

object.

17616

(FunctionCallDotNode::ref): ditto

17617

(FunctionCallDotNode::deref): ditto

17618

(dotExprNotAnObjectString): helper function to avoid global variable access.

17619

(dotExprDoesNotAllowCallsString): ditto

17620

* kjs/nodes.h: Declared new classes.

17621

* kjs/nodes2string.cpp:

17622

(FunctionCallValueNode::streamTo): Added - serializes the appropriate function call

17623

(FunctionCallResolveNode::streamTo): ditto

17624

(FunctionCallBracketNode::streamTo): ditto

17625

(FunctionCallParenBracketNode::streamTo): ditto

17626

(FunctionCallDotNode::streamTo): ditto

17627

(FunctionCallParenDotNode::streamTo): ditto

17628

* kjs/object.h:

17629

(KJS::ObjectImp::isActivation): Change how activation objects are

17630

detected in the scope chain, a virtual function is cheaper than the

17631

old inheritance test.

17632

* kjs/function.h:

17633

(KJS::ActivationImp::isActivation): Ditto.

17634

17635

2005-08-11 Maciej Stachowiak <mjs@apple.com>

17636

17637

- added missing file from earlier checkin

17638

17639

* kjs/grammar_types.h: Added.

17640

(KJS::makeNodePair):

17641

(KJS::makeNodeWithIdent):

17642

17643

2005-08-11 Maciej Stachowiak <mjs@apple.com>

17644

17645

Reviewed by Geoff.

17646

17647

* kjs/date_object.cpp:

17648

(timetUsingCF): Fix one of the date tests my making the CF version of mktime

17649

have the same quirk about the DST field as the real mktime.

17650

* tests/mozilla/expected.html: Updated for newly fixed test.

17651

17652

2005-08-11 Maciej Stachowiak <mjs@apple.com>

17653

17654

- updated for one of the tests that Darin incidentally fixed.

17655

17656

* tests/mozilla/expected.html:

17657

17658

2005-08-10 Maciej Stachowiak <mjs@apple.com>

17659

17660

Reviewed by Geoff.

17661

17662

Refactor assignment grammar to avoid Reference type, and to later

17663

be able to take advantage of writeable PropertySlots, when those

17664

are added. I also fixed a minor bug, turning a function to a

17665

string lost parentheses, I made sure they are printed at least

17666

where semantically significant.

17667

17668

Test cases: see WebCore

17669

17670

* kjs/grammar.y: Change grammar so that assignment expressions are parsed

17671

directly to nodes that know how to set the kind of location being assigned, instead

17672

of having a generic assign node that counts on evaluateReference.

17673

* kjs/lexer.cpp: Include grammar_types.h.

17674

* kjs/nodes.cpp:

17675

(BracketAccessorNode): Renamed from AccessorNode1 for clarity.

17676

(DotAccessorNode): Renamed from AccessorNode2 for clarity.

17677

(combineForAssignment): Inline function for doing the proper kind of

17678

operation for various update assignments like += or *=.

17679

(AssignResolveNode): Node that handles assignment to a bare identifier.

17680

(AssignDotNode): Node that handles assignments of the form EXPR . IDENT = EXPR

17681

(AssignBracketNode): EXPR [ IDENT ] = EXPR

17682

* kjs/nodes.h: Updated for declarations/renames of new classes.

17683

* kjs/nodes2string.cpp:

17684

(GroupNode::streamTo): Fixed to print parens around the expression.

17685

(BracketAccessorNode::streamTo): Renamed.

17686

(DotAccessorNode::streamTo): Renamed.

17687

(AssignResolveNode::streamTo): Added.

17688

(AssignBracketNode::streamTo): Added.

17689

(AssignDotNode::streamTo): Added.

17690

(streamAssignmentOperatorTo): helper function for the above

17691

* kjs/property_slot.h:

17692

(KJS::PropertySlot::isSet): Made this const.

17693

17694

2005-08-10 Adele Peterson <adele@apple.com>

17695

17696

Bumping version to 420+

17697

17698

* Info.plist:

17699

17700

2005-08-10 Geoffrey Garen <ggaren@apple.com>

17701

17702

-fixed <rdar://problem/4151132> REGRESSION: Some applet liveconnect calls

17703

throws privilege exception.

17704

17705

Reviewed by richard and mjs.

17706

17707

-I removed the global static JavaClass cache, since it violated Java

17708

security to cache classes between websites and applets.

17709

17710

* bindings/jni/jni_class.cpp:

17711

-removed global static cache dictionary

17712

-instance constructor and destructor now do the work that used to

17713

be done by static factory methods

17714

-removed obsolete functions

17715

(JavaClass::JavaClass):

17716

(JavaClass::~JavaClass):

17717

* bindings/jni/jni_class.h:

17718

-removed obsolete function declarations

17719

-made copying private since it's unused and it's also not clear

17720

excatly how copying would work with Java security

17721

-made default construction private since it's meaningless

17722

* bindings/jni/jni_instance.cpp:

17723

-removed obsolete functions

17724

(JavaInstance::~JavaInstance):

17725

(JavaInstance::getClass):

17726

* bindings/jni/jni_instance.h:

17727

-made copying private since it's unused and it's also not clear

17728

excatly how copying would work with Java security

17729

-made default construction private since it's meaningless

17730

17731

2005-08-08 Geoffrey Garen <ggaren@apple.com>

17732

17733

-fixed crash caused by fix for http://bugs.webkit.org/show_bug.cgi?id=4313

17734

17735

- exceptionDescription now gets explicitly initialized to NULL in all

17736

the places listed below -- our wrapper classes used to take care of this

17737

automagically

17738

17739

* bindings/jni/jni_instance.cpp:

17740

(JavaInstance::invokeMethod):

17741

* bindings/jni/jni_runtime.cpp:

17742

(JavaField::dispatchValueFromInstance):

17743

(JavaField::dispatchSetValueToInstance):

17744

17745

2005-08-08 Darin Adler <darin@apple.com>

17746

17747

Reviewed by John Sullivan.

17748

17749

- fixed http://bugs.webkit.org/show_bug.cgi?id=4325

17750

Mozilla Date tests have an unnecessary loop that runs 1970 times before each test

17751

17752

* tests/mozilla/ecma/shell.js: Added TIME_YEAR_0 constant.

17753

17754

* tests/mozilla/ecma/Date/15.9.5.10-1.js: Removed the loop and changed code to use the constant.

17755

* tests/mozilla/ecma/Date/15.9.5.10-10.js: Ditto.

17756

* tests/mozilla/ecma/Date/15.9.5.10-11.js: Ditto.

17757

* tests/mozilla/ecma/Date/15.9.5.10-12.js: Ditto.

17758

* tests/mozilla/ecma/Date/15.9.5.10-13.js: Ditto.

17759

* tests/mozilla/ecma/Date/15.9.5.10-2.js: Ditto.

17760

* tests/mozilla/ecma/Date/15.9.5.10-3.js: Ditto.

17761

* tests/mozilla/ecma/Date/15.9.5.10-4.js: Ditto.

17762

* tests/mozilla/ecma/Date/15.9.5.10-5.js: Ditto.

17763

* tests/mozilla/ecma/Date/15.9.5.10-6.js: Ditto.

17764

* tests/mozilla/ecma/Date/15.9.5.10-7.js: Ditto.

17765

* tests/mozilla/ecma/Date/15.9.5.10-8.js: Ditto.

17766

* tests/mozilla/ecma/Date/15.9.5.10-9.js: Ditto.

17767

* tests/mozilla/ecma/Date/15.9.5.11-2.js: Ditto.

17768

* tests/mozilla/ecma/Date/15.9.5.12-1.js: Ditto.

17769

* tests/mozilla/ecma/Date/15.9.5.12-2.js: Ditto.

17770

* tests/mozilla/ecma/Date/15.9.5.12-3.js: Ditto.

17771

* tests/mozilla/ecma/Date/15.9.5.12-4.js: Ditto.

17772

* tests/mozilla/ecma/Date/15.9.5.12-5.js: Ditto.

17773

* tests/mozilla/ecma/Date/15.9.5.12-6.js: Ditto.

17774

* tests/mozilla/ecma/Date/15.9.5.12-7.js: Ditto.

17775

* tests/mozilla/ecma/Date/15.9.5.12-8.js: Ditto.

17776

* tests/mozilla/ecma/Date/15.9.5.13-2.js: Ditto.

17777

* tests/mozilla/ecma/Date/15.9.5.13-8.js: Ditto.

17778

* tests/mozilla/ecma/Date/15.9.5.14.js: Ditto.

17779

* tests/mozilla/ecma/Date/15.9.5.15.js: Ditto.

17780

* tests/mozilla/ecma/Date/15.9.5.16.js: Ditto.

17781

* tests/mozilla/ecma/Date/15.9.5.17.js: Ditto.

17782

* tests/mozilla/ecma/Date/15.9.5.18.js: Ditto.

17783

* tests/mozilla/ecma/Date/15.9.5.19.js: Ditto.

17784

* tests/mozilla/ecma/Date/15.9.5.20.js: Ditto.

17785

* tests/mozilla/ecma/Date/15.9.5.21-1.js: Ditto.

17786

* tests/mozilla/ecma/Date/15.9.5.21-2.js: Ditto.

17787

* tests/mozilla/ecma/Date/15.9.5.21-3.js: Ditto.

17788

* tests/mozilla/ecma/Date/15.9.5.21-4.js: Ditto.

17789

* tests/mozilla/ecma/Date/15.9.5.21-5.js: Ditto.

17790

* tests/mozilla/ecma/Date/15.9.5.21-6.js: Ditto.

17791

* tests/mozilla/ecma/Date/15.9.5.21-7.js: Ditto.

17792

* tests/mozilla/ecma/Date/15.9.5.21-8.js: Ditto.

17793

* tests/mozilla/ecma/Date/15.9.5.22-1.js: Ditto.

17794

* tests/mozilla/ecma/Date/15.9.5.22-2.js: Ditto.

17795

* tests/mozilla/ecma/Date/15.9.5.22-3.js: Ditto.

17796

* tests/mozilla/ecma/Date/15.9.5.22-4.js: Ditto.

17797

* tests/mozilla/ecma/Date/15.9.5.22-5.js: Ditto.

17798

* tests/mozilla/ecma/Date/15.9.5.22-6.js: Ditto.

17799

* tests/mozilla/ecma/Date/15.9.5.22-7.js: Ditto.

17800

* tests/mozilla/ecma/Date/15.9.5.22-8.js: Ditto.

17801

* tests/mozilla/ecma/Date/15.9.5.23-4.js: Ditto.

17802

* tests/mozilla/ecma/Date/15.9.5.23-5.js: Ditto.

17803

* tests/mozilla/ecma/Date/15.9.5.23-6.js: Ditto.

17804

* tests/mozilla/ecma/Date/15.9.5.23-7.js: Ditto.

17805

* tests/mozilla/ecma/Date/15.9.5.23-8.js: Ditto.

17806

* tests/mozilla/ecma/Date/15.9.5.23-9.js: Ditto.

17807

* tests/mozilla/ecma/Date/15.9.5.5.js: Ditto.

17808

* tests/mozilla/ecma/Date/15.9.5.6.js: Ditto.

17809

* tests/mozilla/ecma/Date/15.9.5.7.js: Ditto.

17810

* tests/mozilla/ecma/Date/15.9.5.8.js: Ditto.

17811

* tests/mozilla/ecma/Date/15.9.5.9.js: Ditto.

17812

17813

2005-08-08 Darin Adler <darin@apple.com>

17814

17815

- forgot to delete an obsolete file

17816

17817

* kjs/object_wrapper.h: Deleted.

17818

17819

2005-08-07 Darin Adler <darin@apple.com>

17820

17821

- fixed two problems compiling with gcc 4.0

17822

17823

* kjs/array_object.cpp: (ArrayProtoFuncImp::callAsFunction): Initialized a

17824

variable to quiet an erroneous warning.

17825

* kjs/date_object.cpp: (KJS::makeTime): Removed extraneous KJS:: prefix.

17826

17827

2005-08-07 Darin Adler <darin@apple.com>

17828

17829

Rubber stamped by Maciej.

17830

17831

- fixed http://bugs.webkit.org/show_bug.cgi?id=4313

17832

eliminate KJS::Value and KJS::Object smart pointer wrappers (for simplicity and speed)

17833

17834

* JavaScriptCore.xcodeproj/project.pbxproj: Removed object_wrapper.h.

17835

17836

Global replaces and other wonderful stuff.

17837

17838

* bindings/NP_jsobject.cpp:

17839

(_NPN_Invoke):

17840

(_NPN_Evaluate):

17841

(_NPN_GetProperty):

17842

(_NPN_SetProperty):

17843

(_NPN_HasMethod):

17844

(_NPN_SetException):

17845

* bindings/c/c_instance.cpp:

17846

(KJS::Bindings::CInstance::CInstance):

17847

(KJS::Bindings::CInstance::invokeMethod):

17848

(KJS::Bindings::CInstance::invokeDefaultMethod):

17849

(KJS::Bindings::CInstance::defaultValue):

17850

(KJS::Bindings::CInstance::stringValue):

17851

(KJS::Bindings::CInstance::numberValue):

17852

(KJS::Bindings::CInstance::booleanValue):

17853

(KJS::Bindings::CInstance::valueOf):

17854

* bindings/c/c_instance.h:

17855

* bindings/c/c_runtime.cpp:

17856

(CField::valueFromInstance):

17857

(CField::setValueToInstance):

17858

* bindings/c/c_runtime.h:

17859

* bindings/c/c_utility.cpp:

17860

(convertNPStringToUTF16):

17861

(convertUTF8ToUTF16):

17862

(coerceValueToNPVariantStringType):

17863

(convertValueToNPVariant):

17864

(convertNPVariantToValue):

17865

* bindings/c/c_utility.h:

17866

* bindings/jni/jni_instance.cpp:

17867

(JavaInstance::stringValue):

17868

(JavaInstance::numberValue):

17869

(JavaInstance::booleanValue):

17870

(JavaInstance::invokeMethod):

17871

(JavaInstance::invokeDefaultMethod):

17872

(JavaInstance::defaultValue):

17873

(JavaInstance::valueOf):

17874

* bindings/jni/jni_instance.h:

17875

* bindings/jni/jni_jsobject.cpp:

17876

(JSObject::invoke):

17877

(JSObject::call):

17878

(JSObject::eval):

17879

(JSObject::getMember):

17880

(JSObject::getSlot):

17881

(JSObject::toString):

17882

(JSObject::convertValueToJObject):

17883

(JSObject::convertJObjectToValue):

17884

(JSObject::listFromJArray):

17885

* bindings/jni/jni_jsobject.h:

17886

* bindings/jni/jni_objc.mm:

17887

(KJS::Bindings::dispatchJNICall):

17888

* bindings/jni/jni_runtime.cpp:

17889

(JavaArray::convertJObjectToArray):

17890

(JavaField::dispatchValueFromInstance):

17891

(JavaField::valueFromInstance):

17892

(JavaField::dispatchSetValueToInstance):

17893

(JavaField::setValueToInstance):

17894

(JavaArray::setValueAt):

17895

(JavaArray::valueAt):

17896

* bindings/jni/jni_runtime.h:

17897

(KJS::Bindings::JavaString::ustring):

17898

* bindings/jni/jni_utility.cpp:

17899

(KJS::Bindings::getJavaVM):

17900

(KJS::Bindings::getJNIEnv):

17901

(KJS::Bindings::getMethodID):

17902

(KJS::Bindings::callJNIVoidMethod):

17903

(KJS::Bindings::callJNIObjectMethod):

17904

(KJS::Bindings::callJNIBooleanMethod):

17905

(KJS::Bindings::callJNIStaticBooleanMethod):

17906

(KJS::Bindings::callJNIByteMethod):

17907

(KJS::Bindings::callJNICharMethod):

17908

(KJS::Bindings::callJNIShortMethod):

17909

(KJS::Bindings::callJNIIntMethod):

17910

(KJS::Bindings::callJNILongMethod):

17911

(KJS::Bindings::callJNIFloatMethod):

17912

(KJS::Bindings::callJNIDoubleMethod):

17913

(KJS::Bindings::callJNIVoidMethodA):

17914

(KJS::Bindings::callJNIObjectMethodA):

17915

(KJS::Bindings::callJNIByteMethodA):

17916

(KJS::Bindings::callJNICharMethodA):

17917

(KJS::Bindings::callJNIShortMethodA):

17918

(KJS::Bindings::callJNIIntMethodA):

17919

(KJS::Bindings::callJNILongMethodA):

17920

(KJS::Bindings::callJNIFloatMethodA):

17921

(KJS::Bindings::callJNIDoubleMethodA):

17922

(KJS::Bindings::callJNIBooleanMethodA):

17923

(KJS::Bindings::callJNIVoidMethodIDA):

17924

(KJS::Bindings::callJNIObjectMethodIDA):

17925

(KJS::Bindings::callJNIByteMethodIDA):

17926

(KJS::Bindings::callJNICharMethodIDA):

17927

(KJS::Bindings::callJNIShortMethodIDA):

17928

(KJS::Bindings::callJNIIntMethodIDA):

17929

(KJS::Bindings::callJNILongMethodIDA):

17930

(KJS::Bindings::callJNIFloatMethodIDA):

17931

(KJS::Bindings::callJNIDoubleMethodIDA):

17932

(KJS::Bindings::callJNIBooleanMethodIDA):

17933

(KJS::Bindings::getCharactersFromJString):

17934

(KJS::Bindings::releaseCharactersForJString):

17935

(KJS::Bindings::getCharactersFromJStringInEnv):

17936

(KJS::Bindings::releaseCharactersForJStringInEnv):

17937

(KJS::Bindings::getUCharactersFromJStringInEnv):

17938

(KJS::Bindings::releaseUCharactersForJStringInEnv):

17939

(KJS::Bindings::JNITypeFromClassName):

17940

(KJS::Bindings::signatureFromPrimitiveType):

17941

(KJS::Bindings::JNITypeFromPrimitiveType):

17942

(KJS::Bindings::getJNIField):

17943

(KJS::Bindings::convertValueToJValue):

17944

* bindings/jni/jni_utility.h:

17945

* bindings/objc/WebScriptObject.mm:

17946

(_didExecute):

17947

(-[WebScriptObject _initializeWithObjectImp:originExecutionContext:Bindings::executionContext:Bindings::]):

17948

(-[WebScriptObject _initWithObjectImp:originExecutionContext:Bindings::executionContext:Bindings::]):

17949

(-[WebScriptObject _imp]):

17950

(-[WebScriptObject _executionContext]):

17951

(-[WebScriptObject _setExecutionContext:]):

17952

(-[WebScriptObject _originExecutionContext]):

17953

(-[WebScriptObject _setOriginExecutionContext:]):

17954

(+[WebScriptObject throwException:]):

17955

(listFromNSArray):

17956

(-[WebScriptObject callWebScriptMethod:withArguments:]):

17957

(-[WebScriptObject evaluateWebScript:]):

17958

(-[WebScriptObject setValue:forKey:]):

17959

(-[WebScriptObject valueForKey:]):

17960

(-[WebScriptObject removeWebScriptKey:]):

17961

(-[WebScriptObject stringRepresentation]):

17962

(-[WebScriptObject webScriptValueAtIndex:]):

17963

(-[WebScriptObject setException:]):

17964

(+[WebScriptObject _convertValueToObjcValue:originExecutionContext:executionContext:Bindings::]):

17965

* bindings/objc/WebScriptObjectPrivate.h:

17966

* bindings/objc/objc_class.h:

17967

* bindings/objc/objc_class.mm:

17968

(KJS::Bindings::ObjcClass::fallbackObject):

17969

* bindings/objc/objc_instance.h:

17970

* bindings/objc/objc_instance.mm:

17971

(ObjcInstance::invokeMethod):

17972

(ObjcInstance::invokeDefaultMethod):

17973

(ObjcInstance::setValueOfField):

17974

(ObjcInstance::setValueOfUndefinedField):

17975

(ObjcInstance::getValueOfField):

17976

(ObjcInstance::getValueOfUndefinedField):

17977

(ObjcInstance::defaultValue):

17978

(ObjcInstance::stringValue):

17979

(ObjcInstance::numberValue):

17980

(ObjcInstance::booleanValue):

17981

(ObjcInstance::valueOf):

17982

* bindings/objc/objc_runtime.h:

17983

* bindings/objc/objc_runtime.mm:

17984

(ObjcField::valueFromInstance):

17985

(convertValueToObjcObject):

17986

(ObjcField::setValueToInstance):

17987

(ObjcArray::setValueAt):

17988

(ObjcArray::valueAt):

17989

(ObjcFallbackObjectImp::put):

17990

(ObjcFallbackObjectImp::callAsFunction):

17991

(ObjcFallbackObjectImp::defaultValue):

17992

* bindings/objc/objc_utility.h:

17993

* bindings/objc/objc_utility.mm:

17994

(Bindings::JSMethodNameToObjCMethodName):

17995

(Bindings::convertValueToObjcValue):

17996

(Bindings::convertNSStringToString):

17997

(Bindings::convertObjcValueToValue):

17998

(Bindings::objcValueTypeForType):

17999

(Bindings::createObjcInstanceForValue):

18000

* bindings/runtime.cpp:

18001

(Instance::getValueOfField):

18002

(Instance::setValueOfField):

18003

(Instance::createRuntimeObject):

18004

(Instance::createLanguageInstanceForValue):

18005

* bindings/runtime.h:

18006

(KJS::Bindings::Constructor::~Constructor):

18007

(KJS::Bindings::Field::~Field):

18008

(KJS::Bindings::MethodList::MethodList):

18009

(KJS::Bindings::Class::fallbackObject):

18010

(KJS::Bindings::Class::~Class):

18011

(KJS::Bindings::Instance::Instance):

18012

(KJS::Bindings::Instance::getValueOfUndefinedField):

18013

(KJS::Bindings::Instance::supportsSetValueOfUndefinedField):

18014

(KJS::Bindings::Instance::setValueOfUndefinedField):

18015

(KJS::Bindings::Instance::valueOf):

18016

(KJS::Bindings::Instance::setExecutionContext):

18017

(KJS::Bindings::Instance::~Instance):

18018

(KJS::Bindings::Array::~Array):

18019

* bindings/runtime_array.cpp:

18020

(RuntimeArrayImp::RuntimeArrayImp):

18021

(RuntimeArrayImp::lengthGetter):

18022

(RuntimeArrayImp::indexGetter):

18023

(RuntimeArrayImp::put):

18024

* bindings/runtime_array.h:

18025

* bindings/runtime_method.cpp:

18026

(RuntimeMethodImp::lengthGetter):

18027

(RuntimeMethodImp::callAsFunction):

18028

* bindings/runtime_method.h:

18029

* bindings/runtime_object.cpp:

18030

(RuntimeObjectImp::fallbackObjectGetter):

18031

(RuntimeObjectImp::fieldGetter):

18032

(RuntimeObjectImp::methodGetter):

18033

(RuntimeObjectImp::getOwnPropertySlot):

18034

(RuntimeObjectImp::put):

18035

(RuntimeObjectImp::defaultValue):

18036

(RuntimeObjectImp::callAsFunction):

18037

* bindings/runtime_object.h:

18038

* kjs/array_instance.h:

18039

* kjs/array_object.cpp:

18040

(ArrayInstanceImp::ArrayInstanceImp):

18041

(ArrayInstanceImp::lengthGetter):

18042

(ArrayInstanceImp::getOwnPropertySlot):

18043

(ArrayInstanceImp::put):

18044

(ArrayInstanceImp::propList):

18045

(ArrayInstanceImp::setLength):

18046

(compareByStringForQSort):

18047

(compareWithCompareFunctionForQSort):

18048

(ArrayInstanceImp::sort):

18049

(ArrayInstanceImp::pushUndefinedObjectsToEnd):

18050

(ArrayPrototypeImp::ArrayPrototypeImp):

18051

(ArrayProtoFuncImp::ArrayProtoFuncImp):

18052

(ArrayProtoFuncImp::callAsFunction):

18053

(ArrayObjectImp::ArrayObjectImp):

18054

(ArrayObjectImp::construct):

18055

(ArrayObjectImp::callAsFunction):

18056

* kjs/array_object.h:

18057

* kjs/bool_object.cpp:

18058

(BooleanPrototypeImp::BooleanPrototypeImp):

18059

(BooleanProtoFuncImp::BooleanProtoFuncImp):

18060

(BooleanProtoFuncImp::callAsFunction):

18061

(BooleanObjectImp::BooleanObjectImp):

18062

(BooleanObjectImp::construct):

18063

(BooleanObjectImp::callAsFunction):

18064

* kjs/bool_object.h:

18065

* kjs/collector.cpp:

18066

(KJS::Collector::markStackObjectsConservatively):

18067

(KJS::Collector::collect):

18068

(KJS::className):

18069

* kjs/completion.h:

18070

(KJS::Completion::Completion):

18071

(KJS::Completion::value):

18072

(KJS::Completion::isValueCompletion):

18073

* kjs/context.h:

18074

(KJS::ContextImp::variableObject):

18075

(KJS::ContextImp::setVariableObject):

18076

(KJS::ContextImp::thisValue):

18077

(KJS::ContextImp::activationObject):

18078

(KJS::ContextImp::pushScope):

18079

* kjs/date_object.cpp:

18080

(formatLocaleDate):

18081

(KJS::timeFromArgs):

18082

(KJS::DatePrototypeImp::DatePrototypeImp):

18083

(KJS::DateProtoFuncImp::DateProtoFuncImp):

18084

(KJS::DateProtoFuncImp::callAsFunction):

18085

(KJS::DateObjectImp::DateObjectImp):

18086

(KJS::DateObjectImp::construct):

18087

(KJS::DateObjectImp::callAsFunction):

18088

(KJS::DateObjectFuncImp::DateObjectFuncImp):

18089

(KJS::DateObjectFuncImp::callAsFunction):

18090

(KJS::parseDate):

18091

(KJS::KRFCDate_parseDate):

18092

(KJS::timeClip):

18093

* kjs/date_object.h:

18094

* kjs/debugger.cpp:

18095

(Debugger::exception):

18096

(Debugger::callEvent):

18097

(Debugger::returnEvent):

18098

* kjs/debugger.h:

18099

* kjs/error_object.cpp:

18100

(ErrorPrototypeImp::ErrorPrototypeImp):

18101

(ErrorProtoFuncImp::ErrorProtoFuncImp):

18102

(ErrorProtoFuncImp::callAsFunction):

18103

(ErrorObjectImp::ErrorObjectImp):

18104

(ErrorObjectImp::construct):

18105

(ErrorObjectImp::callAsFunction):

18106

(NativeErrorPrototypeImp::NativeErrorPrototypeImp):

18107

(NativeErrorImp::NativeErrorImp):

18108

(NativeErrorImp::construct):

18109

(NativeErrorImp::callAsFunction):

18110

* kjs/error_object.h:

18111

* kjs/function.cpp:

18112

(KJS::FunctionImp::FunctionImp):

18113

(KJS::FunctionImp::callAsFunction):

18114

(KJS::FunctionImp::processParameters):

18115

(KJS::FunctionImp::argumentsGetter):

18116

(KJS::FunctionImp::lengthGetter):

18117

(KJS::FunctionImp::put):

18118

(KJS::DeclaredFunctionImp::DeclaredFunctionImp):

18119

(KJS::DeclaredFunctionImp::construct):

18120

(KJS::ArgumentsImp::ArgumentsImp):

18121

(KJS::ArgumentsImp::mappedIndexGetter):

18122

(KJS::ArgumentsImp::put):

18123

(KJS::ActivationImp::argumentsGetter):

18124

(KJS::GlobalFuncImp::GlobalFuncImp):

18125

(KJS::encode):

18126

(KJS::decode):

18127

(KJS::GlobalFuncImp::callAsFunction):

18128

* kjs/function.h:

18129

* kjs/function_object.cpp:

18130

(FunctionPrototypeImp::FunctionPrototypeImp):

18131

(FunctionPrototypeImp::callAsFunction):

18132

(FunctionProtoFuncImp::FunctionProtoFuncImp):

18133

(FunctionProtoFuncImp::callAsFunction):

18134

(FunctionObjectImp::FunctionObjectImp):

18135

(FunctionObjectImp::construct):

18136

(FunctionObjectImp::callAsFunction):

18137

* kjs/function_object.h:

18138

* kjs/internal.cpp:

18139

(KJS::UndefinedImp::toPrimitive):

18140

(KJS::UndefinedImp::toObject):

18141

(KJS::NullImp::toPrimitive):

18142

(KJS::NullImp::toObject):

18143

(KJS::BooleanImp::toPrimitive):

18144

(KJS::BooleanImp::toObject):

18145

(KJS::StringImp::toPrimitive):

18146

(KJS::StringImp::toObject):

18147

(KJS::NumberImp::toPrimitive):

18148

(KJS::NumberImp::toObject):

18149

(KJS::NumberImp::getUInt32):

18150

(KJS::LabelStack::push):

18151

(KJS::ContextImp::ContextImp):

18152

(KJS::InterpreterImp::globalInit):

18153

(KJS::InterpreterImp::globalClear):

18154

(KJS::InterpreterImp::InterpreterImp):

18155

(KJS::InterpreterImp::initGlobalObject):

18156

(KJS::InterpreterImp::clear):

18157

(KJS::InterpreterImp::mark):

18158

(KJS::InterpreterImp::evaluate):

18159

(KJS::InternalFunctionImp::hasInstance):

18160

(KJS::roundValue):

18161

(KJS::printInfo):

18162

* kjs/internal.h:

18163

(KJS::InterpreterImp::builtinObject):

18164

(KJS::InterpreterImp::builtinFunction):

18165

(KJS::InterpreterImp::builtinArray):

18166

(KJS::InterpreterImp::builtinBoolean):

18167

(KJS::InterpreterImp::builtinString):

18168

(KJS::InterpreterImp::builtinNumber):

18169

(KJS::InterpreterImp::builtinDate):

18170

(KJS::InterpreterImp::builtinRegExp):

18171

(KJS::InterpreterImp::builtinError):

18172

(KJS::InterpreterImp::builtinObjectPrototype):

18173

(KJS::InterpreterImp::builtinFunctionPrototype):

18174

(KJS::InterpreterImp::builtinArrayPrototype):

18175

(KJS::InterpreterImp::builtinBooleanPrototype):

18176

(KJS::InterpreterImp::builtinStringPrototype):

18177

(KJS::InterpreterImp::builtinNumberPrototype):

18178

(KJS::InterpreterImp::builtinDatePrototype):

18179

(KJS::InterpreterImp::builtinRegExpPrototype):

18180

(KJS::InterpreterImp::builtinErrorPrototype):

18181

(KJS::InterpreterImp::builtinEvalError):

18182

(KJS::InterpreterImp::builtinRangeError):

18183

(KJS::InterpreterImp::builtinReferenceError):

18184

(KJS::InterpreterImp::builtinSyntaxError):

18185

(KJS::InterpreterImp::builtinTypeError):

18186

(KJS::InterpreterImp::builtinURIError):

18187

(KJS::InterpreterImp::builtinEvalErrorPrototype):

18188

(KJS::InterpreterImp::builtinRangeErrorPrototype):

18189

(KJS::InterpreterImp::builtinReferenceErrorPrototype):

18190

(KJS::InterpreterImp::builtinSyntaxErrorPrototype):

18191

(KJS::InterpreterImp::builtinTypeErrorPrototype):

18192

(KJS::InterpreterImp::builtinURIErrorPrototype):

18193

* kjs/interpreter.cpp:

18194

(Context::variableObject):

18195

(Context::thisValue):

18196

(Interpreter::Interpreter):

18197

(Interpreter::globalObject):

18198

(Interpreter::evaluate):

18199

(Interpreter::builtinObject):

18200

(Interpreter::builtinFunction):

18201

(Interpreter::builtinArray):

18202

(Interpreter::builtinBoolean):

18203

(Interpreter::builtinString):

18204

(Interpreter::builtinNumber):

18205

(Interpreter::builtinDate):

18206

(Interpreter::builtinRegExp):

18207

(Interpreter::builtinError):

18208

(Interpreter::builtinObjectPrototype):

18209

(Interpreter::builtinFunctionPrototype):

18210

(Interpreter::builtinArrayPrototype):

18211

(Interpreter::builtinBooleanPrototype):

18212

(Interpreter::builtinStringPrototype):

18213

(Interpreter::builtinNumberPrototype):

18214

(Interpreter::builtinDatePrototype):

18215

(Interpreter::builtinRegExpPrototype):

18216

(Interpreter::builtinErrorPrototype):

18217

(Interpreter::builtinEvalError):

18218

(Interpreter::builtinRangeError):

18219

(Interpreter::builtinReferenceError):

18220

(Interpreter::builtinSyntaxError):

18221

(Interpreter::builtinTypeError):

18222

(Interpreter::builtinURIError):

18223

(Interpreter::builtinEvalErrorPrototype):

18224

(Interpreter::builtinRangeErrorPrototype):

18225

(Interpreter::builtinReferenceErrorPrototype):

18226

(Interpreter::builtinSyntaxErrorPrototype):

18227

(Interpreter::builtinTypeErrorPrototype):

18228

(Interpreter::builtinURIErrorPrototype):

18229

(Interpreter::createLanguageInstanceForValue):

18230

* kjs/interpreter.h:

18231

(KJS::Interpreter::isGlobalObject):

18232

(KJS::ExecState::setException):

18233

(KJS::ExecState::clearException):

18234

(KJS::ExecState::exception):

18235

(KJS::ExecState::hadException):

18236

(KJS::ExecState::ExecState):

18237

* kjs/list.cpp:

18238

(KJS::List::at):

18239

* kjs/list.h:

18240

(KJS::List::operator[]):

18241

(KJS::ListIterator::operator->):

18242

(KJS::ListIterator::operator*):

18243

(KJS::ListIterator::operator++):

18244

(KJS::ListIterator::operator--):

18245

* kjs/lookup.h:

18246

(KJS::staticFunctionGetter):

18247

(KJS::staticValueGetter):

18248

(KJS::lookupPut):

18249

(KJS::cacheGlobalObject):

18250

* kjs/math_object.cpp:

18251

(MathObjectImp::getValueProperty):

18252

(MathFuncImp::MathFuncImp):

18253

(MathFuncImp::callAsFunction):

18254

* kjs/math_object.h:

18255

* kjs/nodes.cpp:

18256

(Node::evaluateReference):

18257

(Node::throwError):

18258

(Node::setExceptionDetailsIfNeeded):

18259

(NullNode::evaluate):

18260

(BooleanNode::evaluate):

18261

(NumberNode::evaluate):

18262

(StringNode::evaluate):

18263

(RegExpNode::evaluate):

18264

(ThisNode::evaluate):

18265

(ResolveNode::evaluate):

18266

(ResolveNode::evaluateReference):

18267

(GroupNode::evaluate):

18268

(ElementNode::evaluate):

18269

(ArrayNode::evaluate):

18270

(ObjectLiteralNode::evaluate):

18271

(PropertyValueNode::evaluate):

18272

(PropertyNode::evaluate):

18273

(AccessorNode1::evaluate):

18274

(AccessorNode1::evaluateReference):

18275

(AccessorNode2::evaluate):

18276

(AccessorNode2::evaluateReference):

18277

(ArgumentListNode::evaluate):

18278

(ArgumentListNode::evaluateList):

18279

(ArgumentsNode::evaluate):

18280

(NewExprNode::evaluate):

18281

(FunctionCallNode::evaluate):

18282

(PostfixNode::evaluate):

18283

(DeleteNode::evaluate):

18284

(VoidNode::evaluate):

18285

(TypeOfNode::evaluate):

18286

(PrefixNode::evaluate):

18287

(UnaryPlusNode::evaluate):

18288

(NegateNode::evaluate):

18289

(BitwiseNotNode::evaluate):

18290

(LogicalNotNode::evaluate):

18291

(MultNode::evaluate):

18292

(AddNode::evaluate):

18293

(ShiftNode::evaluate):

18294

(RelationalNode::evaluate):

18295

(EqualNode::evaluate):

18296

(BitOperNode::evaluate):

18297

(BinaryLogicalNode::evaluate):

18298

(ConditionalNode::evaluate):

18299

(AssignNode::evaluate):

18300

(CommaNode::evaluate):

18301

(StatListNode::execute):

18302

(AssignExprNode::evaluate):

18303

(VarDeclNode::evaluate):

18304

(VarDeclNode::processVarDecls):

18305

(VarDeclListNode::evaluate):

18306

(ExprStatementNode::execute):

18307

(IfNode::execute):

18308

(DoWhileNode::execute):

18309

(WhileNode::execute):

18310

(ForNode::execute):

18311

(ForInNode::execute):

18312

(ContinueNode::execute):

18313

(BreakNode::execute):

18314

(ReturnNode::execute):

18315

(WithNode::execute):

18316

(CaseClauseNode::evaluate):

18317

(ClauseListNode::evaluate):

18318

(CaseBlockNode::evaluate):

18319

(CaseBlockNode::evalBlock):

18320

(SwitchNode::execute):

18321

(ThrowNode::execute):

18322

(CatchNode::execute):

18323

(TryNode::execute):

18324

(ParameterNode::evaluate):

18325

(FuncDeclNode::processFuncDecl):

18326

(FuncExprNode::evaluate):

18327

(SourceElementsNode::execute):

18328

* kjs/nodes.h:

18329

(KJS::StatementNode::evaluate):

18330

* kjs/number_object.cpp:

18331

(NumberPrototypeImp::NumberPrototypeImp):

18332

(NumberProtoFuncImp::NumberProtoFuncImp):

18333

(NumberProtoFuncImp::callAsFunction):

18334

(NumberObjectImp::NumberObjectImp):

18335

(NumberObjectImp::getValueProperty):

18336

(NumberObjectImp::construct):

18337

(NumberObjectImp::callAsFunction):

18338

* kjs/number_object.h:

18339

* kjs/object.cpp:

18340

(KJS::ObjectImp::call):

18341

(KJS::ObjectImp::mark):

18342

(KJS::ObjectImp::classInfo):

18343

(KJS::ObjectImp::get):

18344

(KJS::ObjectImp::getProperty):

18345

(KJS::ObjectImp::getPropertySlot):

18346

(KJS::ObjectImp::put):

18347

(KJS::ObjectImp::hasOwnProperty):

18348

(KJS::ObjectImp::defaultValue):

18349

(KJS::ObjectImp::findPropertyHashEntry):

18350

(KJS::ObjectImp::construct):

18351

(KJS::ObjectImp::callAsFunction):

18352

(KJS::ObjectImp::hasInstance):

18353

(KJS::ObjectImp::propList):

18354

(KJS::ObjectImp::toPrimitive):

18355

(KJS::ObjectImp::toNumber):

18356

(KJS::ObjectImp::toString):

18357

(KJS::ObjectImp::toObject):

18358

(KJS::ObjectImp::putDirect):

18359

(KJS::Error::create):

18360

(KJS::error):

18361

* kjs/object.h:

18362

(KJS::):

18363

(KJS::ObjectImp::getPropertySlot):

18364

(KJS::AllocatedValueImp::isObject):

18365

(KJS::ObjectImp::ObjectImp):

18366

(KJS::ObjectImp::internalValue):

18367

(KJS::ObjectImp::setInternalValue):

18368

(KJS::ObjectImp::prototype):

18369

(KJS::ObjectImp::setPrototype):

18370

(KJS::ObjectImp::inherits):

18371

* kjs/object_object.cpp:

18372

(ObjectPrototypeImp::ObjectPrototypeImp):

18373

(ObjectProtoFuncImp::ObjectProtoFuncImp):

18374

(ObjectProtoFuncImp::callAsFunction):

18375

(ObjectObjectImp::ObjectObjectImp):

18376

(ObjectObjectImp::construct):

18377

(ObjectObjectImp::callAsFunction):

18378

* kjs/object_object.h:

18379

* kjs/operations.cpp:

18380

(KJS::equal):

18381

(KJS::strictEqual):

18382

(KJS::relation):

18383

(KJS::add):

18384

(KJS::mult):

18385

* kjs/operations.h:

18386

* kjs/property_map.cpp:

18387

(KJS::PropertyMap::mark):

18388

(KJS::PropertyMap::addEnumerablesToReferenceList):

18389

(KJS::PropertyMap::addSparseArrayPropertiesToReferenceList):

18390

(KJS::PropertyMap::save):

18391

(KJS::PropertyMap::restore):

18392

* kjs/property_map.h:

18393

* kjs/property_slot.cpp:

18394

(KJS::PropertySlot::undefinedGetter):

18395

* kjs/property_slot.h:

18396

(KJS::PropertySlot::getValue):

18397

* kjs/protect.h:

18398

(KJS::gcUnprotectNullTolerant):

18399

(KJS::ProtectedValue::ProtectedValue):

18400

(KJS::ProtectedValue::~ProtectedValue):

18401

(KJS::ProtectedValue::operator=):

18402

(KJS::ProtectedValue::operator ValueImp *):

18403

(KJS::ProtectedValue::operator->):

18404

* kjs/protected_object.h:

18405

(KJS::ProtectedObject::ProtectedObject):

18406

(KJS::ProtectedObject::operator=):

18407

(KJS::ProtectedObject::operator ValueImp *):

18408

(KJS::ProtectedObject::operator ObjectImp *):

18409

(KJS::ProtectedObject::operator->):

18410

(KJS::ProtectedReference::ProtectedReference):

18411

(KJS::ProtectedReference::~ProtectedReference):

18412

(KJS::ProtectedReference::operator=):

18413

* kjs/protected_values.cpp:

18414

(KJS::ProtectedValues::getProtectCount):

18415

(KJS::ProtectedValues::increaseProtectCount):

18416

(KJS::ProtectedValues::insert):

18417

(KJS::ProtectedValues::decreaseProtectCount):

18418

* kjs/protected_values.h:

18419

* kjs/reference.cpp:

18420

(KJS::Reference::Reference):

18421

(KJS::Reference::makeValueReference):

18422

(KJS::Reference::getBase):

18423

(KJS::Reference::getValue):

18424

(KJS::Reference::putValue):

18425

(KJS::Reference::deleteValue):

18426

* kjs/reference.h:

18427

(KJS::Reference::baseIfMutable):

18428

* kjs/regexp_object.cpp:

18429

(RegExpPrototypeImp::RegExpPrototypeImp):

18430

(RegExpProtoFuncImp::RegExpProtoFuncImp):

18431

(RegExpProtoFuncImp::callAsFunction):

18432

(RegExpObjectImp::RegExpObjectImp):

18433

(RegExpObjectImp::arrayOfMatches):

18434

(RegExpObjectImp::backrefGetter):

18435

(RegExpObjectImp::construct):

18436

(RegExpObjectImp::callAsFunction):

18437

* kjs/regexp_object.h:

18438

* kjs/string_object.cpp:

18439

(StringInstanceImp::lengthGetter):

18440

(StringInstanceImp::indexGetter):

18441

(StringInstanceImp::getOwnPropertySlot):

18442

(StringInstanceImp::put):

18443

(StringPrototypeImp::StringPrototypeImp):

18444

(StringProtoFuncImp::StringProtoFuncImp):

18445

(regExpIsGlobal):

18446

(replace):

18447

(StringProtoFuncImp::callAsFunction):

18448

(StringObjectImp::StringObjectImp):

18449

(StringObjectImp::construct):

18450

(StringObjectImp::callAsFunction):

18451

(StringObjectFuncImp::StringObjectFuncImp):

18452

(StringObjectFuncImp::callAsFunction):

18453

* kjs/string_object.h:

18454

* kjs/testkjs.cpp:

18455

(TestFunctionImp::callAsFunction):

18456

(VersionFunctionImp::callAsFunction):

18457

(main):

18458

* kjs/value.cpp:

18459

(KJS::AllocatedValueImp::operator new):

18460

(KJS::AllocatedValueImp::getUInt32):

18461

(KJS::ValueImp::toInteger):

18462

(KJS::ValueImp::toInt32):

18463

(KJS::ValueImp::toUInt32):

18464

(KJS::ValueImp::toUInt16):

18465

(KJS::ValueImp::toObject):

18466

(KJS::AllocatedValueImp::getBoolean):

18467

(KJS::AllocatedValueImp::getNumber):

18468

(KJS::AllocatedValueImp::getString):

18469

(KJS::AllocatedValueImp::getObject):

18470

(KJS::jsString):

18471

(KJS::jsNumber):

18472

(KJS::ConstantValues::init):

18473

(KJS::ConstantValues::clear):

18474

(KJS::ConstantValues::mark):

18475

* kjs/value.h:

18476

(KJS::):

18477

(KJS::jsUndefined):

18478

(KJS::jsNull):

18479

(KJS::jsBoolean):

18480

(KJS::jsNaN):

18481

(KJS::ValueImp::ValueImp):

18482

(KJS::ValueImp::~ValueImp):

18483

(KJS::AllocatedValueImp::AllocatedValueImp):

18484

(KJS::AllocatedValueImp::~AllocatedValueImp):

18485

(KJS::AllocatedValueImp::isBoolean):

18486

(KJS::AllocatedValueImp::isNumber):

18487

(KJS::AllocatedValueImp::isString):

18488

(KJS::AllocatedValueImp::isObject):

18489

(KJS::AllocatedValueImp::marked):

18490

(KJS::AllocatedValueImp::mark):

18491

(KJS::ValueImp::downcast):

18492

(KJS::ValueImp::isUndefined):

18493

(KJS::ValueImp::isNull):

18494

(KJS::ValueImp::isUndefinedOrNull):

18495

(KJS::ValueImp::isBoolean):

18496

(KJS::ValueImp::isNumber):

18497

(KJS::ValueImp::isString):

18498

(KJS::ValueImp::isObject):

18499

(KJS::ValueImp::getBoolean):

18500

(KJS::ValueImp::getNumber):

18501

(KJS::ValueImp::getString):

18502

(KJS::ValueImp::getObject):

18503

(KJS::ValueImp::getUInt32):

18504

(KJS::ValueImp::mark):

18505

(KJS::ValueImp::marked):

18506

(KJS::ValueImp::type):

18507

(KJS::ValueImp::toPrimitive):

18508

(KJS::ValueImp::toBoolean):

18509

(KJS::ValueImp::toNumber):

18510

(KJS::ValueImp::toString):

18511

(KJS::jsZero):

18512

(KJS::jsOne):

18513

(KJS::jsTwo):

18514

(KJS::Undefined):

18515

(KJS::Null):

18516

(KJS::Boolean):

18517

(KJS::Number):

18518

(KJS::String):

18519

18520

2005-08-06 Maciej Stachowiak <mjs@apple.com>

18521

18522

Reviewed by Darin.

18523

18524

Change over to the new PropertySlot mechanism for property

18525

lookup. This allows the elimination of hasOwnProperty

18526

methods. Also did some of the performance tuning enabled by this

18527

(but not yet all the possible improvements for function calls,

18528

assignment, ++, and so forth). And also much code cleanup.

18529

18530

Net result is about a 2% speedup on the JS iBench.

18531

18532

Also redid Geoff's fix for the chrashing applet by avoiding a NULL

18533

prototype in the bindings code and using the default of Null()

18534

instead.

18535

18536

* JavaScriptCore.xcodeproj/project.pbxproj:

18537

* bindings/objc/objc_runtime.h:

18538

* bindings/objc/objc_runtime.mm:

18539

(ObjcFallbackObjectImp::ObjcFallbackObjectImp):

18540

(ObjcFallbackObjectImp::getOwnPropertySlot):

18541

* bindings/runtime_array.cpp:

18542

(RuntimeArrayImp::lengthGetter):

18543

(RuntimeArrayImp::indexGetter):

18544

(RuntimeArrayImp::getOwnPropertySlot):

18545

* bindings/runtime_array.h:

18546

* bindings/runtime_method.cpp:

18547

(RuntimeMethodImp::lengthGetter):

18548

(RuntimeMethodImp::getOwnPropertySlot):

18549

* bindings/runtime_method.h:

18550

* bindings/runtime_object.cpp:

18551

(RuntimeObjectImp::RuntimeObjectImp):

18552

(RuntimeObjectImp::fallbackObjectGetter):

18553

(RuntimeObjectImp::fieldGetter):

18554

(RuntimeObjectImp::methodGetter):

18555

(RuntimeObjectImp::getOwnPropertySlot):

18556

* bindings/runtime_object.h:

18557

* bindings/runtime_root.h:

18558

* kjs/array_instance.h:

18559

* kjs/array_object.cpp:

18560

(ArrayInstanceImp::lengthGetter):

18561

(ArrayInstanceImp::getOwnPropertySlot):

18562

(ArrayPrototypeImp::getOwnPropertySlot):

18563

* kjs/array_object.h:

18564

* kjs/date_object.cpp:

18565

(DatePrototypeImp::getOwnPropertySlot):

18566

* kjs/date_object.h:

18567

* kjs/function.cpp:

18568

(KJS::FunctionImp::argumentsGetter):

18569

(KJS::FunctionImp::lengthGetter):

18570

(KJS::FunctionImp::getOwnPropertySlot):

18571

(KJS::FunctionImp::put):

18572

(KJS::FunctionImp::deleteProperty):

18573

(KJS::ArgumentsImp::mappedIndexGetter):

18574

(KJS::ArgumentsImp::getOwnPropertySlot):

18575

(KJS::ActivationImp::argumentsGetter):

18576

(KJS::ActivationImp::getArgumentsGetter):

18577

(KJS::ActivationImp::getOwnPropertySlot):

18578

(KJS::ActivationImp::deleteProperty):

18579

* kjs/function.h:

18580

* kjs/internal.cpp:

18581

(InterpreterImp::InterpreterImp):

18582

(InterpreterImp::initGlobalObject):

18583

(InterpreterImp::~InterpreterImp):

18584

(InterpreterImp::evaluate):

18585

* kjs/internal.h:

18586

(KJS::InterpreterImp::globalExec):

18587

* kjs/interpreter.cpp:

18588

(Interpreter::Interpreter):

18589

(Interpreter::createLanguageInstanceForValue):

18590

* kjs/interpreter.h:

18591

(KJS::Interpreter::argumentsIdentifier):

18592

(KJS::Interpreter::specialPrototypeIdentifier):

18593

* kjs/lookup.h:

18594

(KJS::staticFunctionGetter):

18595

(KJS::staticValueGetter):

18596

(KJS::getStaticPropertySlot):

18597

(KJS::getStaticFunctionSlot):

18598

(KJS::getStaticValueSlot):

18599

* kjs/math_object.cpp:

18600

(MathObjectImp::getOwnPropertySlot):

18601

* kjs/math_object.h:

18602

* kjs/nodes.cpp:

18603

(ResolveNode::evaluate):

18604

(ResolveNode::evaluateReference):

18605

(AccessorNode1::evaluate):

18606

(AccessorNode2::evaluate):

18607

* kjs/number_object.cpp:

18608

(NumberObjectImp::getOwnPropertySlot):

18609

* kjs/number_object.h:

18610

* kjs/object.cpp:

18611

(KJS::ObjectImp::get):

18612

(KJS::ObjectImp::getProperty):

18613

(KJS::ObjectImp::getPropertySlot):

18614

(KJS::ObjectImp::getOwnPropertySlot):

18615

(KJS::ObjectImp::put):

18616

(KJS::ObjectImp::hasProperty):

18617

(KJS::ObjectImp::hasOwnProperty):

18618

* kjs/object.h:

18619

(KJS::ObjectImp::getDirectLocation):

18620

(KJS::ObjectImp::getPropertySlot):

18621

(KJS::ObjectImp::getOwnPropertySlot):

18622

* kjs/object_wrapper.h: Added.

18623

(KJS::):

18624

(KJS::Object::Object):

18625

(KJS::Object::operator ObjectImp *):

18626

* kjs/property_map.cpp:

18627

(KJS::PropertyMap::getLocation):

18628

* kjs/property_map.h:

18629

* kjs/property_slot.cpp: Added.

18630

(KJS::PropertySlot::undefinedGetter):

18631

* kjs/property_slot.h: Added.

18632

(KJS::PropertySlot::isSet):

18633

(KJS::PropertySlot::getValue):

18634

(KJS::PropertySlot::setValueSlot):

18635

(KJS::PropertySlot::setStaticEntry):

18636

(KJS::PropertySlot::setCustom):

18637

(KJS::PropertySlot::setCustomIndex):

18638

(KJS::PropertySlot::setUndefined):

18639

(KJS::PropertySlot::slotBase):

18640

(KJS::PropertySlot::staticEntry):

18641

(KJS::PropertySlot::index):

18642

(KJS::PropertySlot::):

18643

* kjs/protect.h:

18644

* kjs/protected_object.h: Added.

18645

(KJS::ProtectedObject::ProtectedObject):

18646

(KJS::ProtectedObject::~ProtectedObject):

18647

(KJS::ProtectedObject::operator=):

18648

(KJS::ProtectedReference::ProtectedReference):

18649

(KJS::ProtectedReference::~ProtectedReference):

18650

(KJS::ProtectedReference::operator=):

18651

* kjs/reference.h:

18652

* kjs/reference_list.cpp:

18653

* kjs/regexp_object.cpp:

18654

(RegExpObjectImp::backrefGetter):

18655

(RegExpObjectImp::getOwnPropertySlot):

18656

* kjs/regexp_object.h:

18657

* kjs/string_object.cpp:

18658

(StringInstanceImp::lengthGetter):

18659

(StringInstanceImp::indexGetter):

18660

(StringInstanceImp::getOwnPropertySlot):

18661

(StringPrototypeImp::getOwnPropertySlot):

18662

* kjs/string_object.h:

18663

18664

2005-08-05 Adele Peterson <adele@apple.com>

18665

18666

Reviewed by Darin.

18667

18668

* JavaScriptCore.xcodeproj/project.pbxproj: Unchecked 'statics are thread safe' option.

18669

18670

2005-08-05 Geoffrey Garen <ggaren@apple.com>

18671

18672

-fixed <rdar://problem/4207220> REGRESSION (DENVER): Crash occurs

18673

after clicking on Hangman applet

18674

18675

Reviewed by darin.

18676

18677

* kjs/object.cpp:

18678

(KJS::ObjectImp::hasProperty): added check for null prototype.

18679

18680

FIXME: The long-term plan is to make runtime objects use JS Null()

18681

instead of null pointers, which will allow us to eliminate null

18682

checks, improving performance.

18683

18684

2005-08-05 Geoffrey Garen <ggaren@apple.com>

18685

18686

Fix by darin, reviewed by me.

18687

18688

- rolled in fix for: <rdar://problem/4161606> JavaScript regular

18689

expressions with certain ranges of Unicode characters cause a crash

18690

18691

Test cases added:

18692

18693

* layout-tests/fast/js/regexp-big-unicode-ranges-expected.txt: Added.

18694

* layout-tests/fast/js/regexp-big-unicode-ranges.html: Added.

18695

18696

* pcre/pcre.c:

18697

(compile_branch): added checks for characters > 255

18698

18699

2005-08-04 Maciej Stachowiak <mjs@apple.com>

18700

18701

- updated expected test results now that we no longer exlude the

18702

date tests (apparently this was overlooked)

18703

18704

* tests/mozilla/expected.html:

18705

18706

2005-07-31 Darin Adler <darin@apple.com>

18707

18708

Reviewed by Maciej.

18709

18710

- remove uses of Mac-OS-X-specific MAX macro

18711

- remove one of the many excess "APPLE_CHANGES" ifdefs

18712

18713

* kjs/collector.cpp: (KJS::Collector::allocate): Use std::max instead of MAX.

18714

* kjs/property_map.cpp: (KJS::PropertyMap::rehash): Ditto.

18715

* kjs/ustring.cpp:

18716

(KJS::UChar::toLower): Take out non-ICU code path.

18717

(KJS::UChar::toUpper): Ditto.

18718

(KJS::UString::spliceSubstringsWithSeparators): Use std::max instead of MAX.

18719

18720

2005-07-27 Geoffrey Garen <ggaren@apple.com>

18721

18722

- fixed http://bugs.webkit.org/show_bug.cgi?id=4147

18723

Array.toString() and toLocaleString() improvements from KDE KJS

18724

(rolled in KDE changes)

18725

18726

Test cases added:

18727

18728

* layout-tests/fast/js/toString-overrides-expected.txt: Added.

18729

* layout-tests/fast/js/toString-overrides.html: Added.

18730

18731

* kjs/array_object.cpp:

18732

(ArrayProtoFuncImp::call):

18733

18734

2005-07-27 Maciej Stachowiak <mjs@apple.com>

18735

18736

Changes by Michael Kahl, reviewed by me.

18737

18738

- fixed <rdar://problem/4194278> Need better debugging support in JavaScriptCore

18739

18740

* JavaScriptCore.xcodeproj/project.pbxproj:

18741

* kjs/debugger.cpp:

18742

(KJS::AttachedInterpreter::AttachedInterpreter):

18743

(KJS::AttachedInterpreter::~AttachedInterpreter):

18744

(Debugger::~Debugger):

18745

(Debugger::attach):

18746

(Debugger::detach):

18747

(Debugger::sourceParsed):

18748

* kjs/debugger.h:

18749

* kjs/function.cpp:

18750

(KJS::FunctionImp::call):

18751

(KJS::GlobalFuncImp::call):

18752

* kjs/function_object.cpp:

18753

(FunctionObjectImp::construct):

18754

* kjs/grammar.y:

18755

* kjs/internal.cpp:

18756

(Parser::parse):

18757

(InterpreterImp::evaluate):

18758

* kjs/internal.h:

18759

(KJS::InterpreterImp::setDebugger):

18760

* kjs/interpreter.cpp:

18761

* kjs/interpreter.h:

18762

(KJS::Interpreter::imp):

18763

* kjs/nodes.cpp:

18764

18765

2005-07-27 Geoffrey Garen <ggaren@apple.com>

18766

18767

- fixed http://bugs.webkit.org/show_bug.cgi?id=3381

18768

Date.prototype.setDate() incorrect for values >=128

18769

18770

- Test cases added:

18771

18772

* layout-tests/fast/js/date-big-setdate-expected.txt: Added.

18773

* layout-tests/fast/js/date-big-setdate.html: Added.

18774

18775

Reviewed by darin.

18776

18777

* kjs/date_object.cpp:

18778

(DateProtoFuncImp::call):

18779

18780

2005-07-27 Geoffrey Garen <ggaren@apple.com>

18781

18782

-rolled in patch by Carsten Guenther <cguenther@gmail.com>

18783

for http://bugs.webkit.org/show_bug.cgi?id=3759

18784

Date object enhancements

18785

18786

Test cases added:

18787

18788

* layout-tests/fast/js/date-preserve-milliseconds-expected.txt: Added.

18789

* layout-tests/fast/js/date-preserve-milliseconds.html: Added.

18790

18791

Reviewed by darin.

18792

18793

* kjs/date_object.cpp:

18794

(timeFromArgs):

18795

(DateProtoFuncImp::call):

18796

(DateObjectImp::construct):

18797

(DateObjectFuncImp::call):

18798

(KJS::makeTime):

18799

* kjs/date_object.h:

18800

* tests/mozilla/expected.html:

18801

18802

2005-07-26 Justin Garcia <justin.garcia@apple.com>

18803

18804

Added a forward declaration to fix gcc4 build error

18805

18806

* kjs/function.h:

18807

18808

2005-07-25 Geoffrey Garen <ggaren@apple.com>

18809

- fixed mistake in my last checkin -- the expected results included

18810

results from a patch that hasn't landed yet.

18811

18812

* tests/mozilla/expected.html:

18813

18814

2005-07-25 Maciej Stachowiak <mjs@apple.com>

18815

18816

- fix mistake in last change that leads to assertion failure in the Development build

18817

18818

* kjs/lookup.h:

18819

(KJS::lookupGetOwnValue):

18820

18821

2005-07-24 Maciej Stachowiak <mjs@apple.com>

18822

18823

Reviewed by Darin.

18824

18825

- http://bugs.webkit.org/show_bug.cgi?id=4124

18826

(change JavaScript property access to avoid double lookup)

18827

18828

- 10% speedup on JavaScript iBench

18829

- 5% speedup on 24fun BenchJS benchmark

18830

18831

Changed all get methods to getOwnProperty - they are no longer

18832

responsible for prototype lookup, and determine if the property

18833

was found as a side efect.

18834

18835

get() is now a nonvirtual ObjectImp method which calls the virtual

18836

getOwnProperty and walks the prototype chain. A few selected

18837

methods were inlined.

18838

18839

Changed ResolveNode::evaluate plus some other places to use

18840

getProperty which does get() and hasProperty() in one lookup.

18841

18842

Also miscellaneous code cleanup.

18843

18844

* bindings/objc/objc_runtime.h:

18845

* bindings/objc/objc_runtime.mm:

18846

(ObjcFallbackObjectImp::ObjcFallbackObjectImp):

18847

(ObjcFallbackObjectImp::getOwnProperty):

18848

* bindings/runtime_array.cpp:

18849

(RuntimeArrayImp::RuntimeArrayImp):

18850

(RuntimeArrayImp::getOwnProperty):

18851

* bindings/runtime_array.h:

18852

* bindings/runtime_method.cpp:

18853

(RuntimeMethodImp::getOwnProperty):

18854

* bindings/runtime_method.h:

18855

* bindings/runtime_object.cpp:

18856

(RuntimeObjectImp::getOwnProperty):

18857

* bindings/runtime_object.h:

18858

* kjs/array_instance.h:

18859

* kjs/array_object.cpp:

18860

(ArrayInstanceImp::getOwnProperty):

18861

(ArrayPrototypeImp::getOwnProperty):

18862

(ArrayProtoFuncImp::call):

18863

* kjs/array_object.h:

18864

* kjs/date_object.cpp:

18865

(DatePrototypeImp::getOwnProperty):

18866

* kjs/date_object.h:

18867

* kjs/function.cpp:

18868

(KJS::FunctionImp::getOwnProperty):

18869

(KJS::ArgumentsImp::getOwnProperty):

18870

(KJS::ActivationImp::getOwnProperty):

18871

* kjs/function.h:

18872

* kjs/lookup.h:

18873

(KJS::lookupGetOwnProperty):

18874

(KJS::lookupGetOwnFunction):

18875

(KJS::lookupGetOwnValue):

18876

* kjs/math_object.cpp:

18877

(MathObjectImp::getOwnProperty):

18878

(MathObjectImp::getValueProperty):

18879

* kjs/math_object.h:

18880

* kjs/nodes.cpp:

18881

(ResolveNode::evaluate):

18882

* kjs/number_object.cpp:

18883

(NumberObjectImp::getOwnProperty):

18884

* kjs/number_object.h:

18885

* kjs/object.cpp:

18886

(KJS::ObjectImp::get):

18887

(KJS::ObjectImp::getOwnProperty):

18888

(KJS::ObjectImp::getProperty):

18889

* kjs/object.h:

18890

(KJS::ObjectImp::getProperty):

18891

(KJS::ObjectImp::getOwnProperty):

18892

* kjs/object_object.cpp:

18893

(ObjectProtoFuncImp::call):

18894

* kjs/regexp_object.cpp:

18895

(RegExpObjectImp::getOwnProperty):

18896

* kjs/regexp_object.h:

18897

* kjs/string_object.cpp:

18898

(StringInstanceImp::getOwnProperty):

18899

(StringPrototypeImp::getOwnProperty):

18900

* kjs/string_object.h:

18901

18902

2005-07-25 Geoffrey Garen <ggaren@apple.com>

18903

18904

- fixed http://bugs.webkit.org/show_bug.cgi?id=3971

18905

JS test suite depends on JS 1.2 behavior

18906

18907

Reviewed by darin.

18908

18909

* tests/mozilla/js1_2/Array/tostring_1.js: now tests only for JS 1.5 behavior

18910

* tests/mozilla/js1_2/Array/tostring_2.js: ditto

18911

* tests/mozilla/expected.html:

18912

18913

2005-07-24 Justin Garcia <justin.garcia@apple.com>

18914

18915

Reviewed by kevin.

18916

18917

Fixes make clean problem introduced in xcode2.1 transition

18918

18919

* Makefile.am:

18920

18921

2005-07-22 Geoffrey Garen <ggaren@apple.com>

18922

18923

Reviewed by darin.

18924

18925

* kjs/date_object.cpp: DatePrototypeImp now identifies itself as a

18926

child class of DateInstanceImp -- this enables calls to Date.ValueOf().

18927

18928

fixes: ecma/Date/15.9.5.js (once we enable the date tests).

18929

18930

2005-07-22 Geoffrey Garen <ggaren@apple.com>

18931

18932

18933

Reviewed by darin.

18934

18935

* tests/mozilla/jsDriver.pl: now takes the path to testkjs as a command-line argument

18936

* tests/mozilla/run-mozilla-tests: Removed.

18937

18938

2005-07-21 Geoffrey Garen <ggaren@apple.com>

18939

18940

* JavaScriptCore.xcodeproj/.cvsignore: Added.

18941

18942

2005-07-21 Geoffrey Garen <ggaren@apple.com>

18943

18944

* JavaScriptCore.pbproj/project.pbxproj: Removed.

18945

* JavaScriptCore.xcodeproj/ggaren.pbxuser: Added.

18946

* JavaScriptCore.xcodeproj/ggaren.perspective: Added.

18947

* JavaScriptCore.xcodeproj/project.pbxproj: Added.

18948

* Makefile.am:

18949

18950

2005-07-20 Maciej Stachowiak <mjs@apple.com>

18951

18952

Patch from Trey Matteson <trey@usa.net>, reviewed by me.

18953

18954

- fixed http://bugs.webkit.org/show_bug.cgi?id=3956

18955

some of WebKit builds with symbols, some doesn't

18956

18957

* JavaScriptCore.pbproj/project.pbxproj: Generate symbols even for

18958

Deployment.

18959

18960

2005-07-19 Geoffrey Garen <ggaren@apple.com>

18961

18962

-fixed http://bugs.webkit.org/show_bug.cgi?id=3991

18963

JSC doesn't implement Array.prototype.toLocaleString()

18964

18965

-test failure: ecma_3/Array/15.4.4.3-1.js

18966

18967

Reviewed by mjs.

18968

18969

* kjs/array_object.cpp:

18970

(ArrayProtoFuncImp::call): now searches for toString and

18971

toLocaleString overrides in the array's elements

18972

18973

* tests/mozilla/expected.html: failures are under 100! woohoo!

18974

18975

2005-07-19 Darin Adler <darin@apple.com>

18976

18977

- fixed the build

18978

18979

* kjs/lookup.h: (KJS::lookupPut): Remove bogus const; was preventing WebCore from

18980

compiling (not sure why this didn't affect my other build machine).

18981

18982

- one other tiny tweak (so sue me)

18983

18984

* bindings/runtime_root.cpp: Remove unneeded declaration.

18985

18986

2005-07-19 Darin Adler <darin@apple.com>

18987

18988

Reviewed by Geoff Garen.

18989

18990

- eliminated try wrappers for get/put/call since we don't use C++ exceptions any more

18991

18992

* kjs/lookup.h: Changed tryCall in IMPLEMENT_PROTOFUNC here to call. It doesn't make

18993

sense for this macro to use the name tryCall anyway, since that's specific to how

18994

WebCore used this, so this is good anyway. On the other hand, it might be a problem

18995

for KDOM or KSVG, in which case we'll need another macro for them, since JavaScriptCore

18996

should presumably not have the C++ exception support.

18997

18998

2005-07-18 Geoffrey Garen <ggaren@apple.com>

18999

19000

-fixed http://bugs.webkit.org/show_bug.cgi?id=4008

19001

Error objects report incorrect length

19002

19003

Reviewed by darin.

19004

19005

* kjs/error_object.cpp: Error objects now include a length property

19006

(ErrorObjectImp::ErrorObjectImp):

19007

19008

* tests/mozilla/expected.html: updated expected results to reflect fix

19009

* tests/mozilla/js1_5/Exceptions/regress-123002.js: test now expects

19010

ecma compliant results

19011

19012

2005-07-15 Geoffrey Garen <ggaren@apple.com>

19013

19014

-rolled in KDE fixes for http://bugs.webkit.org/show_bug.cgi?id=3601

19015

Error instance type info

19016

19017

Reviewed by mjs.

19018

19019

* kjs/error_object.cpp:

19020

- Created ErrorInstanceImp class for Error() objects.

19021

- Changed parent object for Native Errors to "Function" (matches

19022

ECMA spec).

19023

(ErrorInstanceImp::ErrorInstanceImp):

19024

(ErrorProtoFuncImp::call):

19025

(ErrorObjectImp::construct):

19026

(NativeErrorImp::construct):

19027

19028

* kjs/error_object.h:

19029

(KJS::ErrorInstanceImp::classInfo):

19030

* kjs/object.h: made comment more informative about ClassInfo

19031

19032

* tests/mozilla/expected.html:

19033

19034

2005-07-14 Geoffrey Garen <ggaren@apple.com>

19035

19036

- fixed: JS test suite expects an out of memory error

19037

that our memory efficiency avoids

19038

19039

Reviewed by mjs.

19040

19041

* tests/mozilla/js1_5/Array/regress-157652.js:

19042

test now expects normal execution

19043

19044

* tests/mozilla/expected.html:

19045

19046

2005-07-14 Geoffrey Garen <ggaren@apple.com>

19047

- fixed http://bugs.webkit.org/show_bug.cgi?id=4006

19048

testkjs doesn't implement gc()

19049

19050

- test failure:

19051

ecma_3/Function/regress-104584.js

19052

19053

Reviewed by mjs.

19054

19055

* kjs/interpreter.cpp:

19056

(Interpreter::finalCheck): removed misleading while && comment

19057

19058

* kjs/testkjs.cpp: added "gc" function to global object

19059

(TestFunctionImp::):

19060

(TestFunctionImp::call):

19061

(main):

19062

19063

* tests/mozilla/expected.html:

19064

19065

2005-07-14 Geoffrey Garen <ggaren@apple.com>

19066

19067

-rolled in patches for http://bugs.webkit.org/show_bug.cgi?id=3945

19068

[PATCH] Safe merges of comments and other trivialities from KDE's kjs

19069

19070

-patch by Martijn Klingens <klingens@kde.org>

19071

19072

* kjs/array_instance.h:

19073

* kjs/array_object.cpp:

19074

* kjs/array_object.h:

19075

* kjs/bool_object.cpp:

19076

* kjs/bool_object.h:

19077

* kjs/collector.cpp:

19078

* kjs/collector.h:

19079

* kjs/completion.h:

19080

* kjs/context.h:

19081

* kjs/date_object.cpp:

19082

* kjs/date_object.h:

19083

* kjs/debugger.cpp:

19084

* kjs/debugger.h:

19085

* kjs/dtoa.h:

19086

* kjs/error_object.cpp:

19087

* kjs/error_object.h:

19088

* kjs/function.cpp:

19089

* kjs/function.h:

19090

* kjs/function_object.cpp:

19091

* kjs/function_object.h:

19092

* kjs/grammar.y:

19093

* kjs/identifier.cpp:

19094

* kjs/identifier.h:

19095

* kjs/internal.cpp:

19096

* kjs/internal.h:

19097

* kjs/interpreter.cpp:

19098

* kjs/interpreter.h:

19099

* kjs/interpreter_map.cpp:

19100

* kjs/interpreter_map.h:

19101

* kjs/lexer.cpp:

19102

* kjs/lexer.h:

19103

* kjs/list.cpp:

19104

* kjs/list.h:

19105

* kjs/lookup.cpp:

19106

* kjs/lookup.h:

19107

* kjs/math_object.cpp:

19108

* kjs/math_object.h:

19109

* kjs/nodes.cpp:

19110

* kjs/nodes.h:

19111

* kjs/nodes2string.cpp:

19112

* kjs/number_object.cpp:

19113

* kjs/number_object.h:

19114

* kjs/object.cpp:

19115

* kjs/object.h:

19116

* kjs/object_object.cpp:

19117

* kjs/object_object.h:

19118

* kjs/operations.cpp:

19119

* kjs/operations.h:

19120

* kjs/property_map.cpp:

19121

* kjs/property_map.h:

19122

* kjs/reference.cpp:

19123

* kjs/reference.h:

19124

* kjs/reference_list.cpp:

19125

* kjs/reference_list.h:

19126

* kjs/regexp.cpp:

19127

* kjs/regexp.h:

19128

* kjs/regexp_object.cpp:

19129

* kjs/regexp_object.h:

19130

* kjs/scope_chain.cpp:

19131

* kjs/scope_chain.h:

19132

* kjs/simple_number.h:

19133

* kjs/string_object.cpp:

19134

* kjs/string_object.h:

19135

* kjs/testkjs.cpp:

19136

* kjs/types.h:

19137

* kjs/ustring.cpp:

19138

* kjs/ustring.h:

19139

* kjs/value.cpp:

19140

* kjs/value.h:

19141

19142

2005-07-14 Geoffrey Garen <ggaren@apple.com>

19143

19144

-fixed http://bugs.webkit.org/show_bug.cgi?id=3970

19145

throw statements fail inside eval statements

19146

19147

Reviewed by mjs.

19148

19149

* kjs/function.cpp:

19150

(KJS::GlobalFuncImp::call):

19151

Big change since I fixed the tabbing. The important part is:

19152

if (c.complType() == Throw)

19153

exec->setException(c.value());

19154

19155

* kjs/nodes.cpp:

19156

(ThrowNode::execute): removed duplicate KJS_CHECKEXCEPTION

19157

(TryNode::execute):

19158

try now clears the exception state before the finally block executes,

19159

and checks the state after the block executes, so that exceptions in

19160

finally code get caught.

19161

19162

* tests/mozilla/expected.html:

19163

19164

2005-07-14 Geoffrey Garen <ggaren@apple.com>

19165

19166

-landed fix for http://bugs.webkit.org/show_bug.cgi?id=3412

19167

Object.prototype is missing toLocaleString

19168

19169

- patch by Mark Rowe (bdash) <opendarwin.org@bdash.net.nz>

19170

19171

-layout test info in webcore changelog

19172

19173

Reviewed by mjs.

19174

19175

* kjs/object_object.cpp:

19176

(ObjectPrototypeImp::ObjectPrototypeImp):

19177

(ObjectProtoFuncImp::call):

19178

* kjs/object_object.h:

19179

(KJS::ObjectProtoFuncImp::):

19180

19181

2005-07-12 Geoffrey Garen <ggaren@apple.com>

19182

19183

Reviewed by mjs.

19184

19185

* kjs/function.cpp:

19186

(KJS::IndexToNameMap::operator[]): fixed infinite recursion

19187

bug in last checkin

19188

19189

2005-07-12 Geoffrey Garen <ggaren@apple.com>

19190

19191

-fixed http://bugs.webkit.org/show_bug.cgi?id=3881

19192

arguments object should share values with function parameters

19193

19194

Reviewed by mjs.

19195

19196

ArgumentsImp now uses a simple hash lookup to share values

19197

with the activation object.

19198

19199

* kjs/function.cpp:

19200

(KJS::FunctionImp::getParameterName):

19201

(KJS::IndexToNameMap::IndexToNameMap):

19202

(KJS::IndexToNameMap::~IndexToNameMap):

19203

(KJS::IndexToNameMap::isMapped):

19204

(KJS::IndexToNameMap::unMap):

19205

(KJS::IndexToNameMap::operator[]):

19206

(KJS::ArgumentsImp::ArgumentsImp):

19207

(KJS::ArgumentsImp::mark):

19208

(KJS::ArgumentsImp::get):

19209

(KJS::ArgumentsImp::put):

19210

(KJS::ArgumentsImp::deleteProperty):

19211

(KJS::ArgumentsImp::hasOwnProperty):

19212

(KJS::ActivationImp::createArgumentsObject):

19213

* kjs/function.h:

19214

* tests/mozilla/expected.html: updated results

19215

19216

2005-07-09 Maciej Stachowiak <mjs@apple.com>

19217

19218

- backing out my earlier collector change, it causes a performance regression in TOT

19219

19220

* kjs/collector.cpp:

19221

(KJS::Collector::allocate):

19222

19223

2005-07-08 Eric Seidel <eseidel@apple.com>

19224

19225

Reviewed by mjs/hyatt (only in concept).

19226

19227

* JavaScriptCore.pbproj/project.pbxproj: Added JavaScriptCore+SVG

19228

Turns on RTTI support for JavaScriptCore.framework when

19229

building the JavaScriptCore+SVG target. This is needed as

19230

kdom (part of WebCore+SVG) requires RTTI for the time being.

19231

19232

2005-07-08 Maciej Stachowiak <mjs@apple.com>

19233

19234

Reviewed by hyatt.

19235

19236

- When there are many live objects, GC less often, to try to make

19237

GC cost proportional to garbage, not proportional to total memory used.

19238

19239

* kjs/collector.cpp:

19240

(KJS::Collector::allocate):

19241

19242

2005-07-08 Vicki Murley <vicki@apple.com>

19243

19244

Fix from Carsten Guenther, reviewed by Maciej

19245

19246

- fixed http://bugs.webkit.org/show_bug.cgi?id=3644 (Error string representation)

19247

19248

Switch from "-" to ":" in error strings.

19249

19250

* kjs/error_object.cpp:

19251

(ErrorProtoFuncImp::call):

19252

* tests/mozilla/expected.html:

19253

19254

2005-07-08 Geoffrey Garen <ggaren@apple.com>

19255

19256

-rolled in patch for http://bugs.webkit.org/show_bug.cgi?id=3878

19257

arguments object should be an object not an array

19258

19259

Reviewed by mjs.

19260

19261

* kjs/function.cpp:

19262

(KJS::ArgumentsImp::ArgumentsImp): now manually handles initialization

19263

we used to get for free by inheriting from ArrayInstanceImp

19264

* kjs/function.h: ArgumentsImp now inherits from ObjectImp

19265

* tests/mozilla/expected.html: updated expected test results

19266

19267

2005-07-07 Eric Seidel <eseidel@apple.com>

19268

19269

Reviewed by mjs.

19270

19271

* kjs/grammar.y: removed #define YYMAXDEPTH 0 for bison 2.0

19272

http://bugs.webkit.org/show_bug.cgi?id=3882

19273

19274

2005-07-03 Maciej Stachowiak <mjs@apple.com>

19275

19276

Original patch from Mark Rowe <opendarwin.org@bdash.net.nz>, reviewed by me.

19277

Fixes to patch by me, reviewed by John Sullivan.

19278

19279

- fixed http://bugs.webkit.org/show_bug.cgi?id=3293

19280

19281

Test cases added:

19282

* tests/mozilla/expected.html: Two tests newly pass.

19283

19284

* bindings/objc/objc_runtime.h:

19285

* bindings/objc/objc_runtime.mm:

19286

(ObjcFallbackObjectImp::hasOwnProperty):

19287

* bindings/runtime_array.cpp:

19288

(RuntimeArrayImp::hasOwnProperty):

19289

* bindings/runtime_array.h:

19290

* bindings/runtime_object.cpp:

19291

(RuntimeObjectImp::hasOwnProperty):

19292

* bindings/runtime_object.h:

19293

* kjs/array_instance.h:

19294

* kjs/array_object.cpp:

19295

(ArrayInstanceImp::hasOwnProperty):

19296

* kjs/function.cpp:

19297

(KJS::FunctionImp::hasOwnProperty):

19298

(KJS::ActivationImp::hasOwnProperty):

19299

* kjs/function.h:

19300

* kjs/lookup.h:

19301

* kjs/object.cpp:

19302

(KJS::ObjectImp::hasProperty):

19303

(KJS::ObjectImp::hasOwnProperty):

19304

* kjs/object.h:

19305

(KJS::Object::hasOwnProperty):

19306

* kjs/object_object.cpp:

19307

(ObjectPrototypeImp::ObjectPrototypeImp):

19308

(ObjectProtoFuncImp::call):

19309

* kjs/object_object.h:

19310

(KJS::ObjectProtoFuncImp::):

19311

* kjs/string_object.cpp:

19312

(StringInstanceImp::hasOwnProperty):

19313

* kjs/string_object.h:

19314

19315

2005-07-01 Geoffrey Garen <ggaren@apple.com>

19316

19317

-landed patch by Eric Seidel <macdome@opendarwin.org>

19318

19319

-for http://bugs.webkit.org/show_bug.cgi?id=3657

19320

GroundWork: Moving some functions from khtml->jsc following kjs TOT

19321

19322

- no layout test necessary yet - only groundwork

19323

19324

Reviewed by darin.

19325

19326

* kjs/lookup.h:

19327

(KJS::cacheGlobalObject):

19328

19329

2005-07-01 Geoffrey Garen <ggaren@apple.com>

19330

19331

-landed patch by Carsten Guenther <cguenther@gmail.com>

19332

19333

-fixes http://bugs.webkit.org/show_bug.cgi?id=3477

19334

some US-centric date formats not parsed by JavaScript (clock at news8austin.com)

19335

19336

-relevant tests:

19337

mozilla/ecma_3/Date/15.9.5.5.js

19338

layout-tests/fast/js/date-parse-test.html

19339

19340

Reviewed by darin.

19341

19342

* kjs/date_object.cpp:

19343

(formatLocaleDate):

19344

(day):

19345

(dayFromYear):

19346

(daysInYear):

19347

(timeFromYear):

19348

(yearFromTime):

19349

(weekDay):

19350

(timeZoneOffset):

19351

(DateProtoFuncImp::call):

19352

(DateObjectImp::construct):

19353

(KJS::parseDate):

19354

(ymdhms_to_seconds):

19355

(KJS::makeTime):

19356

(findMonth):

19357

(KJS::KRFCDate_parseDate):

19358

* kjs/date_object.h:

19359

* tests/mozilla/expected.html: updated expected results to reflect fix

19360

19361

2005-07-01 Geoffrey Garen <ggaren@apple.com>

19362

19363

-fixed <rdar://problem/4168186> JavaScript fails to throw exceptions

19364

for invalid return statements

19365

19366

relevant tests:

19367

ecma/Statements/12.9-1-n.js

19368

ecma_2/Exceptions/lexical-052.js

19369

ecma_2/Exceptions/statement-009.js

19370

19371

Reviewed by sullivan.

19372

19373

* kjs/nodes.cpp:

19374

(ReturnNode::execute): now throws exception if return is not inside

19375

a function.

19376

19377

* tests/mozilla/expected.html: updated to reflect fix

19378

19379

2005-07-01 Geoffrey Garen <ggaren@apple.com>

19380

19381

Reviewed by sullivan.

19382

19383

* tests/mozilla/expected.html: Updated test results for last fix.

19384

19385

2005-07-01 Geoffrey Garen <ggaren@apple.com>

19386

19387

-fixed <rdar://problem/4168161> JavaScript fails to throw an exception

19388

for invalid function calls

19389

19390

Reviewed by sullivan.

19391

19392

Relevant mozilla test: ecma_3/Exceptions/regress-95101.js

19393

19394

* kjs/nodes.cpp:

19395

(FunctionCallNode::evaluate): evaluate now checks for an exception

19396

after resolving a function name (in case the function is undefined)

19397

19398

2005-07-01 Eric Seidel <eseidel@apple.com>

19399

19400

Reviewed by darin.

19401

19402

* kjs/interpreter.h:

19403

(KJS::Context::curStmtFirstLine): stub for compatibility with KDE

19404

* kjs/value.h:

19405

(KJS::Value::isValid): compatibility with KDE

19406

http://bugs.webkit.org/show_bug.cgi?id=3687

19407

19408

2005-07-01 Eric Seidel <eseidel@apple.com>

19409

19410

Reviewed by darin.

19411

19412

* kjs/create_hash_table: rolled in changes from KDE, including

19413

-n <namespace> support from KDOM and support for newer comments

19414

http://bugs.webkit.org/show_bug.cgi?id=3771

19415

19416

2005-06-30 Geoffrey Garen <ggaren@apple.com>

19417

19418

-rolled in KDE fix to <rdar://problem/4167660> JavaScript fails to

19419

throw exceptions for invalid break/continue statements

19420

19421

No layout tests because it's already covered by the Mozilla suite

19422

19423

Reviewed by mjs.

19424

19425

* kjs/internal.h: LabelStack now tracks where you are relative to

19426

switch and iteration (loop) statements

19427

19428

(KJS::LabelStack::LabelStack):

19429

(KJS::LabelStack::pushIteration):

19430

(KJS::LabelStack::popIteration):

19431

(KJS::LabelStack::inIteration):

19432

(KJS::LabelStack::pushSwitch):

19433

(KJS::LabelStack::popSwitch):

19434

(KJS::LabelStack::inSwitch):

19435

19436

* kjs/nodes.cpp:

19437

These files were updated to use the new LabelStack:

19438

(DoWhileNode::execute):

19439

(WhileNode::execute):

19440

(ForNode::execute):

19441

(ForInNode::execute):

19442

(SwitchNode::execute):

19443

19444

These files were updated to throw exceptions for invalid

19445

break/continue statements:

19446

(BreakNode::execute):

19447

(ContinueNode::execute):

19448

19449

* tests/mozilla/expected.html: Updated expected results to reflect fix

19450

19451

2005-06-30 Kevin Decker <kdecker@apple.com>

19452

19453

Reviewed by rjw.

19454

19455

fixed: <rdar://problem/4166838> failed assertion in`Interpreter::lockCount() > 0

19456

19457

no layout test added; this is in the bindings code.

19458

19459

* bindings/objc/WebScriptObject.mm:

19460

(+[WebScriptObject _convertValueToObjcValue:KJS::originExecutionContext:Bindings::executionContext:Bindings::]): make sure to lock and unlock the interpreter around allocations.

19461

19462

2005-06-29 Geoffrey Garen <ggaren@apple.com>

19463

19464

Patch by Francisco Tolmasky <tolmasky@gmail.com>

19465

19466

- fixes http://bugs.webkit.org/show_bug.cgi?id=3667

19467

Core JavaScript 1.5 Reference:Objects:Array:forEach

19468

19469

See WebCore Changelog for layout tests added.

19470

19471

Reviewed by darin.

19472

19473

* kjs/array_object.cpp:

19474

(ArrayProtoFuncImp::call):

19475

* kjs/array_object.h:

19476

(KJS::ArrayProtoFuncImp::):

19477

19478

2005-06-29 Geoffrey Garen <ggaren@apple.com>

19479

19480

Patch contributed by Oliver Hunt <ojh16@student.canterbury.ac.nz>

19481

19482

-fixed http://bugs.webkit.org/show_bug.cgi?id=3743

19483

Incorrect error message given for certain calls

19484

19485

See WebCore Changelog for layout test added.

19486

19487

Reviewed by mjs.

19488

19489

* kjs/object.cpp:

19490

(KJS::ObjectImp::defaultValue):

19491

19492

2005-06-29 Geoffrey Garen <ggaren@apple.com>

19493

19494

Rolling out date patch from 6-28-05 because it breaks

19495

fast/js/date-parse-test

19496

19497

* kjs/date_object.cpp:

19498

(formatLocaleDate):

19499

(DateProtoFuncImp::call):

19500

(DateObjectImp::construct):

19501

(KJS::parseDate):

19502

(ymdhms_to_seconds):

19503

(isSpaceOrTab):

19504

(KJS::KRFCDate_parseDate):

19505

* kjs/date_object.h:

19506

* tests/mozilla/expected.html:

19507

19508

2005-06-29 Geoffrey Garen <ggaren@apple.com>

19509

19510

Reviewed by Darin.

19511

19512

-fixes http://bugs.webkit.org/show_bug.cgi?id=3750

19513

build fails with KJS_VERBOSE set

19514

19515

* kjs/nodes.cpp: changed debug print statement to use UString

19516

(VarDeclNode::evaluate):

19517

* kjs/reference.cpp: ditto

19518

(KJS::Reference::putValue):

19519

19520

2005-06-28 Geoffrey Garen <ggaren@apple.com>

19521

19522

Patch contributed by Carsten Guenther <cguenther@gmail.com>.

19523

19524

-fixes http://bugs.webkit.org/show_bug.cgi?id=3477

19525

some US-centric date formats not parsed by JavaScript (clock at news8austin.com)

19526

19527

Reviewed by darin.

19528

19529

* kjs/date_object.cpp:

19530

(formatLocaleDate):

19531

(day):

19532

(dayFromYear):

19533

(daysInYear):

19534

(timeFromYear):

19535

(yearFromTime):

19536

(weekDay):

19537

(timeZoneOffset):

19538

(DateProtoFuncImp::call):

19539

(DateObjectImp::construct):

19540

(KJS::parseDate):

19541

(ymdhms_to_seconds):

19542

(KJS::makeTime):

19543

(findMonth):

19544

(KJS::KRFCDate_parseDate):

19545

* kjs/date_object.h:

19546

* tests/mozilla/expected.html: updated expected test results to reflect fix

19547

19548

2005-06-26 Maciej Stachowiak <mjs@apple.com>

19549

19550

Reviewed by Darin.

19551

19552

- replace hash functions with better ones

19553

19554

* JavaScriptCore.pbproj/project.pbxproj: Add new file to build.

19555

* kjs/interpreter_map.cpp:

19556

(KJS::InterpreterMap::computeHash): Use shared pointer hash.

19557

* kjs/pointer_hash.h: Added.

19558

(KJS::pointerHash): Pointer hash based on 32-bit mix and 64-bit mix hashes.

19559

* kjs/protected_values.cpp:

19560

(KJS::ProtectedValues::computeHash): Use shared pointer hash.

19561

* kjs/ustring.cpp:

19562

(KJS::UString::Rep::computeHash): Use SuperFastHash algorithm.

19563

19564

2005-06-22 Darin Adler <darin@apple.com>

19565

19566

Change by Anders Carlsson.

19567

Reviewed by me.

19568

19569

- fixed <http://bugs.webkit.org/show_bug.cgi?id=3294>

19570

String.prototype.replace() fails with function as second param

19571

19572

* kjs/string_object.cpp: (replace): Added code to handle functions.

19573

19574

* tests/mozilla/expected.html: Updated since ecma_3/RegExp/regress-209067.js is fixed now.

19575

19576

* tests/mozilla/run-mozilla-tests: Fix a minor coding style issue that leads to a warning each

19577

time we run the tests.

19578

19579

2005-06-21 Adele Peterson <adele@apple.com>

19580

19581

rolling out fix for http://bugs.webkit.org/show_bug.cgi?id=3293, since it caused layout test failures.

19582

fast/forms/element-by-name

19583

fast/loader/loadInProgress

19584

19585

* ChangeLog:

19586

* bindings/objc/objc_runtime.h:

19587

* bindings/objc/objc_runtime.mm:

19588

(ObjcFallbackObjectImp::hasProperty):

19589

* bindings/runtime_array.cpp:

19590

(RuntimeArrayImp::hasProperty):

19591

* bindings/runtime_array.h:

19592

* bindings/runtime_object.cpp:

19593

(RuntimeObjectImp::hasProperty):

19594

* bindings/runtime_object.h:

19595

* kjs/array_instance.h:

19596

* kjs/array_object.cpp:

19597

(ArrayInstanceImp::hasProperty):

19598

* kjs/function.cpp:

19599

(KJS::FunctionImp::hasProperty):

19600

(KJS::ActivationImp::hasProperty):

19601

* kjs/function.h:

19602

* kjs/object.cpp:

19603

(KJS::ObjectImp::hasProperty):

19604

* kjs/object.h:

19605

* kjs/object_object.cpp:

19606

(ObjectPrototypeImp::ObjectPrototypeImp):

19607

(ObjectProtoFuncImp::call):

19608

* kjs/object_object.h:

19609

(KJS::ObjectProtoFuncImp::):

19610

* kjs/string_object.cpp:

19611

(StringInstanceImp::hasProperty):

19612

* kjs/string_object.h:

19613

* tests/mozilla/expected.html:

19614

19615

2005-06-21 Darin Adler <darin@apple.com>

19616

19617

* JavaScriptCore.pbproj/project.pbxproj: Switched to a build rule rather than a build phase for

19618

.y files -- this gets rid of the problem where modifying the .y file would not cause sufficient

19619

compilation.

19620

19621

* kjs/grammar_wrapper.cpp: Removed.

19622

19623

2005-06-21 Adele Peterson <adele@apple.com>

19624

19625

Patch from Anders Carlsson <andersca@mac.com>, reviewed by Darin.

19626

19627

Fixed: <http://bugs.webkit.org/show_bug.cgi?id=3450>

19628

<rdar://problem/3881901> String.replace() method not working when regex pattern contains {n, m}

19629

19630

* pcre/pcre.c: (pcre_compile): Remember the last char length so it can be subtracted correctly if needed.

19631

19632

2005-06-21 Geoffrey Garen <ggaren@apple.com>

19633

19634

- fixed <rdar://problem/4155532> 'delete' succeeds on functions

19635

- fixed <rdar://problem/4155049> javascript function named as "opener" doesn't get called because of window.opener property

19636

19637

Reviewed by cblu.

19638

19639

* kjs/nodes.cpp:

19640

(FuncDeclNode::processFuncDecl): Functions now have DontDelete and Internal attributes set when appropriate.

19641

19642

Test cases:

19643

* tests/mozilla/expected.html: Updated for one new success.

19644

- see also test case added in WebCore.

19645

19646

2005-06-20 Maciej Stachowiak <mjs@apple.com>

19647

19648

Reviewed by Darin(first pass) and Hyatt.

19649

19650

- fixed http://bugs.webkit.org/show_bug.cgi?id=3576

19651

(roll in support for "const" keyword from KDE tree)

19652

- make processVarDecls handle deletability of variables declared

19653

in an eval block the same as evaluate would

19654

- make eval() call processVarDecls - needed to match mozilla and

19655

to make the second change testable

19656

19657

I started with the KDE implementation of const but I ended up changing it a bit

19658

to avoid the use of a global variable. Now instead of the global variable it distinguishes

19659

const and var at the grammar level so the appropriate node can know the right kind of

19660

declaration.

19661

19662

Test cases:

19663

* tests/mozilla/expected.html: Updated for one new test that is

19664

failing - we used to bail on it entirely because it checks for

19665

const support before starting.

19666

- see also test cases added in WebCore

19667

19668

* kjs/grammar.y: Add rules for const declarations.

19669

* kjs/keywords.table: Add const keyword.

19670

* kjs/nodes.cpp:

19671

(VarDeclNode::VarDeclNode): Add parameter.

19672

(VarDeclNode::evaluate): Add const support.

19673

(VarDeclNode::processVarDecls): Add const support.

19674

(VarStatementNode::execute): Irrelevant change.

19675

(ForInNode::ForInNode): Tell our variable node that it's a variable.

19676

* kjs/nodes.h:

19677

(KJS::VarDeclNode::): Add declaration of type enum, extra constructor parameter.

19678

(KJS::VarStatementNode::VarStatementNode): Irrelevant change.

19679

* kjs/function.cpp:

19680

(KJS::GlobalFuncImp::call): Process var decls before evaluating.

19681

19682

2005-06-20 Maciej Stachowiak <mjs@apple.com>

19683

19684

Patch from Mark Rowe <opendarwin.org@bdash.net.nz>, reviewed by me.

19685

19686

- fixed http://bugs.webkit.org/show_bug.cgi?id=3293

19687

19688

Test cases added:

19689

* tests/mozilla/expected.html: Updated for two fixed tests.

19690

- also added a layout test

19691

19692

* bindings/objc/objc_runtime.h:

19693

* bindings/objc/objc_runtime.mm:

19694

(ObjcFallbackObjectImp::hasOwnProperty):

19695

* bindings/runtime_array.cpp:

19696

(RuntimeArrayImp::hasOwnProperty):

19697

* bindings/runtime_array.h:

19698

* bindings/runtime_object.cpp:

19699

(RuntimeObjectImp::hasOwnProperty):

19700

* bindings/runtime_object.h:

19701

* kjs/array_instance.h:

19702

* kjs/array_object.cpp:

19703

(ArrayInstanceImp::hasOwnProperty):

19704

* kjs/function.cpp:

19705

(KJS::FunctionImp::hasOwnProperty):

19706

(KJS::ActivationImp::hasOwnProperty):

19707

* kjs/function.h:

19708

* kjs/object.cpp:

19709

(KJS::ObjectImp::hasProperty):

19710

(KJS::ObjectImp::hasOwnProperty):

19711

* kjs/object.h:

19712

(KJS::Object::hasOwnProperty):

19713

* kjs/object_object.cpp:

19714

(ObjectPrototypeImp::ObjectPrototypeImp):

19715

(ObjectProtoFuncImp::call):

19716

* kjs/object_object.h:

19717

(KJS::ObjectProtoFuncImp::):

19718

* kjs/string_object.cpp:

19719

(StringInstanceImp::hasOwnProperty):

19720

* kjs/string_object.h:

19721

19722

2005-06-18 Darin Adler <darin@apple.com>

19723

19724

Reviewed by Eric Seidel.

19725

19726

* pcre/get.c: (pcre_get_substring): Fix some computations so this works for UTF-16.

19727

This is unused in the current JavaScriptCore, but still good to fix.

19728

19729

2005-06-18 Darin Adler <darin@apple.com>

19730

19731

Change by Finlay Dobbie.

19732

Reviewed by me.

19733

19734

- fixed <http://bugs.webkit.org/show_bug.cgi?id=3331>

19735

10.3.9 Build Failure: NSString may not respond to `+stringWithCString:encoding:'

19736

19737

* bindings/objc/WebScriptObject.mm: (-[WebScriptObject stringRepresentation]):

19738

Undo change we did a while back to work around the gcc 3.3 compiler error.

19739

It no longer seems to happen, and the workaround code was 10.4-specific.

19740

19741

2005-06-16 Geoffrey Garen <ggaren@apple.com>

19742

19743

Fixed: <rdar://problem/4151759> 'delete' fails on variables declared inside 'eval' statements.

19744

19745

Reviewed by cblu.

19746

19747

* kjs/context.h:

19748

(KJS::ContextImp::codeType): Added code type accessor for execution context objects.

19749

* kjs/internal.cpp:

19750

(ContextImp::ContextImp): Reflects change to ContextImp::codeType.

19751

* kjs/nodes.cpp:

19752

(VarDeclNode::evaluate): Added separate code path for variable declarations inside 'eval' statements.

19753

* tests/mozilla/expected.html: Updated expected test results to reflect fix.

19754

19755

2005-06-14 Geoffrey Garen <ggaren@apple.com>

19756

19757

Updated expected.html to reflect fix to <rdar://problem/4147745>.

19758

19759

Reviewed by cblu.

19760

19761

* tests/mozilla/expected.html:

19762

19763

2005-06-14 Geoffrey Garen <ggaren@apple.com>

19764

19765

Fixed: <rdar://problem/4147745> JavaScript discards locally defined "arguments" property

19766

19767

No layout tests added because this change fixes existing tests:

19768

ecma/ExecutionContexts/10.1.6.js

19769

ecma_3/Function/regress-94506.js

19770

js1_4/Functions/function-001.js

19771

19772

Reviewed by cblu.

19773

19774

* kjs/function.cpp:

19775

(KJS::ActivationImp::get): get now checks for an "arguments" property defined in the local variable object

19776

before trying to return the built-in arguments array.

19777

19778

* kjs/function.h: ActivationImp::put no longer overrides ObjectImp::put

19779

19780

2005-06-10 Darin Adler <darin@apple.com>

19781

19782

Change by Mark Rowe <opendarwin.org@bdash.net.nz>.

19783

Reviewed by me.

19784

19785

- further improvements to exception file/line number fix

19786

19787

* kjs/nodes.h: Added setExceptionDetailsIfNeeded function.

19788

* kjs/nodes.cpp: Updated macros to call the new setExceptionDetailsIfNeeded function.

19789

(Node::setExceptionDetailsIfNeeded): Added.

19790

19791

2005-06-09 Darin Adler <darin@apple.com>

19792

19793

Change by Mark Rowe <opendarwin.org@bdash.net.nz>

19794

Reviewed by me.

19795

19796

* kjs/nodes.cpp: Get rid of unneeded this->.

19797

19798

2005-06-08 Maciej Stachowiak <mjs@apple.com>

19799

19800

Change by Mark Rowe <opendarwin.org@bdash.net.nz>

19801

Reviewed by me.

19802

19803

- fixed http://bugs.webkit.org/show_bug.cgi?id=3327

19804

(Exception When Setting Style to Invalid Value Lacks Line/File Information)

19805

19806

* kjs/nodes.cpp: Include source file and line number when making exception in

19807

KJS_CHECKEXCEPTIONVALUE.

19808

19809

2005-06-07 Darin Adler <darin@apple.com>

19810

19811

Change by Toby Peterson <toby@opendarwin.org>.

19812

Reviewed by me.

19813

19814

* JavaScriptCore.pbproj/project.pbxproj: Allow bison 2.0, which generates the file

19815

with a different name.

19816

19817

2005-06-07 Darin Adler <darin@apple.com>

19818

19819

Change by Toby Peterson <toby@opendarwin.org>.

19820

Reviewed by me.

19821

19822

* kjs/grammar.y: Remove bogus extra line from grammar.y. Toby got this change from KDE KJS.

19823

19824

2005-06-06 Darin Adler <darin@apple.com>

19825

19826

* tests/mozilla/run-mozilla-tests: Wrote a perl version of this so we don't require

19827

the "jst" tool to run the tests.

19828

19829

2005-06-04 Darin Adler <darin@apple.com>

19830

19831

Reviewed by Maciej.

19832

19833

- add libicu headers

19834

19835

* JavaScriptCore.pbproj/project.pbxproj: Added icu directory to header search path.

19836

19837

* icu/README: Added.

19838

* icu/unicode/platform.h: Added.

19839

* icu/unicode/uchar.h: Added.

19840

* icu/unicode/uconfig.h: Added.

19841

* icu/unicode/umachine.h: Added.

19842

* icu/unicode/urename.h: Added.

19843

* icu/unicode/utf.h: Added.

19844

* icu/unicode/utf16.h: Added.

19845

* icu/unicode/utf8.h: Added.

19846

* icu/unicode/utf_old.h: Added.

19847

* icu/unicode/utypes.h: Added.

19848

* icu/unicode/uversion.h: Added.

19849

19850

2005-05-19 Darin Adler <darin@apple.com>

19851

19852

Reviewed by Maciej.

19853

19854

- turned off exceptions and RTTI; seems to cut JavaScriptCore code size by about 22%

19855

19856

* JavaScriptCore.pbproj/project.pbxproj: Turn off exceptions and RTTI for both

19857

the framework and testkjs tool.

19858

19859

2005-05-18 Darin Adler <darin@apple.com>

19860

19861

Reviewed by Maciej.

19862

19863

- got rid of code that depended on RTTI

19864

19865

* kjs/collector.cpp:

19866

(KJS::className): Added. Gets class name in a KJS way, rather than a C++ RTTI way.

19867

(KJS::Collector::rootObjectClasses): Use className instead of typeid names.

19868

19869

2005-05-18 Darin Adler <darin@apple.com>

19870

19871

Reviewed by Maciej.

19872

19873

- fix a failure seen in the Mozilla JavaScript tests where a live object was garbage-collected

19874

when the only reference to it was in an argList on the stack

19875

19876

* kjs/list.h: Moved the operator= function into the .cpp file since it's too big to be

19877

a good choice to inline.

19878

* kjs/list.cpp: (KJS::List::operator=): Moved this formerly-inline function into a separate

19879

file and added missing code to update valueRefCount. It's the latter that fixes the bug.

19880

19881

2005-05-16 Darin Adler <darin@apple.com>

19882

19883

Reviewed by Adele.

19884

19885

- fixed issues preventing us from compiling with newer versions of gcc 4.0

19886

19887

* kjs/ustring.cpp:

19888

(KJS::operator==): Remove redundant and illegal KJS:: prefix on this function's definition.

19889

(KJS::operator<): Ditto.

19890

(KJS::compare): Ditto.

19891

19892

2005-05-09 Darin Adler <darin@apple.com>

19893

19894

Reviewed by John.

19895

19896

- turn on conservative GC unconditionally and start on SPI changes to

19897

eliminate the now-unneeded smart pointers since we don't ref count any more

19898

19899

* kjs/value.h: Removed macros to turn conservative GC on and off.

19900

Removed ref and deref functions.

19901

(KJS::ValueImp::ValueImp): Removed non-conservative-GC code path.

19902

(KJS::ValueImp::isUndefined): Added. New SPI to make it easier to deal with ValueImp directly.

19903

(KJS::ValueImp::isNull): Ditto.

19904

(KJS::ValueImp::isBoolean): Ditto.

19905

(KJS::ValueImp::isNumber): Ditto.

19906

(KJS::ValueImp::isString): Ditto.

19907

(KJS::ValueImp::isObject): Ditto.

19908

(KJS::Value::Value): Removed non-conservative-GC code path and made constructor no

19909

longer explicit so we can quietly create Value wrappers from ValueImp *; inexpensive with

19910

conservative GC and eases the transition.

19911

(KJS::Value::operator ValueImp *): Added. Quietly creates ValueImp * from Value.

19912

(KJS::ValueImp::marked): Removed non-conservative-GC code path.

19913

19914

* kjs/value.cpp:

19915

(KJS::ValueImp::mark): Removed non-conservative-GC code path.

19916

(KJS::ValueImp::isUndefinedOrNull): Added. New SPI to make it easier to deal with ValueImp directly.

19917

(KJS::ValueImp::isBoolean): Ditto.

19918

(KJS::ValueImp::isNumber): Ditto.

19919

(KJS::ValueImp::isString): Ditto.

19920

(KJS::ValueImp::asString): Ditto.

19921

(KJS::ValueImp::isObject): Ditto.

19922

(KJS::undefined): Ditto.

19923

(KJS::null): Ditto.

19924

(KJS::boolean): Ditto.

19925

(KJS::string): Ditto.

19926

(KJS::zero): Ditto.

19927

(KJS::one): Ditto.

19928

(KJS::two): Ditto.

19929

(KJS::number): Ditto.

19930

19931

* kjs/object.h: Made constructor no longer explicit so we can quietly create Object

19932

wrappers from ObjectImp *; inexpensive with conservative GC and eases the transition.

19933

(KJS::Object::operator ObjectImp *): Added. Quietly creates ObjectImp * from Object.

19934

(KJS::ValueImp::isObject): Added. Implementation of new object-related ValueImp function.

19935

(KJS::ValueImp::asObject): Ditto.

19936

19937

* kjs/object.cpp:

19938

(KJS::ObjectImp::setInternalValue): Remove non-conservative-GC code path.

19939

(KJS::ObjectImp::putDirect): Ditto.

19940

(KJS::error): Added. Function in the new SPI style to create an error object.

19941

19942

* kjs/internal.h: Added the new number-constructing functions as friends of NumberImp.

19943

There may be a more elegant way to do this later; what's important now is the new SPI.

19944

19945

* kjs/collector.h: Remove non-conservative-GC code path and also take out some

19946

unneeded APPLE_CHANGES.

19947

19948

* bindings/runtime_root.cpp:

19949

(KJS::Bindings::addNativeReference): Remove non-conservative-GC code path.

19950

(KJS::Bindings::removeNativeReference): Ditto.

19951

(RootObject::removeAllNativeReferences): Ditto.

19952

* bindings/runtime_root.h:

19953

(KJS::Bindings::RootObject::~RootObject): Ditto.

19954

(KJS::Bindings::RootObject::setRootObjectImp): Ditto.

19955

* kjs/collector.cpp:

19956

(KJS::Collector::allocate): Ditto.

19957

(KJS::Collector::collect): Ditto.

19958

(KJS::Collector::numGCNotAllowedObjects): Ditto.

19959

(KJS::Collector::numReferencedObjects): Ditto.

19960

(KJS::Collector::rootObjectClasses): Ditto.

19961

* kjs/internal.cpp:

19962

(NumberImp::create): Ditto.

19963

(InterpreterImp::globalInit): Ditto.

19964

(InterpreterImp::globalClear): Ditto.

19965

* kjs/list.cpp:

19966

(KJS::List::markProtectedLists): Ditto.

19967

(KJS::List::clear): Ditto.

19968

(KJS::List::append): Ditto.

19969

* kjs/list.h:

19970

(KJS::List::List): Ditto.

19971

(KJS::List::deref): Ditto.

19972

(KJS::List::operator=): Ditto.

19973

* kjs/protect.h:

19974

(KJS::gcProtect): Ditto.

19975

(KJS::gcUnprotect): Ditto.

19976

19977

2005-05-09 Chris Blumenberg <cblu@apple.com>

19978

19979

Workaround gcc 3.3 internal compiler errors.

19980

19981

Reviewed by darin.

19982

19983

* bindings/objc/WebScriptObject.mm:

19984

(-[WebScriptObject stringRepresentation]): call [NSString stringWithCString:encoding] rather than using @""

19985

19986

2005-05-09 Darin Adler <darin@apple.com>

19987

19988

* Makefile.am: Don't set up PBXIntermediatesDirectory explicitly;

19989

Not needed to make builds work, spews undesirable error messages too.

19990

19991

2005-05-06 Darin Adler <darin@apple.com>

19992

19993

Reviewed by Maciej.

19994

19995

- make building multiple trees with make work better

19996

19997

* Makefile.am: Set up Xcode build directory before invoking xcodebuild.

19998

19999

2005-05-04 Maciej Stachowiak <mjs@apple.com>

20000

20001

Reviewed by Darin.

20002

20003

<rdar://problem/4086570> Crash in JavaScriptCore with RSS Visualizer

20004

20005

* kjs/internal.cpp:

20006

(InterpreterImp::mark): mark staticNaN, it is usually protected by the Number

20007

prototype but there is a small window where it can get collected.

20008

20009

2005-05-04 Darin Adler <darin@apple.com>

20010

20011

Reviewed by Dave Hyatt.

20012

20013

- another gcc-4.0-related fix

20014

20015

* bindings/runtime_root.h: Take off extra namespace prefixes that apparently cause problems

20016

compiling with gcc 4.0, although I have not observed the problems.

20017

20018

2005-05-04 Darin Adler <darin@apple.com>

20019

20020

Reviewed by Dave Hyatt.

20021

20022

- fixed build rules to match other projects

20023

20024

* JavaScriptCore.pbproj/project.pbxproj: Set deployment target to 10.3 in the build styles.

20025

When built without a build style (by Apple B&I) we want to get the target from the

20026

environment. But when built with a build style (by Safari engineers and others), we want

20027

to use 10.3.

20028

20029

* Makefile.am: Took out extra parameters that make command-line building different from

20030

Xcode building. Now that this is fixed, you should not get a full rebuild if you switch

20031

from command line to Xcode or back.

20032

20033

2005-05-04 Maciej Stachowiak <mjs@apple.com>

20034

20035

- revert presumably accidental change to mozilla JS test expected results, this

20036

was making the tests fail.

20037

20038

* tests/mozilla/expected.html:

20039

20040

2005-05-03 Richard Williamson <rjw@apple.com>

20041

20042

Fixed <rdar://problem/4102644> Crash in LiveConnect below KJS::Bindings::JavaInstance::stringValue() const

20043

20044

Correctly handle accessing nil objects from a Java object array.

20045

20046

Reviewed by John.

20047

20048

* bindings/jni/jni_runtime.cpp:

20049

(JavaArray::valueAt):

20050

20051

2005-05-01 Darin Adler <darin@apple.com>

20052

20053

- move to Xcode native targets and stop checking in generated files

20054

20055

* JavaScriptCore.pbproj/project.pbxproj: Updated to use native targets and generate all the generated

20056

files, so we don't have to check them in any more.

20057

* Info.plist: Added. Native targets use a separate file for this.

20058

20059

* Makefile.am: Removed pcre and kjs SUBDIRS. Also removed code that deleted the embedded copy of this

20060

framework, since we haven't been embedding it for some time.

20061

20062

* kjs/grammar_wrapper.cpp: Added. Shell used to compile grammar.cpp since we can't add a generated file

20063

easily to the list of files to be compiled.

20064

20065

* kjs/.cvsignore: Removed.

20066

* kjs/Makefile.am: Removed.

20067

* kjs/array_object.lut.h: Removed.

20068

* kjs/date_object.lut.h: Removed.

20069

* kjs/grammar.cpp: Removed.

20070

* kjs/grammar.cpp.h: Removed.

20071

* kjs/grammar.h: Removed.

20072

* kjs/lexer.lut.h: Removed.

20073

* kjs/math_object.lut.h: Removed.

20074

* kjs/number_object.lut.h: Removed.

20075

* kjs/string_object.lut.h: Removed.

20076

* pcre/.cvsignore: Removed.

20077

* pcre/Makefile.am: Removed.

20078

* pcre/chartables.c: Removed.

20079

20080

2005-04-28 Darin Adler <darin@apple.com>

20081

20082

Reviewed by Dave Harrison.

20083

20084

- fixed problems preventing us from compiling with gcc 4.0

20085

20086

* JavaScriptCore.pbproj/project.pbxproj: Removed -Wmissing-prototypes from

20087

WARNING_CPLUSPLUSFLAGS since it's now a C-only warning.

20088

20089

* bindings/jni/jni_jsobject.cpp:

20090

(JSObject::getSlot): Changed some %d to %ld where the parameters where long ints.

20091

(JSObject::setSlot): Ditto.

20092

* bindings/jni/jni_utility.cpp:

20093

(KJS::Bindings::getJavaVM): Ditto.

20094

(KJS::Bindings::getJNIEnv): Ditto.

20095

* bindings/objc/objc_utility.mm: Fixed include of <JavascriptCore/internal.h> that needed the

20096

letter "S" capitalized.

20097

* kjs/bool_object.cpp: (BooleanProtoFuncImp::call): Rearranged how this function returns to

20098

avoid incorrect gcc 4.0 warning.

20099

* kjs/collector.cpp: (KJS::Collector::markStackObjectsConservatively): Changed code to check

20100

the alignment of the passed-in pointers to only require pointer-level alignment, not 8-byte alignment.

20101

Prevents a crash on garbage collect when compiled with gcc 4.0.

20102

* kjs/nodes.cpp:

20103

(WhileNode::execute): Added a redundant return after an infinite loop to work around incorrect gcc 4.0 warning.

20104

(ForNode::execute): Ditto.

20105

(SwitchNode::execute):Rearranged how this function returns to avoid incorrect gcc 4.0 warning.

20106

(LabelNode::execute): Ditto.

20107

* kjs/string_object.cpp: (replace): Ditto.

20108

20109

2005-04-26 Richard Williamson <rjw@apple.com>

20110

20111

Fixed <rdar://problem/4098713> Scripting API is incompatible with Mozilla

20112

20113

We were incompatible with Mozilla's implementation of the scripting APIs in

20114

two ways:

20115

20116

Their NPN_SetException has the following signature:

20117

20118

void NPN_SetException(NPObject *npobj, const NPUTF8 *message);

20119

20120

ours has:

20121

20122

void NPN_SetException (NPObject * npobj, const NPString *message);

20123

20124

Also, they expect the string returned from NPN_UTF8FromIdentifier() to be freed by caller.

20125

We do not.

20126

20127

I changed both behaviors to match Mozilla.

20128

20129

Reviewed by Chris.

20130

20131

* bindings/NP_jsobject.cpp:

20132

(_NPN_SetException):

20133

* bindings/npruntime.cpp:

20134

(_NPN_UTF8FromIdentifier):

20135

(_NPN_IntFromIdentifier):

20136

(_NPN_SetExceptionWithUTF8):

20137

* bindings/npruntime.h:

20138

* bindings/npruntime_impl.h:

20139

20140

2005-04-26 Maciej Stachowiak <mjs@apple.com>

20141

20142

Reviewed by Chris.

20143

20144

<rdar://problem/4092136> reproducible crash in KJS::kjs_fast_realloc loading maps.google.com

20145

20146

* kjs/string_object.cpp:

20147

(StringObjectFuncImp::call): Allocate adopted ustring buffer properly.

20148

20149

2005-04-22 Darin Adler <darin@apple.com>

20150

20151

Reviewed by Maciej.

20152

20153

* kjs/ustring.cpp: (KJS::UString::UTF8String): Fix off-by-one error in surrogate pair logic.

20154

20155

2005-04-22 Darin Adler <darin@apple.com>

20156

20157

Reviewed by John.

20158

20159

- fixed <rdar://problem/4090046> JavaScript throw statement causes parse error when no semicolon is present

20160

20161

* kjs/grammar.y: Added an additional rule for throw like the ones we have for all the other semicolon rules.

20162

Not sure why we missed this one earlier.

20163

20164

* kjs/grammar.cpp: Regenerated.

20165

20166

=== JavaScriptCore-412.1 ===

20167

20168

2005-04-20 Darin Adler <darin@apple.com>

20169

20170

Reviewed by Maciej.

20171

20172

- speedups, total 12% on JavaScript iBench

20173

20174

I ran the benchmark under Shark and followed its advice a lot, mainly.

20175

20176

* kjs/collector.cpp:

20177

(KJS::Collector::allocate): Take out special case for 0; costing speed but unexercised.

20178

Use numLiveObjectsAtLastCollect instead of numAllocationsSinceLastCollect so we don't

20179

have to bump it each time we call allocate. Put numLiveObjects into a local variable to

20180

cut down on global variable accesses. Make "next" cell pointer be a byte offset rather

20181

than a pointer so we don't need a special case for NULL. Allow freeList to point to some

20182

bogus item when the entire block is full rather than going out of our way to make it

20183

point to NULL.

20184

(KJS::Collector::markProtectedObjects): Get table size and pointer into locals outside

20185

the loop to avoid re-loading them over and over again.

20186

(KJS::Collector::collect): Put numLiveObjects into a local variable to cut down on global

20187

variable accesses. Make "next" cell pointer be a byte offset as above. Put numLiveObjects

20188

into a local variable to cut down on global variable accesses. Set numLiveObjectsAtLastCollect

20189

rather than numAllocationsSinceLastCollect.

20190

(KJS::Collector::numReferencedObjects): Get table size and pointer into locals outside

20191

the loop to avoid re-loading them over and over again.

20192

(KJS::Collector::rootObjectClasses): Ditto.

20193

20194

* kjs/internal.h: Make Value be a friend of NumberImp so it can construct number objects

20195

directly, avoiding the conversion from Number to Value.

20196

20197

* kjs/internal.cpp: (StringImp::toObject): Don't use Object::dynamicCast, because we know

20198

the thing is an object and we don't want to do all the extra work; just cast directly.

20199

20200

* kjs/list.cpp: (KJS::List::List): Construct valueRefCount in a way that avoids the need for

20201

a branch -- in the hot case this just meant avoiding checking a variable we just set to false.

20202

20203

* kjs/lookup.cpp: (keysMatch): Marked this inline.

20204

20205

* kjs/nodes.cpp: Disabled KJS_BREAKPOINT, to avoid calling hitStatement all the time.

20206

(BooleanNode::evaluate): Make a Value directly, rather than making a Boolean which is converted

20207

into a Value.

20208

(NumberNode::evaluate): Ditto.

20209

(StringNode::evaluate): Ditto.

20210

(ArrayNode::evaluate): Ditto.

20211

(FunctionCallNode::evaluate): Use new inline baseIfMutable to avoid unnecessary getBase function.

20212

Also just use a pointer for func, rather than an Object.

20213

(PostfixNode::evaluate): Change code so that it doesn't make an excess Number, and so that it

20214

passes a "known to be integer" boolean in, often avoiding a conversion from floating point to

20215

integer and back.

20216

(DeleteNode::evaluate): Make a Value directly.

20217

(TypeOfNode::evaluate): Use new inline baseIfMutable and make Value directly.

20218

(PrefixNode::evaluate): Change code so that it doesn't make an excess Number, and so that it

20219

passes a "known to be integer" boolean in, often avoiding a conversion from floating point to

20220

integer and back.

20221

(UnaryPlusNode::evaluate): Make a Value directly.

20222

(NegateNode::evaluate): Change code so that it doesn't make an excess Number, and so that it

20223

passes a "known to be integer" boolean in, often avoiding a conversion from floating point to

20224

integer and back.

20225

(BitwiseNotNode::evaluate): Make a Value directly.

20226

(LogicalNotNode::evaluate): Ditto.

20227

(ShiftNode::evaluate): Don't convert to a double before making a Value.

20228

(RelationalNode::evaluate): Make a Value directly.

20229

(EqualNode::evaluate): Ditto.

20230

(BitOperNode::evaluate): Ditto.

20231

(AssignNode::evaluate): Make a Value directly. Change code so that it passes a "known to be integer"

20232

boolean in, often avoiding a conversion from floating point to integer and back.

20233

(VarDeclNode::evaluate): Make a Value directly.

20234

(ForNode::execute): Remove unused local variable.

20235

20236

* kjs/operations.h:

20237

(KJS::isNaN): Inlined.

20238

(KJS::isInf): Ditto.

20239

(KJS::isPosInf): Ditto.

20240

(KJS::isNegInf): Ditto.

20241

20242

* kjs/operations.cpp: Change isNaN, isInf, isPosInf, and isNegInf to be inlines.

20243

(KJS::equal): Rewrite to avoid creating values and recursing back into the function.

20244

(KJS::relation): Rearranged code so that we don't need explicit isNaN checks.

20245

(KJS::add): Changed code to make Value directly, and so that it passes a "known to be integer"

20246

boolean in, often avoiding a conversion from floating point to integer and back.

20247

(KJS::mult): Ditto.

20248

20249

* kjs/property_map.cpp:

20250

(KJS::PropertyMap::~PropertyMap): Get size and entries pointer outside loop to avoid

20251

re-getting them inside the loop.

20252

(KJS::PropertyMap::clear): Ditto. Clear value pointer in addition to key, so we can just

20253

look at the value pointer in the mark function.

20254

(KJS::PropertyMap::get): Get sizeMask and entries pointer outside loop to avoid

20255

re-getting them inside the loop.

20256

(KJS::PropertyMap::put): Ditto.

20257

(KJS::PropertyMap::insert): Ditto.

20258

(KJS::PropertyMap::remove): Ditto.

20259

(KJS::PropertyMap::mark): Get size and entries pointer outside loop to avoid

20260

re-getting them inside the loop. Don't bother checking key for 0, since we already have

20261

to check value for 0. (Also had to change clear() to set value to 0.)

20262

(KJS::PropertyMap::addEnumerablesToReferenceList): Get size and entries pointer outside

20263

loop to avoid re-getting them inside the loop.

20264

(KJS::PropertyMap::addSparseArrayPropertiesToReferenceList): Ditto.

20265

(KJS::PropertyMap::save): Ditto.

20266

20267

- other changes

20268

20269

* kjs/protected_values.h: Remove unneeded class name qualifiers.

20270

20271

* kjs/reference.h:

20272

(KJS::Reference::baseIfMutable): New inline function: replaces isMutable().

20273

(KJS::Reference::Reference): Inlined.

20274

* kjs/reference.cpp:

20275

(KJS::Reference::getValue): Rewrite to not use getBase.

20276

(KJS::Reference::putValue): Ditto.

20277

(KJS::Reference::deleteValue): Dittol

20278

20279

* kjs/simple_number.h:

20280

(KJS::SimpleNumber::integerFits): Added. For use when the parameter is known to be integral.

20281

20282

* kjs/string_object.cpp: (StringProtoFuncImp::call): Create the number without first converting

20283

to double in various cases that involve integers.

20284

20285

* kjs/ustring.h:

20286

(KJS::UString::attach): Inlined.

20287

(KJS::UString::release): Inlined.

20288

* kjs/ustring.cpp:

20289

(KJS::UString::find): Get first character outside the loop instead of re-fetching it each time.

20290

20291

* kjs/value.cpp:

20292

(Value::Value): Added overloads for all the various specific types of values, so you don't have

20293

to convert from, say, Number to Value, just to create one.

20294

(Number::Number): Added an overload that takes a boolean to indicate the number is already

20295

known to be an integer.

20296

20297

* kjs/value.h: Added more Value constructors, added a version of toNumber that returns

20298

a boolean to indicate if the number is known to be an integer (because it was a "simple number").

20299

(KJS::ValueImp::marked): Inlined.

20300

(KJS::ValueImp::dispatchType): Inlined.

20301

(KJS::ValueImp::dispatchToPrimitive): Inlined.

20302

(KJS::ValueImp::dispatchToBoolean): Inlined.

20303

(KJS::ValueImp::dispatchToNumber): Inlined.

20304

(KJS::ValueImp::dispatchToString): Inlined.

20305

(KJS::ValueImp::dispatchToUInt32): Inlined.

20306

20307

2005-04-14 Maciej Stachowiak <mjs@apple.com>

20308

20309

- make fast_malloc.h a private header, not project

20310

20311

* JavaScriptCore.pbproj/project.pbxproj:

20312

20313

2005-04-12 Maciej Stachowiak <mjs@apple.com>

20314

20315

Reviewed by Richard.

20316

20317

<rdar://problem/4089734> JavaScript iBench can be sped up ~10% with custom allocator

20318

20319

- use custom single-threaded malloc for all non-GC JavaScriptCore

20320

allocations, for a 9.1% speedup on JavaScript iBench

20321

20322

* JavaScriptCore.pbproj/project.pbxproj:

20323

* kjs/collector.cpp:

20324

(KJS::Collector::allocate): Use dlmalloc to allocate the collector blocks.

20325

(KJS::Collector::collect): And dlfree to free it.

20326

* kjs/fast_malloc.cpp: Added, just the standard dlmalloc here.

20327

* kjs/fast_malloc.h: Added. Declarations for the functions. Also added a handy

20328

macro to give a class custom operator new/delete

20329

* kjs/identifier.cpp:

20330

(KJS::Identifier::add): Use dlmalloc/dlfree.

20331

* kjs/nodes.h: make nodes KJS_FAST_ALLOCATED.

20332

* kjs/property_map.cpp:

20333

(KJS::PropertyMap::~PropertyMap): Use dlmalloc/dlfree.

20334

(KJS::PropertyMap::rehash): ditto

20335

* kjs/scope_chain.h:

20336

* kjs/ustring.cpp:

20337

(KJS::UString::Rep::createCopying): New named constructor that copies a passed-in

20338

buffer, to hide allocation details from webcore.

20339

(KJS::UString::UString): use createCopying when appropriate.

20340

(KJS::UString::Rep::destroy): Use dlmalloc/dlfree.

20341

(KJS::UString::expandedSize): likewise

20342

(KJS::UString::expandCapacity): likewise

20343

(KJS::UString::expandPreCapacity): likewise

20344

(KJS::UString::spliceSubstringsWithSeparators): likewise

20345

(KJS::UString::append): likewise

20346

(KJS::UString::operator=): likewise

20347

(KJS::UString::detach): likewise

20348

* kjs/ustring.h: make UString and UString::Rep KJS_FAST_ALLOCATED.

20349

20350

2005-04-11 Maciej Stachowiak <mjs@apple.com>

20351

20352

Reviewed by John.

20353

20354

<rdar://problem/4086819> Avoid using protect count hash table so much for 5.6% JS iBench speedup

20355

20356

- Avoid using protected values hash for the two most common cases

20357

- Bump up ListImp high water mark, new testing shows 508 ListImps are

20358

created during JS iBench.

20359

20360

Net result is a 5.6% speedup on JavaScript iBench

20361

20362

* kjs/collector.cpp:

20363

(KJS::Collector::collect): mark protected lists as appropriate.

20364

* kjs/context.h:

20365

* kjs/list.cpp:

20366

(KJS::ListImp::markValues): Moved implementation from List::markValues

20367

(KJS::List::markProtectedLists): Implemented - scan pool and overflow

20368

list.

20369

(KJS::allocateListImp): link lists outside the pool into a separate

20370

doubly linked list to be able to mark protected lists

20371

(KJS::deallocateListImp): do the corresponding delinking

20372

(KJS::List::derefValues): do nothing in conservative GC mode

20373

(KJS::List::refValues): do nothing in conservative GC mode

20374

(KJS::List::markValues): call ListImp version

20375

(KJS::List::append):

20376

* kjs/list.h:

20377

20378

=== Safari-412 ===

20379

20380

=== Safari-411 ===

20381

20382

=== Safari-410 ===

20383

20384

=== Safari-409 ===

20385

20386

=== Safari-408 ===

20387

20388

=== Safari-407 ===

20389

20390

2005-03-16 Jens Alfke <jens@apple.com>

20391

20392

Reviewed by Kevin.

20393

20394

Fix for <rdar://problem/4025212> "REGRESSION (163-164): search not performed correctly; united.com"

20395

JavaScript unescape("") was returning a messed-up String object that appeared identical to an empty string, but would in some cases act as 'null' when passed to native functions, in this case the Option() constructor.

20396

In the implementation of unescape, the UString holding the result was not initialized to "", so it started out as a null string. If nothing was appended to it, it remained null, resulting in a JavaScript String object with some bad behaviors (namely, converting it to a DOMStringImpl results in a NULL pointer.)

20397

Darin says this regression occurred when we replaced our own implementation of unescape() with code from KJS.

20398

20399

* kjs/function.cpp:

20400

(KJS::GlobalFuncImp::call):

20401

20402

2005-03-15 Richard Williamson <rjw@apple.com>

20403

20404

Fixed <rdar://problem/4053276> WebScripting protocol in WebKit cannot convert Boolean in Javascript to BOOL in Objective-C

20405

20406

Added JavaScript boolean to type that can be converted to

20407

ObjC scalar parameters.

20408

20409

Reviewed by Ken Kocienda.

20410

20411

* bindings/objc/objc_utility.mm:

20412

(KJS::Bindings::convertValueToObjcValue):

20413

20414

=== Safari-406 ===

20415

20416

=== Safari-405 ===

20417

20418

=== Safari-403 ===

20419

20420

=== Safari-402 ===

20421

20422

=== Safari-401 ===

20423

20424

=== Safari-400 ===

20425

20426

=== Safari-188 ===

20427

20428

2005-02-21 Darin Adler <darin@apple.com>

20429

20430

* kjs/date_object.cpp: (timetUsingCF): Fixed indenting.

20431

20432

2005-02-17 Richard Williamson <rjw@apple.com>

20433

20434

Fixed <rdar://problem/4003251> Safari crashed at www.icelandair.com in LiveConnect code converting a Java object to a string

20435

20436

Added nil check.

20437

20438

Reviewed by John Sullivan.

20439

20440

* bindings/jni/jni_runtime.cpp:

20441

(JavaField::valueFromInstance):

20442

20443

=== Safari-187 ===

20444

20445

2005-02-11 Richard Williamson <rjw@apple.com>

20446

20447

Fixed <rdar://problem/3985118> DOM objects not being marshaled on JS->native calls

20448

20449

Re-factored how 'native' wrappers for JS objects are created. The interpreter now

20450

creates these wrappers. The WebCore subclass of the interpreter now overrides

20451

createLanguageInstanceForValue() and creates a DOM ObjC wrapper for DOM objects.

20452

20453

Reviewed by Ken.

20454

20455

* bindings/c/c_utility.cpp:

20456

(convertValueToNPVariant):

20457

* bindings/jni/jni_instance.cpp:

20458

(JavaInstance::invokeMethod):

20459

* bindings/jni/jni_objc.mm:

20460

(KJS::Bindings::dispatchJNICall):

20461

* bindings/jni/jni_runtime.cpp:

20462

(JavaField::valueFromInstance):

20463

(JavaArray::valueAt):

20464

* bindings/objc/WebScriptObject.mm:

20465

(-[WebScriptObject _setExecutionContext:KJS::Bindings::]):

20466

(+[WebScriptObject _convertValueToObjcValue:KJS::originExecutionContext:Bindings::executionContext:Bindings::]):

20467

* bindings/objc/WebScriptObjectPrivate.h:

20468

* bindings/objc/objc_utility.h:

20469

* bindings/objc/objc_utility.mm:

20470

(KJS::Bindings::convertObjcValueToValue):

20471

(KJS::Bindings::createObjcInstanceForValue):

20472

* bindings/runtime.cpp:

20473

(Instance::createBindingForLanguageInstance):

20474

(Instance::createRuntimeObject):

20475

(Instance::createLanguageInstanceForValue):

20476

* bindings/runtime.h:

20477

* kjs/interpreter.cpp:

20478

(Interpreter::createLanguageInstanceForValue):

20479

* kjs/interpreter.h:

20480

20481

=== Safari-186 ===

20482

20483

2005-02-10 Darin Adler <darin@apple.com>

20484

20485

"Reviewed" by Richard (he told me the file was obsolete).

20486

20487

- got rid of an obsolete file

20488

20489

* bindings/npsap.h: Removed.

20490

20491

=== Safari-185 ===

20492

20493

=== Safari-183 ===

20494

20495

2005-02-03 Richard Williamson <rjw@apple.com>

20496

20497

Fixed <rdar://problem/3972905> CrashTracer: ...36 crashes at com.apple.WebCore: khtml::CSSStyleSelector::applyDeclarations + 120

20498

20499

Revert to old (and correct) behavior of returning runtime object

20500

when passed as a parameter, rather than it's corresponding DOM

20501

object.

20502

20503

Reviewed by Chris.

20504

20505

* bindings/objc/WebScriptObject.mm:

20506

(+[WebScriptObject _convertValueToObjcValue:KJS::originExecutionContext:Bindings::executionContext:Bindings::]):

20507

20508

=== Safari-182 ===

20509

20510

2005-01-28 Richard Williamson <rjw@apple.com>

20511

20512

Fixed <rdar://problem/3980389> JavaScript bindings access incorrect runtime object

20513

20514

Only use special 'back door' property to get the runtime object if thisObj isn't

20515

already a runtime object.

20516

20517

<gratuitous> Cleaned up a couple of strcmp on ClassInfo name. Used == on

20518

ClassInfo pointer instead.

20519

20520

Reviewed by Chris.

20521

20522

* bindings/c/c_utility.cpp:

20523

(convertValueToNPVariant):

20524

* bindings/objc/WebScriptObject.mm:

20525

(+[WebScriptObject _convertValueToObjcValue:KJS::originExecutionContext:Bindings::executionContext:Bindings::]):

20526

* bindings/runtime_method.cpp:

20527

(RuntimeMethodImp::call):

20528

20529

=== Safari-181 ===

20530

20531

2005-01-26 Richard Williamson <rjw@apple.com>

20532

20533

Fixed <rdar://problem/3972522> (179-180) 40% slowdown on iBench JavaScript test

20534

20535

I added a member variable to ObjectImp. This changed it's size and consequently

20536

hampered the optimizations built into the garbage collector. Objects no longer

20537

fit within the allocators cell size, and thus allocation fell back to a slower

20538

allocator.

20539

20540

As a result of this fix I also dramatically cleaned up how runtime objects are

20541

accessed. The path mostly *removes* code.

20542

20543

Reviewed by Chris.

20544

20545

* bindings/runtime_method.cpp:

20546

(RuntimeMethodImp::call):

20547

* bindings/runtime_object.cpp:

20548

(RuntimeObjectImp::get):

20549

(RuntimeObjectImp::put):

20550

(RuntimeObjectImp::canPut):

20551

(RuntimeObjectImp::hasProperty):

20552

(RuntimeObjectImp::defaultValue):

20553

* bindings/runtime_object.h:

20554

* kjs/object.cpp:

20555

(KJS::ObjectImp::ObjectImp):

20556

* kjs/object.h:

20557

20558

2005-01-20 Darin Adler <darin@apple.com>

20559

20560

Reviewed by me, changes by Han Ming Ong.

20561

20562

- <rdar://problem/3964302> SWB: A few files need to be updated to be compilable under GCC 4.0

20563

20564

* bindings/objc/WebScriptObjectPrivate.h: Make members public.

20565

* kjs/lookup.h: Change "value.h" to "object.h" because we need KJS::Object to compile a template.

20566

20567

2005-01-20 Richard Williamson <rjw@apple.com>

20568

20569

Fixed <rdar://problem/3964634> undefined property value from binding seems to evaluate to true in an if statement

20570

20571

The comprehensive fix for this problem requires new API, as described in 3965326. However,

20572

given that we can't add new API at this point, the 'ObjcFallbackObjectImp' will behave

20573

like and Undefined object if invokeUndefinedMethodFromWebScript:withArguments: isn't

20574

implemented on the bound object.

20575

20576

Reviewed by Chris.

20577

20578

* bindings/objc/objc_runtime.h:

20579

* bindings/objc/objc_runtime.mm:

20580

(ObjcFallbackObjectImp::type):

20581

(ObjcFallbackObjectImp::implementsCall):

20582

(ObjcFallbackObjectImp::toBoolean):

20583

* bindings/testbindings.mm:

20584

(+[MyFirstInterface isSelectorExcludedFromWebScript:]):

20585

(+[MyFirstInterface isKeyExcludedFromWebScript:]):

20586

20587

=== Safari-180 ===

20588

20589

2005-01-19 Richard Williamson <rjw@apple.com>

20590

20591

Fixed <rdar://problem/3853676> Browser Crash when accessing CCWeb Progress Page - KJS::Bindings::convertValueToJValue

20592

20593

Fixed the following problems with LiveConnect that are demonstrated by the application

20594

described in 3853676.

20595

20596

1. If a nil object is passed in an array from Java to JavaScript we will crash.

20597

2. We sometimes will incorrectly attempt to access a generic JavaScript as a Java runtime object wrapper.

20598

3. We will sometimes fail to find the correct static method ID.

20599

20600

Reviewed by Maciej.

20601

20602

* bindings/jni/jni_jsobject.cpp:

20603

(JSObject::convertJObjectToValue):

20604

(JSObject::listFromJArray):

20605

* bindings/jni/jni_runtime.cpp:

20606

(JavaField::valueFromInstance):

20607

(JavaField::setValueToInstance):

20608

* bindings/jni/jni_utility.cpp:

20609

(KJS::Bindings::getMethodID):

20610

(KJS::Bindings::convertValueToJValue):

20611

* bindings/runtime_array.h:

20612

20613

2005-01-18 Richard Williamson <rjw@apple.com>

20614

20615

Fixed several issues all arising from analysis of plugin detection code at ifilm.com:

20616

20617

Fixed <rdar://problem/3958592> can't script plug-ins if plug-in is invoked with <object> element instead of <embed>

20618

Fixed <rdar://problem/3958597> <object> elements with IDs do not show up as named properties of the document

20619

Fixed <rdar://problem/3960973> DOM objects for plugin elements are not accessible

20620

Fixed <rdar://problem/3958601> need an additional class ID in WebCore for the Real plug-in

20621

20622

We now support accessing scriptable plugin objects that are specified with <applet>, <embed>, or <object>

20623

tags. Also, if any of these elements are named they can be accessed from the document or window objects.

20624

Finally, DOM methods are properties will be forwarded appropriately for the plugin's root scriptable object.

20625

20626

Reviewed by Chris.

20627

20628

* bindings/objc/objc_instance.h:

20629

* bindings/objc/objc_instance.mm:

20630

(ObjcInstance::supportsSetValueOfUndefinedField):

20631

* bindings/runtime.h:

20632

(KJS::Bindings::Instance::supportsSetValueOfUndefinedField):

20633

* bindings/runtime_object.cpp:

20634

(RuntimeObjectImp::RuntimeObjectImp):

20635

(RuntimeObjectImp::get):

20636

(RuntimeObjectImp::put):

20637

(RuntimeObjectImp::canPut):

20638

(RuntimeObjectImp::hasProperty):

20639

(RuntimeObjectImp::defaultValue):

20640

* bindings/runtime_object.h:

20641

(KJS::RuntimeObjectImp::fallbackObject):

20642

* kjs/object.cpp:

20643

(KJS::ObjectImp::ObjectImp):

20644

* kjs/object.h:

20645

(KJS::ObjectImp::forwardingScriptMessage):

20646

(KJS::ObjectImp::setForwardingScriptMessage):

20647

20648

2005-01-18 Richard Williamson <rjw@apple.com>

20649

20650

Back out a change that was incorrectly committed yesterday.

20651

20652

Reviewed by Chris.

20653

20654

* bindings/objc/objc_utility.mm:

20655

(KJS::Bindings::convertValueToObjcValue):

20656

20657

2005-01-17 Richard Williamson <rjw@apple.com>

20658

20659

Fixed <rdar://problem/3753030> Need to ensure same origin for plugin binding invocations (origin security rules)

20660

20661

Keep track of originating execution context and target execution

20662

context for native JS object wrappers, and perform appropriate

20663

security checks.

20664

20665

Reviewed by David Harrison.

20666

20667

* bindings/NP_jsobject.cpp:

20668

(_isSafeScript):

20669

(_NPN_CreateScriptObject):

20670

(_NPN_Invoke):

20671

(_NPN_Evaluate):

20672

(_NPN_GetProperty):

20673

(_NPN_SetProperty):

20674

(_NPN_RemoveProperty):

20675

(_NPN_HasProperty):

20676

(_NPN_HasMethod):

20677

(_NPN_SetException):

20678

* bindings/NP_jsobject.h:

20679

* bindings/c/c_instance.cpp:

20680

(CInstance::CInstance):

20681

(CInstance::stringValue):

20682

* bindings/c/c_instance.h:

20683

* bindings/c/c_utility.cpp:

20684

(convertValueToNPVariant):

20685

* bindings/jni/jni_instance.cpp:

20686

(JavaInstance::JavaInstance):

20687

(JavaInstance::valueOf):

20688

* bindings/jni/jni_instance.h:

20689

* bindings/objc/WebScriptObject.mm:

20690

(-[WebScriptObject _initializeWithObjectImp:KJS::originExecutionContext:Bindings::executionContext:Bindings::]):

20691

(-[WebScriptObject _initWithObjectImp:KJS::originExecutionContext:Bindings::executionContext:Bindings::]):

20692

(-[WebScriptObject KJS::Bindings::]):

20693

(-[WebScriptObject _setOriginExecutionContext:KJS::Bindings::]):

20694

(-[WebScriptObject _isSafeScript]):

20695

(-[WebScriptObject callWebScriptMethod:withArguments:]):

20696

(-[WebScriptObject evaluateWebScript:]):

20697

(-[WebScriptObject setValue:forKey:]):

20698

(-[WebScriptObject valueForKey:]):

20699

(-[WebScriptObject removeWebScriptKey:]):

20700

(-[WebScriptObject stringRepresentation]):

20701

(-[WebScriptObject webScriptValueAtIndex:]):

20702

(-[WebScriptObject setWebScriptValueAtIndex:value:]):

20703

(+[WebScriptObject _convertValueToObjcValue:KJS::originExecutionContext:Bindings::executionContext:Bindings::]):

20704

* bindings/objc/WebScriptObjectPrivate.h:

20705

* bindings/objc/objc_instance.h:

20706

* bindings/objc/objc_runtime.mm:

20707

(convertValueToObjcObject):

20708

* bindings/objc/objc_utility.mm:

20709

(KJS::Bindings::convertValueToObjcValue):

20710

* bindings/runtime.cpp:

20711

(Instance::Instance):

20712

(Instance::operator=):

20713

* bindings/runtime.h:

20714

(KJS::Bindings::Instance::Instance):

20715

(KJS::Bindings::Instance::setExecutionContext):

20716

(KJS::Bindings::Instance::executionContext):

20717

* bindings/runtime_root.cpp:

20718

(RootObject::setInterpreter):

20719

* bindings/runtime_root.h:

20720

* kjs/interpreter.h:

20721

(KJS::Interpreter::isGlobalObject):

20722

(KJS::Interpreter::interpreterForGlobalObject):

20723

(KJS::Interpreter::isSafeScript):

20724

20725

=== Safari-179 ===

20726

20727

2005-01-13 Vicki Murley <vicki@apple.com>

20728

20729

Reviewed by Adele.

20730

20731

- fix <rdar://problem/3946836> Safari about box lists 2004 instead of 2005

20732

20733

* JavaScriptCore.pbproj/project.pbxproj: bump "2004" to "2005"

20734

20735

2005-01-12 Richard Williamson <rjw@apple.com>

20736

20737

Avoid additional work on dealloc by adding early out to

20738

removeNativeReference(). (This will save time on dealloc

20739

for all ObjC DOM objects.)

20740

20741

Reviewed by Darin.

20742

20743

* bindings/runtime_root.cpp:

20744

(KJS::Bindings::removeNativeReference):

20745

20746

2005-01-12 Richard Williamson <rjw@apple.com>

20747

20748

Fixed <rdar://problem/3923356> REGRESSION: Java/JavaScript security checks working incorrectly

20749

20750

We were always returning the first "root" object for all runtime

20751

objects. Changed 0 in loop to i, the index.

20752

20753

Reviewed by David Harrison.

20754

20755

* bindings/runtime_root.cpp:

20756

(KJS::Bindings::rootForImp):

20757

20758

2005-01-11 Richard Williamson <rjw@apple.com>

20759

20760

Fixed <rdar://problem/3887930> Must use new Java plug-in API to get/set fields so exception handling works (fixes many LiveConnect crashes)

20761

20762

Use the new dispatching API to invoke JNI, rather than calling JNI

20763

directly.

20764

20765

Reviewed by David Harrison.

20766

20767

* bindings/jni/jni_instance.cpp:

20768

(JavaInstance::invokeMethod):

20769

* bindings/jni/jni_runtime.cpp:

20770

(JavaField::dispatchValueFromInstance):

20771

(JavaField::valueFromInstance):

20772

(JavaField::dispatchSetValueToInstance):

20773

(JavaField::setValueToInstance):

20774

* bindings/jni/jni_runtime.h:

20775

* bindings/jni/jni_utility.cpp:

20776

(KJS::Bindings::convertValueToJValue):

20777

20778

=== Safari-178 ===

20779

20780

=== Safari-177 ===

20781

20782

=== Safari-176 ===

20783

20784

2004-12-17 Maciej Stachowiak <mjs@apple.com>

20785

20786

Reviewed by Kevin.

20787

20788

<rdar://problem/3926869> Opening caches window after running PLT causes crash

20789

20790

* kjs/protected_values.cpp:

20791

(KJS::ProtectedValues::getProtectCount): Don't include simple numbers in

20792

the protected value table.

20793

(KJS::ProtectedValues::increaseProtectCount): Ditto.

20794

(KJS::ProtectedValues::decreaseProtectCount): Ditto.

20795

20796

2004-12-16 Darin Adler <darin@apple.com>

20797

20798

Reviewed by Maciej.

20799

20800

- fixed <rdar://problem/3920764> Unimplemented String methods toLocaleLowerCase and toLocaleUpperCase

20801

20802

* kjs/string_object.h: Added toLocaleLowerCase and toLocaleUpperCase.

20803

* kjs/string_object.cpp: (StringProtoFuncImp::call): Made locale versions be synonmyms for the

20804

non-locale-specific versions.

20805

* kjs/string_object.lut.h: Regenerated.

20806

20807

2004-12-14 Richard Williamson <rjw@apple.com>

20808

20809

Pass URL of plugin view when call into JNI.

20810

20811

Reviewed by Chris.

20812

20813

* bindings/jni/jni_objc.mm:

20814

(KJS::Bindings::dispatchJNICall):

20815

20816

2004-12-13 Richard Williamson <rjw@apple.com>

20817

20818

Fixed <rdar://problem/3827799> repro. crash with IBM Rational ClearCase Web under Safari (Java/LiveConnect-related)

20819

20820

Add support for calling static Java methods from JavaScript.

20821

20822

Reviewed by Maciej.

20823

20824

* bindings/jni/jni_instance.cpp:

20825

(JavaInstance::invokeMethod):

20826

* bindings/jni/jni_runtime.cpp:

20827

(JavaMethod::JavaMethod):

20828

* bindings/jni/jni_runtime.h:

20829

(KJS::Bindings::JavaMethod::isStatic):

20830

* bindings/jni/jni_utility.cpp:

20831

(callJNIStaticMethod):

20832

(KJS::Bindings::callJNIBooleanMethod):

20833

(KJS::Bindings::callJNIStaticBooleanMethod):

20834

* bindings/jni/jni_utility.h:

20835

20836

2004-12-13 Richard Williamson <rjw@apple.com>

20837

20838

Fixed <rdar://problem/3887767> LiveConnect doesn't propagate Java exceptions back to JavaScript (prevents security suite from running)

20839

20840

Reviewed by John.

20841

20842

* bindings/jni/jni_instance.cpp:

20843

(JavaInstance::invokeMethod):

20844

* bindings/jni/jni_objc.mm:

20845

(KJS::Bindings::dispatchJNICall):

20846

* bindings/jni/jni_runtime.h:

20847

* bindings/jni/jni_utility.h:

20848

20849

=== Safari-175 ===

20850

20851

2004-12-07 Maciej Stachowiak <mjs@apple.com>

20852

20853

Reviewed by Darin.

20854

20855

<rdar://problem/3908017> REGRESSION (172-173): assertion in ObjectImp::construct trying to create JS error (24hourfitness.com)

20856

20857

The fix was to implement copy constructor and assignment operator,

20858

the ones that worked on the base class did not replace the

20859

defaults apparently!

20860

20861

* kjs/protect.h:

20862

(KJS::ProtectedValue::ProtectedValue):

20863

(KJS::ProtectedValue::operator=):

20864

(KJS::ProtectedObject::ProtectedObject):

20865

(KJS::ProtectedObject::operator=):

20866

20867

Also fixed a bug in the GC test mode that compares the results of

20868

the old collector and the new collector.

20869

20870

* kjs/value.cpp:

20871

(ValueImp::mark):

20872

20873

=== Safari-173 ===

20874

20875

2004-11-23 Richard Williamson <rjw@apple.com>

20876

20877

Fixed <rdar://problem/3890385> field and method cache incorrectly capped (c bindings)

20878

20879

Reviewed by Ken.

20880

20881

* bindings/c/c_class.cpp:

20882

(CClass::_commonInit):

20883

20884

2004-11-21 Maciej Stachowiak <mjs@apple.com>

20885

20886

Reviewed by Ken.

20887

20888

<rdar://problem/3889696> Enable conservative garbage collection for JavaScript

20889

20890

* kjs/collector.cpp:

20891

(KJS::Collector::Thread::Thread):

20892

(KJS::destroyRegisteredThread):

20893

(KJS::initializeRegisteredThreadKey):

20894

(KJS::Collector::registerThread):

20895

(KJS::Collector::markStackObjectsConservatively):

20896

(KJS::Collector::markCurrentThreadConservatively):

20897

(KJS::Collector::markOtherThreadConservatively):

20898

* kjs/collector.h:

20899

* kjs/internal.cpp:

20900

(lockInterpreter):

20901

* kjs/value.h:

20902

20903

=== Safari-172 ===

20904

20905

2004-11-15 Richard Williamson <rjw@apple.com>

20906

20907

Fixed <rdar://problem/3880561> Default string value of ObjC object in JS should be [obj description].

20908

20909

Reviewed by Hyatt.

20910

20911

* bindings/objc/objc_instance.mm:

20912

(ObjcInstance::stringValue):

20913

* bindings/objc/objc_utility.h:

20914

* bindings/objc/objc_utility.mm:

20915

(KJS::Bindings::convertNSStringToString):

20916

(KJS::Bindings::convertObjcValueToValue):

20917

20918

=== Safari-171 ===

20919

20920

2004-11-09 Chris Blumenberg <cblu@apple.com>

20921

20922

Fixed: <rdar://problem/3872724> soft link against JavaVM to save ~2MB RSHRD

20923

20924

Reviewed by rjw.

20925

20926

* ChangeLog:

20927

* JavaScriptCore.pbproj/project.pbxproj: don't link against JavaVM

20928

* bindings/softlinking.c: Added.

20929

(loadFramework): new

20930

(getFunctionPointer): new

20931

(JNI_GetCreatedJavaVMs): load JavaVM if not already loaded, get _JNI_GetCreatedJavaVMs symbol if we don't already have it, call JNI_GetCreatedJavaVMs

20932

20933

=== Safari-170 ===

20934

20935

2004-11-04 Darin Adler <darin@apple.com>

20936

20937

Reviewed by Ken.

20938

20939

- fixed <rdar://problem/3865365> since -[WebScriptObject dealloc] does not call [super dealloc], the build will fail due to a warning

20940

- fixed behavior so that [[WebScriptObject alloc] initWithCoder:] doesn't leak WebUndefined instances

20941

and incidentally so that [[WebScriptObject alloc] init] returns the single shared instance rather

20942

than allocating a new one

20943

20944

* bindings/objc/WebScriptObject.mm: Removed some stray semicolons.

20945

(+[WebUndefined allocWithZone:]): Made this the common bottleneck that returns the single instance

20946

of WebUndefined, since it's the single method that normally allocates new instances. Calls super to

20947

actually allocate only the very first time it's called.

20948

(-[WebUndefined initWithCoder:]): Simplified to just return self (no reason to re-lookup the single

20949

shared instance since there can be only one).

20950

(-[WebUndefined copyWithZone:]): Ditto.

20951

(-[WebUndefined retain]): Ditto.

20952

(-[WebUndefined retainCount]): Use UINT_MAX constant here (matches usage in NSObject.m for retain count

20953

of class).

20954

(-[WebUndefined autorelease]): Simplified to just return self (see above).

20955

(-[WebUndefined copy]): No need to override this since it just turns around and calls copyWithZone:.

20956

(-[WebUndefined dealloc]): Added an assertion since this method should never be called. Also added

20957

a call to [super dealloc] after return; to make the new -Wdealloc-check compiler happy (fixing the

20958

bug mentioned above).

20959

(+[WebUndefined undefined]): Reimplemented; calls allocWithZone:NULL to get to the shared instance.

20960

No need to call init, since that's a no-op for this class.

20961

20962

2004-11-03 David Harrison <harrison@apple.com>

20963

20964

Reviewed by Darin.

20965

20966

Eliminate the use of a marker file to determine how to build.

20967

20968

* .cvsignore:

20969

* Makefile.am:

20970

20971

2004-11-01 Richard Williamson <rjw@apple.com>

20972

20973

Fixed <rdar://problem/3861469> Latest Real player crashes Safari on some sites.

20974

20975

Reviewed by Ken.

20976

20977

* bindings/c/c_instance.cpp:

20978

(CInstance::invokeMethod):

20979

(CInstance::invokeDefaultMethod):

20980

Initialize out parameters to void type.

20981

20982

* bindings/c/c_runtime.cpp:

20983

(CField::valueFromInstance):

20984

(CField::setValueToInstance):

20985

Initialize out parameters to void type.

20986

Also added additional checks to protect against classes that

20987

don't implement all functions.

20988

20989

2004-11-01 Richard Williamson <rjw@apple.com>

20990

20991

Fixed <rdar://problem/3861257> WebUndefined should be returned for undefined values

20992

20993

Reviewed by John.

20994

20995

* ChangeLog:

20996

* bindings/objc/WebScriptObject.mm:

20997

(+[WebScriptObject _convertValueToObjcValue:KJS::root:Bindings::]):

20998

Added additional conversion Undefined -> WebUndefined.

20999

* bindings/objc/objc_utility.mm:

21000

(KJS::Bindings::convertObjcValueToValue):

21001

Added additional conversion WebUndefined -> Undefined.

21002

21003

2004-11-01 Darin Adler <darin@apple.com>

21004

21005

- fixed <rdar://problem/3855573> Remove reference to "WebScriptMethods" from WebScriptObject.h comments

21006

21007

* bindings/objc/WebScriptObject.h: Removed unneeded #ifdef protection for multiple includes (since

21008

this is an Objective-C header and we use #import for those). Fixed comments as requested in the bug

21009

report to match the contents of the file.

21010

21011

=== Safari-169 ===

21012

21013

=== Safari-168 ===

21014

21015

2004-10-22 Ken Kocienda <kocienda@apple.com>

21016

21017

Reviewed by me

21018

21019

* JavaScriptCore.pbproj/project.pbxproj:

21020

Add GCC_ENABLE_OBJC_GC and GCC_FAST_OBJC_DISPATCH flags.

21021

21022

=== Safari-167 ===

21023

21024

2004-10-13 Richard Williamson <rjw@apple.com>

21025

21026

Moved boolean checks prior to NSNumber checks. booleans are

21027

NSNumbers.

21028

21029

Follow on to <rdar://problem/3821515> binding layer needs to convert NSNumber-bools to js type boolean not number.

21030

21031

Reviewed by John.

21032

21033

* bindings/objc/objc_utility.mm:

21034

(KJS::Bindings::convertObjcValueToValue):

21035

21036

2004-10-12 Richard Williamson <rjw@apple.com>

21037

21038

Fixed access to DOM object via WebScriptObject API.

21039

The execution context for DOM objects wasn't being found.

21040

<rdar://problem/3831372> The valueForKey method for @"offsetLeft" on a paragraph element causes a crash.

21041

21042

Reviewed by Chris.

21043

21044

* bindings/objc/WebScriptObject.mm:

21045

(_didExecute):

21046

(-[WebScriptObject KJS::Bindings::]):

21047

(-[WebScriptObject callWebScriptMethod:withArguments:]):

21048

(-[WebScriptObject evaluateWebScript:]):

21049

(-[WebScriptObject setValue:forKey:]):

21050

(-[WebScriptObject valueForKey:]):

21051

(-[WebScriptObject stringRepresentation]):

21052

* bindings/objc/WebScriptObjectPrivate.h:

21053

21054

2004-10-09 Darin Adler <darin@apple.com>

21055

21056

Reviewed by Kevin.

21057

21058

- fixed <rdar://problem/3804661> REGRESSION: JavaScriptCore framework now has two init routines

21059

21060

* bindings/NP_jsobject.cpp: Fixed unnecessarily-complex globals set up that was

21061

creating an init routine.

21062

21063

* kjs/ustring.cpp: Changed around the UString::Rep::empty construction to not

21064

require a global constructor that creates an init routine.

21065

21066

2004-10-09 Darin Adler <darin@apple.com>

21067

21068

Reviewed by Kevin.

21069

21070

- fixed <rdar://problem/3822618> REGRESSION (164-165): expedia.com's popup help doesn't work

21071

21072

* kjs/reference.cpp: (Reference::putValue): Change so that references not found in any object

21073

work with the window object of the page the function is in, not the page of the caller. This

21074

is what all other browsers do. This code was hidden before by the "everything is defined on

21075

window object" hack in WebCore.

21076

21077

2004-10-07 Richard Williamson <rjw@apple.com>

21078

21079

Added simple JavaScript call tracing. Very useful for

21080

debugging complex pages.

21081

21082

Tracing is only available in development builds and is

21083

enabled by:

21084

21085

(gdb) set traceJavaScript = 1

21086

21087

or programatically

21088

21089

setTraceJavaScript(true)

21090

21091

Function, args, and return values are printed to console. Very

21092

verbose.

21093

21094

Reviewed by Ken.

21095

21096

* kjs/function_object.cpp:

21097

(FunctionProtoFuncImp::call):

21098

* kjs/object.cpp:

21099

(KJS::Object::call):

21100

21101

=== Safari-166 ===

21102

21103

2004-10-05 Richard Williamson <rjw@apple.com>

21104

21105

Fixed <rdar://problem/3819234> NPN_SetException (and throwException:) isn't implemented

21106

21107

Reviewed by Chris.

21108

21109

* bindings/NP_jsobject.cpp:

21110

(_NPN_SetException):

21111

* bindings/npruntime.cpp:

21112

(_NPN_SetExceptionWithUTF8):

21113

* bindings/objc/WebScriptObject.mm:

21114

(+[WebScriptObject throwException:]):

21115

* kjs/internal.h:

21116

(KJS::InterpreterImp::context):

21117

21118

2004-10-05 Richard Williamson <rjw@apple.com>

21119

21120

Fixed <rdar://problem/3821515> binding layer needs to convert NSNumber-bools to js type boolean not number

21121

21122

Reviewed by Ken.

21123

21124

* bindings/objc/objc_utility.mm:

21125

(KJS::Bindings::convertObjcValueToValue):

21126

21127

2004-10-04 Darin Adler <darin@apple.com>

21128

21129

Reviewed by Ken.

21130

21131

- rolled in a fix the KDE folks did for the operations that generate HTML fragments

21132

21133

* kjs/string_object.cpp: (StringProtoFuncImp::call): Added quote marks to generated HTML.

21134

21135

- rolled out an old workaround we don't need any more

21136

21137

* JavaScriptCore.pbproj/project.pbxproj: Remove -Wno-long-double because the <math.h> issue that

21138

required it is no longer there.

21139

21140

2004-09-30 Richard Williamson <rjw@apple.com>

21141

21142

Fixed <rdar://problem/3821215> NPN hasMethod and hasProperty functions should take NPObjects, not NPClass

21143

21144

Reviewed by Chris.

21145

21146

* bindings/NP_jsobject.cpp:

21147

(_NPN_GetProperty):

21148

(_NPN_HasProperty):

21149

(_NPN_HasMethod):

21150

* bindings/c/c_class.cpp:

21151

(CClass::methodsNamed):

21152

(CClass::fieldNamed):

21153

* bindings/c/c_class.h:

21154

* bindings/c/c_instance.cpp:

21155

(CInstance::invokeMethod):

21156

* bindings/jni/jni_class.cpp:

21157

(JavaClass::methodsNamed):

21158

* bindings/jni/jni_class.h:

21159

* bindings/npruntime.h:

21160

* bindings/objc/objc_class.h:

21161

* bindings/objc/objc_class.mm:

21162

(ObjcClass::methodsNamed):

21163

* bindings/runtime.h:

21164

* bindings/runtime_object.cpp:

21165

(RuntimeObjectImp::get):

21166

(RuntimeObjectImp::hasProperty):

21167

21168

2004-09-29 Chris Blumenberg <cblu@apple.com>

21169

21170

Prepended underscores to NPN methods so that when the QT plug-in loads these symbols, it uses the non-underscore versions in WebKit. Without this, the QT plug-in was failing to load when launching Safari from the command-line.

21171

21172

Reviewed by rjw.

21173

21174

* JavaScriptCore.pbproj/project.pbxproj:

21175

* bindings/NP_jsobject.cpp:

21176

(_NPN_CreateScriptObject):

21177

(_NPN_InvokeDefault):

21178

(_NPN_Invoke):

21179

(_NPN_Evaluate):

21180

(_NPN_GetProperty):

21181

(_NPN_SetProperty):

21182

(_NPN_RemoveProperty):

21183

(_NPN_HasProperty):

21184

(_NPN_HasMethod):

21185

* bindings/c/c_class.cpp:

21186

(CClass::methodsNamed):

21187

(CClass::fieldNamed):

21188

* bindings/c/c_instance.cpp:

21189

(CInstance::CInstance):

21190

(CInstance::~CInstance):

21191

(CInstance::operator=):

21192

(CInstance::invokeMethod):

21193

(CInstance::invokeDefaultMethod):

21194

* bindings/c/c_runtime.cpp:

21195

* bindings/c/c_runtime.h:

21196

(KJS::Bindings::CField::name):

21197

(KJS::Bindings::CMethod::name):

21198

* bindings/npruntime.cpp:

21199

(_NPN_GetStringIdentifier):

21200

(_NPN_GetStringIdentifiers):

21201

(_NPN_GetIntIdentifier):

21202

(_NPN_IdentifierIsString):

21203

(_NPN_UTF8FromIdentifier):

21204

(_NPN_IntFromIdentifier):

21205

(NPN_InitializeVariantWithObject):

21206

(_NPN_ReleaseVariantValue):

21207

(_NPN_CreateObject):

21208

(_NPN_RetainObject):

21209

(_NPN_ReleaseObject):

21210

(_NPN_SetExceptionWithUTF8):

21211

(_NPN_SetException):

21212

21213

2004-09-26 Darin Adler <darin@apple.com>

21214

21215

* kjs/string_object.cpp: (StringProtoFuncImp::call): Remove strange use of high() and

21216

low() to get Unicode value of character, and just use unicode().

21217

21218

2004-09-26 Darin Adler <darin@apple.com>

21219

21220

- refine charAt/charCodeAt fix slightly

21221

21222

* kjs/string_object.cpp: (StringProtoFuncImp::call): Treat undefined the same was as an

21223

omitted parameter, as we do everywhere else, and as other browsers do here.

21224

21225

2004-09-26 Darin Adler <darin@apple.com>

21226

21227

Reviewed by Kevin.

21228

21229

- fixed <rdar://problem/3816097> REGRESSION: mailblocks, and presumably many other pages, failing because variable not found

21230

21231

* kjs/internal.cpp: (InterpreterImp::evaluate): Process variable declarations before executing

21232

the program. We were doing this properly for functions, but not entire programs.

21233

21234

- fixed <rdar://problem/3814706> REGRESSION: text fields in mailblocks wizards do not accept keystrokes due to use of charCodeAt()

21235

21236

* kjs/string_object.cpp: (StringProtoFuncImp::call): Changed the implementation of charAt

21237

and charCodeAt to treat a missing parameter as an index of 0, rather than an invalid index.

21238

21239

* tests/mozilla/expected.html: Update for two tests that now pass with these changes.

21240

21241

=== Safari-165 ===

21242

21243

=== Safari-164 ===

21244

21245

2004-09-14 Richard Williamson <rjw@apple.com>

21246

21247

1. Add class parameter to object allocation function. This is somewhat redundant, given that

21248

the allocation function is in the class function vector, but people wanted to use the same

21249

allocation function for different classes.

21250

21251

2. Renamed NPN_Class to NPN_Invoke to match the name in the function vector.

21252

21253

3. Add support for a default function on an object. This is a feature that ActiveX supports,

21254

and will allow JavaScript code to be written that will look exactly the same for both ActiveX

21255

plugins and Netscape or WebKit plugins. There are implementations included for the 'C' and

21256

'Objective-C' bindings.

21257

21258

There bugs are covered by

21259

21260

<rdar://problem/3776343> Support for default functions in the JavaScript bindings

21261

<rdar://problem/3779186> NPN_Call needs to be renamed to NPN_Invoke

21262

<rdar://problem/3674754> Need to implement latest npruntime.h

21263

21264

Reviewed by John.

21265

21266

* bindings/NP_jsobject.cpp:

21267

(jsAllocate):

21268

(NPN_InvokeDefault):

21269

(NPN_Invoke):

21270

* bindings/c/c_class.cpp:

21271

* bindings/c/c_instance.cpp:

21272

(CInstance::CInstance):

21273

(CInstance::operator=):

21274

(CInstance::invokeMethod):

21275

(CInstance::invokeDefaultMethod):

21276

* bindings/c/c_instance.h:

21277

* bindings/c/c_runtime.cpp:

21278

* bindings/c/c_runtime.h:

21279

* bindings/jni/jni_instance.cpp:

21280

(JavaInstance::invokeDefaultMethod):

21281

* bindings/jni/jni_instance.h:

21282

* bindings/npruntime.cpp:

21283

(NPN_CreateObject):

21284

* bindings/npruntime.h:

21285

* bindings/objc/WebScriptObject.h:

21286

* bindings/objc/objc_class.mm:

21287

(ObjcClass::fallbackObject):

21288

* bindings/objc/objc_instance.h:

21289

* bindings/objc/objc_instance.mm:

21290

(ObjcInstance::invokeDefaultMethod):

21291

* bindings/objc/objc_runtime.h:

21292

* bindings/objc/objc_runtime.mm:

21293

(ObjcFallbackObjectImp::ObjcFallbackObjectImp):

21294

(ObjcFallbackObjectImp::get):

21295

(ObjcFallbackObjectImp::put):

21296

(ObjcFallbackObjectImp::canPut):

21297

(ObjcFallbackObjectImp::implementsCall):

21298

(ObjcFallbackObjectImp::call):

21299

(ObjcFallbackObjectImp::hasProperty):

21300

(ObjcFallbackObjectImp::deleteProperty):

21301

(ObjcFallbackObjectImp::defaultValue):

21302

* bindings/runtime.h:

21303

(KJS::Bindings::Class::fallbackObject):

21304

(KJS::Bindings::Instance::getValueOfUndefinedField):

21305

(KJS::Bindings::Instance::setValueOfUndefinedField):

21306

(KJS::Bindings::Instance::valueOf):

21307

* bindings/runtime_object.cpp:

21308

(RuntimeObjectImp::implementsCall):

21309

(RuntimeObjectImp::call):

21310

* bindings/runtime_object.h:

21311

21312

2004-09-13 Maciej Stachowiak <mjs@apple.com>

21313

21314

Reviewed by Darin.

21315

21316

<rdar://problem/3794735> Gmail- sending a very long message with Safari is so slow it seems like a hang

21317

21318

* kjs/string_object.cpp:

21319

(StringProtoFuncImp::call): Replaced implementation of replace()

21320

method with function below...

21321

(replace): In order to avoid excessive allocation and copying,

21322

figure out the ranges of the original string and replacement

21323

strings to be assembled, instead of constantly creating new

21324

strings at each substitution. The old behavior is basically O(N^2)

21325

for a global replace on a pattern that matches many places in the

21326

string.

21327

(regExpIsGlobal): Helper function for the above.

21328

(expandSourceRanges): ditto

21329

(pushSourceRange): ditto

21330

(expandReplacements): ditto

21331

(pushReplacement): ditto

21332

* kjs/ustring.cpp:

21333

(KJS::UString::spliceSubstringsWithSeparators): New method that

21334

pieces together substring ranges of this string together with

21335

specified separators, all at one go.

21336

* kjs/ustring.h:

21337

(KJS::UString::Range::Range): Added new helper class to represent

21338

substring choices.

21339

21340

2004-09-14 Maciej Stachowiak <mjs@apple.com>

21341

21342

Reviewed by Darin.

21343

21344

- fixed <rdar://problem/3800315> encode-URI-test layout test is failing

21345

21346

* kjs/function.cpp:

21347

(KJS::GlobalFuncImp::call): Make sure to escape null

21348

characters. This is a bug in the new code that made part of the

21349

test fail.

21350

21351

2004-09-13 Darin Adler <darin@apple.com>

21352

21353

Reviewed by Kevin and Maciej.

21354

21355

- new function to support fix for DIG bug in WebCore

21356

21357

* kjs/scope_chain.h: Added new push function that pushes another entire scope chain.

21358

* kjs/scope_chain.cpp: (KJS::ScopeChain::push): Ditto.

21359

21360

2004-09-12 Darin Adler <darin@apple.com>

21361

21362

* tests/mozilla/expected.html: Updated test results for 3 more tests that pass with the new version

21363

of escape and unescape.

21364

21365

2004-09-12 Darin Adler <darin@apple.com>

21366

21367

Reviewed by Maciej.

21368

21369

- fixed <rdar://problem/3798209> any non-ASCII characters are garbled in the result of toLocaleString

21370

21371

* kjs/date_object.cpp:

21372

(formatLocaleDate): Replaced two old functions that used LongDateTime with this one new function that

21373

uses CFDateFormatter.

21374

(DateProtoFuncImp::call): Call the new formatLocaleDate instead of both formatLocaleDate and formatLocaleTime.

21375

21376

2004-09-09 Maciej Stachowiak <mjs@apple.com>

21377

21378

Reviewed by Richard.

21379

21380

<rdar://problem/3493140> REGRESSION (85-100): cedille displays %-escaped in JavaScript message at hotmail.com

21381

21382

* kjs/function.cpp:

21383

(KJS::GlobalFuncImp::call): Replace our escape() and unescape() implementations with

21384

ones from KDE KJS, which have the proper latin-1 behavior to match Win IE.

21385

* kjs/lexer.cpp:

21386

(Lexer::isHexDigit): Made static and non-const.

21387

* kjs/lexer.h:

21388

21389

=== Safari-163 ===

21390

21391

2004-09-06 Darin Adler <darin@apple.com>

21392

21393

* JavaScriptCore.pbproj/project.pbxproj: Bump MACOSX_DEPLOYMENT_TARGET to 10.3.

21394

21395

=== Safari-162 ===

21396

21397

2004-09-01 Richard Williamson <rjw@apple.com>

21398

21399

Add pid to exception messages (to help debug dashboard clients).

21400

21401

Reviewed by Chris.

21402

21403

* kjs/interpreter.cpp:

21404

(Interpreter::evaluate):

21405

21406

=== Safari-161 ===

21407

21408

2004-08-20 Richard Williamson <rjw@apple.com>

21409

21410

Implemented new JNI abstraction. We no longer invoke Java methods

21411

directly with JNI, rather we call into the plugin. This allows the

21412

plugin to dispatch the call to the appropriate VM thread. This

21413

change should (will?) fix a whole class of threading related problems with

21414

the Java VM.

21415

21416

Reviewed by Hyatt.

21417

21418

* JavaScriptCore.pbproj/project.pbxproj:

21419

* bindings/c/c_instance.h:

21420

(KJS::Bindings::CInstance::setExecutionContext):

21421

(KJS::Bindings::CInstance::executionContext):

21422

* bindings/jni/jni_instance.cpp:

21423

(JavaInstance::JavaInstance):

21424

(JavaInstance::invokeMethod):

21425

(JavaInstance::setExecutionContext):

21426

(JavaInstance::executionContext):

21427

* bindings/jni/jni_instance.h:

21428

* bindings/jni/jni_jsobject.cpp:

21429

(JSObject::convertJObjectToValue):

21430

* bindings/jni/jni_runtime.cpp:

21431

(JavaField::JavaField):

21432

(JavaArray::convertJObjectToArray):

21433

(JavaField::valueFromInstance):

21434

(JavaArray::JavaArray):

21435

(JavaArray::valueAt):

21436

* bindings/jni/jni_runtime.h:

21437

(KJS::Bindings::JavaArray::operator=):

21438

(KJS::Bindings::JavaArray::executionContext):

21439

* bindings/jni/jni_utility.h:

21440

* bindings/objc/objc_instance.h:

21441

(KJS::Bindings::ObjcInstance::setExecutionContext):

21442

(KJS::Bindings::ObjcInstance::executionContext):

21443

* bindings/runtime.cpp:

21444

(Instance::createBindingForLanguageInstance):

21445

* bindings/runtime.h:

21446

* bindings/runtime_root.h:

21447

(KJS::Bindings::RootObject::nativeHandle):

21448

21449

=== Safari-158 ===

21450

21451

2004-08-19 Vicki Murley <vicki@apple.com>

21452

21453

Reviewed by John.

21454

21455

* kjs/property_map.cpp:

21456

(KJS::PropertyMap::put): initialize deletedElementIndex to zero, to make the compiler happy

21457

21458

2004-08-17 Darin Adler <darin@apple.com>

21459

21460

Reviewed by Adele.

21461

21462

- fixed <rdar://problem/3746676> SAP WebDynpro app hangs inside JavaScript property map hash table code (deleted sentinel problem)

21463

21464

* kjs/property_map.h: Added some private functions.

21465

* kjs/property_map.cpp:

21466

(KJS::PropertyMap::clear): Set sentinelCount to 0.

21467

(KJS::PropertyMap::put): Complete search for the element before choosing to use the deleted-element sentinel.

21468

Also keep sentinel count up to date when we destroy a sentinel by overwriting with a new added element.

21469

(KJS::PropertyMap::expand): Added. Calls rehash with a size 2x the old size, or 16.

21470

(KJS::PropertyMap::rehash): Added. Refactored the rehash code into a separate function.

21471

(KJS::PropertyMap::remove): Add one to sentinelCount, and rehash if 1/4 or more of the elements are

21472

deleted-element sentinels.

21473

(KJS::PropertyMap::checkConsistency): Check the sentinelCount.

21474

21475

2004-08-16 Maciej Stachowiak <mjs@apple.com>

21476

21477

Code change by Eric Albert, reviewd by me.

21478

21479

<rdar://problem/3571960> washingtonpost.com claims I don't have cookies enabled and won't let me read articles

21480

21481

* kjs/date_object.cpp:

21482

(timetUsingCF): Clamp time to LONG_MAX (getting rid of time_t

21483

entirely would be even better, but is not required to fix this bug.

21484

21485

=== Safari-157 ===

21486

21487

2004-08-16 Richard Williamson <rjw@apple.com>

21488

21489

Fixed <rdar://problem/3581092> cash in KJS::Bindings::JSObject::eval at tcvetantcvetkov.com

21490

21491

Adds bullet proofing to protect against evaluation of bogus JS in all the flavors of bindings (Java, C, and ObjC).

21492

21493

Reviewed by Chris.

21494

21495

* bindings/NP_jsobject.cpp:

21496

(NPN_Evaluate):

21497

* bindings/jni/jni_jsobject.cpp:

21498

(JSObject::eval):

21499

* bindings/objc/WebScriptObject.mm:

21500

(-[WebScriptObject evaluateWebScript:]):

21501

21502

2004-08-15 Richard Williamson <rjw@apple.com>

21503

21504

More updates to np headers. Implemented new NPN functions.

21505

21506

Reviewed by Darin.

21507

21508

* bindings/NP_jsobject.cpp:

21509

(NPN_HasProperty):

21510

(NPN_HasMethod):

21511

* bindings/npapi.h:

21512

* bindings/npruntime.h:

21513

21514

2004-08-13 Darin Adler <darin@apple.com>

21515

21516

- fix build so we can compile again

21517

21518

* bindings/npapi.h: Added. Richard forgot to check this in. The one I'm checking in here

21519

is good enough so that we can compile, but it's only a stopgap measure, because I think

21520

Richard has a newer one he wants to check in.

21521

21522

2004-08-12 Richard Williamson <rjw@apple.com>

21523

21524

Bring npruntime.h and friends closer to compliance with

21525

latest spec.

21526

21527

Reviewed by Maciej.

21528

21529

* JavaScriptCore.pbproj/project.pbxproj:

21530

* bindings/NP_jsobject.cpp:

21531

(jsAllocate):

21532

(_NPN_CreateScriptObject):

21533

(NPN_Call):

21534

(NPN_Evaluate):

21535

(NPN_GetProperty):

21536

(NPN_SetProperty):

21537

(NPN_RemoveProperty):

21538

* bindings/NP_jsobject.h:

21539

* bindings/c/c_instance.cpp:

21540

(CInstance::invokeMethod):

21541

* bindings/c/c_utility.cpp:

21542

(convertNPVariantToValue):

21543

* bindings/npruntime.cpp:

21544

(NPN_IdentifierIsString):

21545

(NPN_VariantIsVoid):

21546

(NPN_VariantIsNull):

21547

(NPN_VariantIsUndefined):

21548

(NPN_VariantIsBool):

21549

(NPN_VariantIsInt32):

21550

(NPN_VariantIsDouble):

21551

(NPN_VariantIsString):

21552

(NPN_VariantIsObject):

21553

(NPN_VariantToBool):

21554

(NPN_VariantToString):

21555

(NPN_VariantToInt32):

21556

(NPN_VariantToDouble):

21557

(NPN_VariantToObject):

21558

(NPN_InitializeVariantAsVoid):

21559

(NPN_InitializeVariantAsNull):

21560

(NPN_InitializeVariantAsUndefined):

21561

(NPN_InitializeVariantWithBool):

21562

(NPN_InitializeVariantWithInt32):

21563

(NPN_InitializeVariantWithDouble):

21564

(NPN_InitializeVariantWithString):

21565

(NPN_InitializeVariantWithStringCopy):

21566

(NPN_InitializeVariantWithObject):

21567

(NPN_InitializeVariantWithVariant):

21568

(NPN_ReleaseVariantValue):

21569

(NPN_CreateObject):

21570

* bindings/npruntime.h:

21571

(_NPString::):

21572

(_NPString::_NPVariant::):

21573

* bindings/npruntime_priv.h: Added.

21574

21575

2004-08-12 Darin Adler <darin@apple.com>

21576

21577

Reviewed by Adele.

21578

21579

- fixed 3 problems with parse functions that I just wrote, fixing 3 more Mozilla JavaScript tests

21580

21581

* kjs/function.cpp:

21582

(KJS::parseDigit): Fix typo, 'Z' instead of 'z', that prevented lowercase hex digits from working.

21583

(KJS::parseInt): Add octal support. Specification says it's optional, but I guess not.

21584

(KJS::parseFloat): Fix check for "0x" in parseFloat to return 0 rather than NaN. Also add code

21585

to skip leading "+" or "-".

21586

21587

=== Safari-156 ===

21588

21589

2004-08-12 Darin Adler <darin@apple.com>

21590

21591

Reviewed by Ken.

21592

21593

- fixed 43 Mozilla JavaScript tests

21594

21595

* kjs/date_object.h: Change parseDate and timeClip to take and return doubles.

21596

* kjs/date_object.cpp:

21597

(DateObjectImp::construct): Change to use a timeClip function that takes and returns a double rather

21598

than constructing a number object to pass to it.

21599

(DateObjectFuncImp::call): Change to use a parseDate function that returns a double.

21600

(KJS::parseDate): Change to return a double instead of creating the Number object here.

21601

(KJS::timeClip): Implement this as specified in the language standard.

21602

21603

* kjs/error_object.cpp: (NativeErrorImp::NativeErrorImp): Set the DontDelete, ReadOnly, and DontEnum

21604

flags on the prototype property.

21605

21606

* kjs/function.cpp:

21607

(KJS::FunctionImp::get): Return null rather than undefined for arguments when the function is not

21608

currently in scope.

21609

(KJS::isStrWhiteSpace): Added. Matches specification for StrWhiteSpace. Could move it to some utility

21610

file later.

21611

(KJS::parseDigit): Added. Helper function for parseInt.

21612

(KJS::parseInt): Added. Integer parser that puts result in a double so we're not limited to what

21613

strtoll can handle. Also matches standard more closely.

21614

(KJS::parseFloat): Added. Handles "0x" properly and passes flag to make empty string turn into NaN

21615

instead of 0.

21616

(KJS::GlobalFuncImp::call): Use the new parseInt and parseFloat.

21617

21618

* kjs/function_object.cpp: (FunctionPrototypeImp::FunctionPrototypeImp): Add a length property.

21619

21620

* kjs/lexer.h: Added error flag and sawError() function for detecting errors.

21621

* kjs/lexer.cpp:

21622

(Lexer::setCode): Clear error state.

21623

(Lexer::lex): Set error state if the lexer encounters an error

21624

21625

* kjs/internal.cpp:

21626

(NumberImp::toString): Roll in change from KDE version to special case 0 so we handle -0 correctly.

21627

(Parser::parse): Use new lexer error method so those errors are treated like parser errors.

21628

21629

* kjs/math_object.cpp: (MathFuncImp::call): Change min and max to treat -0 as less than +0.

21630

Change round to round values between -0.5 and -0 to -0 instead of +0.

21631

21632

* kjs/nodes.h: Add evaluateReference function to GroupNode.

21633

* kjs/nodes.cpp: (GroupNode::evaluateReference): Pass references through groups (parenthesized

21634

expressions) properly so that expressions like "delete (x.y)" work. Before, the parentheses

21635

would change x.y into a value that can't be deleted as a side effect.

21636

21637

* kjs/string_object.cpp: Change parameter count for indexOf and lastIndexOf from 2 to 1 to match

21638

the specification.

21639

21640

* kjs/testkjs.cpp: Rolled in changes from KDE to add a "quit" function to the test tool and

21641

get rid of the fixed size limit for code.

21642

21643

* kjs/ustring.cpp: (KJS::UString::substr): Added optimized case for substr(0, length) so it just

21644

returns the string without creating a new Rep, since I'm using substr in a place where it will

21645

often be passed a 0.

21646

21647

* tests/mozilla/ecma/String/15.5.4.11-1.js: Fixed one wrong entry in the Unicode table I added to

21648

the other day that was making a couple tests fail.

21649

* tests/mozilla/ecma/String/15.5.4.12-1.js: Ditto.

21650

* tests/mozilla/ecma/String/15.5.4.12-2.js: Ditto.

21651

* tests/mozilla/ecma/String/15.5.4.12-3.js: Ditto.

21652

* tests/mozilla/ecma/String/15.5.4.12-4.js: Ditto.

21653

* tests/mozilla/ecma/String/15.5.4.12-5.js: Ditto.

21654

21655

* kjs/string_object.lut.h: Regenerated.

21656

21657

2004-08-11 Darin Adler <darin@apple.com>

21658

21659

- fixed a tiny problem with the UTF-16 PCRE check-in

21660

21661

* pcre/maketables.c: (pcre_maketables): Fix mistake in table-generating code that sometimes caused

21662

the ctype_meta flag to get set in items that should not have it.

21663

21664

* pcre/chartables.c: Regenerated.

21665

21666

2004-08-10 Richard Williamson <rjw@apple.com>

21667

21668

Fixed <rdar://problem/3674747> Need to implement invokeUndefinedMethodFromWebScript:withArguments:

21669

21670

The following WebScripting methods are now supported on bound

21671

objects:

21672

21673

- (id)invokeUndefinedMethodFromWebScript:(NSString *)name withArguments:(NSArray *)args;

21674

- (void)setValue:(id)value forUndefinedKey:(NSString *)key

21675

- (id)valueForUndefinedKey:(NSString *)key

21676

21677

Reviewed by Chris.

21678

21679

* bindings/c/c_class.cpp:

21680

(CClass::fieldNamed):

21681

* bindings/c/c_class.h:

21682

* bindings/jni/jni_class.cpp:

21683

(JavaClass::fieldNamed):

21684

* bindings/jni/jni_class.h:

21685

* bindings/objc/objc_class.h:

21686

(KJS::Bindings::ObjcClass::isa):

21687

* bindings/objc/objc_class.mm:

21688

(ObjcClass::methodsNamed):

21689

(ObjcClass::fieldNamed):

21690

(ObjcClass::fallbackObject):

21691

* bindings/objc/objc_instance.h:

21692

* bindings/objc/objc_instance.mm:

21693

(ObjcInstance::invokeMethod):

21694

(ObjcInstance::setValueOfField):

21695

(ObjcInstance::setValueOfUndefinedField):

21696

(ObjcInstance::getValueOfField):

21697

(ObjcInstance::getValueOfUndefinedField):

21698

* bindings/objc/objc_runtime.h:

21699

(KJS::Bindings::ObjcField::~ObjcField):

21700

(KJS::Bindings::ObjcField::ObjcField):

21701

(KJS::Bindings::ObjcField::operator=):

21702

(KJS::Bindings::FallbackObjectImp::classInfo):

21703

* bindings/objc/objc_runtime.mm:

21704

(ObjcField::ObjcField):

21705

(ObjcField::name):

21706

(ObjcField::type):

21707

(ObjcField::valueFromInstance):

21708

(ObjcField::setValueToInstance):

21709

(FallbackObjectImp::FallbackObjectImp):

21710

(FallbackObjectImp::get):

21711

(FallbackObjectImp::put):

21712

(FallbackObjectImp::canPut):

21713

(FallbackObjectImp::implementsCall):

21714

(FallbackObjectImp::call):

21715

(FallbackObjectImp::hasProperty):

21716

(FallbackObjectImp::deleteProperty):

21717

(FallbackObjectImp::defaultValue):

21718

* bindings/runtime.h:

21719

(KJS::Bindings::Class::fallbackObject):

21720

(KJS::Bindings::Instance::getValueOfUndefinedField):

21721

(KJS::Bindings::Instance::setValueOfUndefinedField):

21722

* bindings/runtime_object.cpp:

21723

(RuntimeObjectImp::get):

21724

(RuntimeObjectImp::put):

21725

(RuntimeObjectImp::canPut):

21726

(RuntimeObjectImp::hasProperty):

21727

* bindings/testbindings.mm:

21728

(-[MyFirstInterface valueForUndefinedKey:]):

21729

(-[MyFirstInterface setValue:forUndefinedKey:]):

21730

21731

2004-08-10 Darin Adler <darin@apple.com>

21732

21733

Reviewed by Dave.

21734

21735

- switch PCRE to do UTF-16 directly instead of converting to/from UTF-8 for speed

21736

21737

* pcre/pcre.h: Added PCRE_UTF16 switch, set to 1. Added pcre_char typedef, which is char

21738

or uint16_t depending on the mode, and used appropriate in the 7 public functions

21739

that need to use it.

21740

* pcre/pcre.c: Add UTF-16 support to all functions.

21741

* pcre/study.c: Ditto.

21742

21743

* pcre/internal.h: Added ichar typedef, which is unsigned char or uint16_t depending on

21744

the mode. Changed declarations to use symbolic constants and typedefs so we size

21745

things to ichar when needed.

21746

21747

* pcre/maketables.c: (pcre_maketables): Change code to make tables that are

21748

sized to 16-bit characters instead of 8-bit.

21749

21750

* pcre/get.c:

21751

(pcre_copy_substring): Use pcre_char instead of char.

21752

(pcre_get_substring_list): Ditto.

21753

(pcre_free_substring_list): Ditto.

21754

(pcre_get_substring): Ditto.

21755

(pcre_free_substring): Ditto.

21756

21757

* pcre/dftables.c: (main): Used a bit more const, and use ICHAR sizes instead

21758

of hard-coding 8-bit table sizes.

21759

21760

* pcre/chartables.c: Regenerated.

21761

21762

* kjs/ustring.h: Remove functions that convert UTF-16 to/from UTF-8 offsets.

21763

* kjs/ustring.cpp: Change the shared empty string to have a unicode pointer that

21764

is not null. The null string still has a null pointer. This prevents us from

21765

passing a null through to the regular expression engine (which results in a null

21766

error even when the string length is 0).

21767

21768

* kjs/regexp.cpp:

21769

(KJS::RegExp::RegExp): Null-terminate the pattern and pass it.

21770

(KJS::RegExp::match): Use the 16-bit string directly, no need to convert to UTF-8.

21771

21772

2004-08-09 Darin Adler <darin@apple.com>

21773

21774

Reviewed by Maciej.

21775

21776

- fixed 28 Mozilla JavaScript tests

21777

21778

* kjs/array_object.cpp: (ArrayProtoFuncImp::call): Check for undefined rather than

21779

checking the number of arguments for the join method.

21780

21781

* kjs/lexer.cpp: (Lexer::lex): Parse hexadecimal and octal constants in doubles rather

21782

than integers, so we aren't limited to 32 bits.

21783

21784

* kjs/math_object.cpp: (MathFuncImp::call): Get rid of many unneeded special cases in

21785

the implementation of the pow operation. Also simplied a case that was handling positive

21786

and negative infinity separately.

21787

21788

* kjs/nodes.cpp: (ShiftNode::evaluate): Keep the result of shifts in a double instead of

21789

putting them in a long, so that unsigned shift will work properly.

21790

21791

* kjs/number_object.cpp: Add the DontDelete and ReadOnly flags to the numeric constants.

21792

21793

* kjs/operations.cpp:

21794

(KJS::isPosInf): Added an implementation inside APPLE_CHANGES that does not depend on the

21795

sign of isinf; our isinf function returns +1 even for negative infinity.

21796

(KJS::isNegInf): And again.

21797

(KJS::relation): Put in a nice simple implementation of comparison inside APPLE_CHANGES.

21798

Our floating point already handles the various infinity cases correctly.

21799

21800

* kjs/regexp_object.cpp:

21801

(RegExpProtoFuncImp::call): Add missing return before Null() in Exec method.

21802

(RegExpObjectImp::arrayOfMatches): Put undefined rather than an empty string into the

21803

array in cases where we did not match.

21804

(RegExpObjectImp::construct): Set the DontDelete, ReadOnly, and DontEnum flags for

21805

"global", "ignoreCase", "multiline", and "source".

21806

21807

* kjs/string_object.cpp: (StringProtoFuncImp::call): For the match method, turn a null

21808

string into undefined rather than an empty string. For the slice method, handle an

21809

undefined parameter for the limit properly as decribed in the specification, and add

21810

the limit to one case that didn't have the limit at all. For the methods that generate

21811

HTML strings, use lowercase tags instead of uppercase.

21812

21813

* kjs/ustring.cpp:

21814

(KJS::UChar::toLower): Use u_tolower from the ICU library.

21815

(KJS::UChar::toUpper): Use u_toupper from the ICU library.

21816

(KJS::UString::append): Fix some math that caused a buffer overflow.

21817

(KJS::convertUTF16OffsetsToUTF8Offsets): Ignore negative numbers (-1 is used as a special

21818

flag) rather than converting them all to 0.

21819

(KJS::convertUTF8OffsetsToUTF16Offsets): Ditto.

21820

21821

* tests/mozilla/jsDriver.pl: Fixed the relative links to point to our actual test files.

21822

21823

* tests/mozilla/ecma/String/15.5.4.11-1.js: Fixed the Unicode table in this test to match

21824

the Unicode specification in a few cases where it was wrong before.

21825

* tests/mozilla/ecma/String/15.5.4.11-2.js: Ditto.

21826

* tests/mozilla/ecma/String/15.5.4.11-3.js: Ditto.

21827

* tests/mozilla/ecma/String/15.5.4.11-5.js: Ditto.

21828

* tests/mozilla/ecma/String/15.5.4.11-6.js: Ditto.

21829

* tests/mozilla/ecma/String/15.5.4.12-1.js: Ditto.

21830

* tests/mozilla/ecma/String/15.5.4.12-2.js: Ditto.

21831

* tests/mozilla/ecma/String/15.5.4.12-3.js: Ditto.

21832

* tests/mozilla/ecma/String/15.5.4.12-4.js: Ditto.

21833

* tests/mozilla/ecma/String/15.5.4.12-5.js: Ditto.

21834

21835

* JavaScriptCore.pbproj/project.pbxproj: Link to libicu.

21836

21837

* kjs/number_object.lut.h: Regenerated.

21838

21839

2004-08-09 Darin Adler <darin@apple.com>

21840

21841

Reviewed by Maciej.

21842

21843

- fixed <rdar://problem/3753467> REGRESSION (137-138): reproducible buffer overrun in UString manipulation code

21844

21845

* kjs/ustring.cpp: (KJS::UString::append): Fix incorrect size computation. Without it

21846

we get a buffer overflow.

21847

21848

=== Safari-155 ===

21849

21850

2004-08-05 Richard Williamson <rjw@apple.com>

21851

21852

Fixed part of 3674747. The QT guys need this for feature freeze.

21853

21854

This patch implements support for the

21855

21856

- (id)invokeUndefinedMethodFromWebScript:(NSString *)name withArguments:(NSArray *)args

21857

21858

method of objects bound to JavaScript.

21859

21860

Reviewed by John.

21861

21862

* ChangeLog:

21863

* bindings/objc/objc_class.mm:

21864

(ObjcClass::methodsNamed):

21865

(ObjcClass::fieldNamed):

21866

* bindings/objc/objc_instance.mm:

21867

(ObjcInstance::invokeMethod):

21868

* bindings/objc/objc_runtime.h:

21869

(KJS::Bindings::ObjcMethod::~ObjcMethod):

21870

(KJS::Bindings::ObjcMethod::isFallbackMethod):

21871

(KJS::Bindings::ObjcMethod::javaScriptName):

21872

* bindings/objc/objc_runtime.mm:

21873

(ObjcMethod::ObjcMethod):

21874

(ObjcMethod::getMethodSignature):

21875

(ObjcMethod::setJavaScriptName):

21876

* bindings/testbindings.mm:

21877

21878

2004-08-04 Vicki Murley <vicki@apple.com>

21879

21880

Reviewed by mjs.

21881

21882

- fix <rdar://problem/3649789> SAP WebGUI has problems loading first page because of parse error

21883

21884

* kjs/lexer.cpp:

21885

(Lexer::lex): if the current character is a '\' and the next character is a line terminator,

21886

go to the next line and continue parsing the string (instead of failing). This matches

21887

behavior in Mac IE and Mozilla.

21888

21889

2004-08-03 Kevin Decker <kdecker@apple.com>

21890

21891

Reviewed by Darin.

21892

21893

Rolled in changes from the latest KJS sources that support additional

21894

Number.prototype functions.

21895

21896

Specifically this patch covers the follow parts of the ECMA 3 spec:

21897

15.7.4.5, 15.7.4.6, and 15.7.4.7

21898

21899

Fixes:

21900

<rdar://problem/3663716> missing Number.toFixed (and toPrecision, toExponential)

21901

<rdar://problem/3749492> missing Number.toPrecision prototype implementation

21902

<rdar://problem/3749591> missing Number.toExponential prototype implementation

21903

21904

* kjs/identifier.h: Added toFixed, toPrecision, and toExponential to the

21905

list of supported identifiers (a macro).

21906

* kjs/number_object.cpp: Implemented support for toFixed(), toPrecision(),

21907

and toExponential().

21908

(NumberPrototypeImp::NumberPrototypeImp):

21909

(NumberProtoFuncImp::call):

21910

* kjs/number_object.h: Added property names for toFixed, toPrecision,

21911

and toExponential.

21912

(KJS::NumberProtoFuncImp::):

21913

* tests/mozilla/expected.html: Update results.

21914

21915

2004-08-03 Darin Adler <darin@apple.com>

21916

21917

Reviewed by Ken.

21918

21919

- added support for copying RegExp objects so 7 more Mozilla regexp tests pass

21920

21921

* kjs/regexp_object.cpp: (RegExpObjectImp::construct): Check for case where

21922

we are supposed to just copy the regular expression object, and do so.

21923

Also tighten up arguments check to handle case where an actual "undefined"

21924

is passed rather than just omitting an argument.

21925

21926

* tests/mozilla/expected.html: Update results.

21927

21928

2004-08-02 Darin Adler <darin@apple.com>

21929

21930

* tests/mozilla/.cvsignore: Added.

21931

* tests/mozilla/expected.html: Update results.

21932

21933

2004-08-02 Darin Adler <darin@apple.com>

21934

21935

Reviewed by Ken.

21936

21937

- fixed RegExp.toString so 3 more Mozilla regexp tests pass

21938

21939

* kjs/regexp_object.cpp: (RegExpProtoFuncImp::call):

21940

Append the flags here so more tests paseed.

21941

21942

2004-08-02 Darin Adler <darin@apple.com>

21943

21944

Reviewed by Ken.

21945

21946

- fixed a couple things making 5 Mozilla regexp tests pass

21947

21948

* kjs/regexp_object.cpp: (RegExpProtoFuncImp::call): Implement toString

21949

for the prototype.

21950

(RegExpObjectImp::construct): Fix bug where the string "undefined" would

21951

be used as the flags string when no parameter was passed.

21952

21953

* kjs/regexp_object.h: (KJS::RegExpPrototypeImp::classInfo):

21954

Added a class info object for RegExp prototype so it can return

21955

a string instead of raising an exception when converting to a string.

21956

21957

* tests/mozilla/expected.html: Update results.

21958

21959

2004-08-02 Darin Adler <darin@apple.com>

21960

21961

Reviewed by Kevin.

21962

21963

- fix crashes in mozilla tests due to mishandling NaN

21964

21965

* kjs/array_object.cpp: (ArrayProtoFuncImp::call): Rerranged range checks after

21966

calls to toInteger so that NaN will get turned into something that fits in an integer.

21967

These were the ones John already fixed, but his fix used isnan and the new fix is

21968

more efficient.

21969

21970

* kjs/number_object.cpp: (NumberProtoFuncImp::call): Rearranged radix range checks

21971

after a call to toInteger to handle NaN properly. Also removed separate check

21972

for undefined that's not needed.

21973

21974

* kjs/string_object.cpp: (StringProtoFuncImp::call): More of the same kinds of changes

21975

as in the above two files, but for a lot more functions. Also changed one place with

21976

an explicit check for undefined to instead just check isNaN.

21977

21978

* tests/mozilla/run-mozilla-tests: Changed to invoke jst using $SYMROOTS for people

21979

like me who don't keep $SYMROOTS in their $PATH.

21980

21981

=== Safari-154 ===

21982

21983

=== Safari-153 ===

21984

21985

2004-07-26 Kevin Decker <kdecker@apple.com>

21986

21987

Changes done by Darin, reviewed by Kevin.

21988

21989

- changed testkjs to build in Xcode rather than from Makefile

21990

21991

* .cvsignore: Removed obsolete files from this list.

21992

* Makefile.am: Removed code to build testkjs; we do this in Xcode now.

21993

Changed to build target "All" rather than default target. This makes us

21994

build the testkjs test tool.

21995

* dummy.cpp: Removed.

21996

* kjs/.cvsignore: Removed obsolete files from this list, including

21997

the testkjs tool, which is now built in the symroots directory.

21998

* kjs/testkjs.cpp: Added copyright notice that was missing, since we have

21999

changed this file. Also this has the nice side effect of causing the tool

22000

to be rebuilt in the new location even if there are no other changes in

22001

your tree when you check this out.

22002

* tests/mozilla/run-mozilla-tests: Invoke perl explicitly so this works

22003

without setting the execute bit on jsDriver.pl.

22004

22005

2004-07-22 Kevin Decker <kdecker@apple.com>

22006

22007

Reviewed by Darin

22008

22009

Fixed <rdar://problem/3682340> (error console does not include source urls or line numbers of event exceptions).

22010

22011

* kjs/function_object.cpp:

22012

(FunctionObjectImp::construct):

22013

* kjs/function_object.h:

22014

* kjs/object.cpp:

22015

(KJS::ObjectImp::construct):

22016

* kjs/object.h:

22017

(KJS::Object::construct):

22018

22019

2004-07-21 Darin Adler <darin@apple.com>

22020

22021

* bindings/npruntime.h: Fixed typo.

22022

22023

2004-07-19 John Sullivan <sullivan@apple.com>

22024

22025

Reviewed by Maciej.

22026

22027

- bulletproofed array.slice() against NAN arguments. Harri noticed this

22028

vulnerability in my patch for 3714644

22029

22030

* kjs/array_object.cpp:

22031

(ArrayProtoFuncImp::call):

22032

handle NAN parameters passed to slice() by clamping to 0 and length.

22033

22034

2004-07-19 Richard Williamson <rjw@apple.com>

22035

22036

Fixed 3733349. Prevent Java applet callbacks into JavaScript after applet

22037

has been destroyed.

22038

22039

Reviewed by John.

22040

22041

* bindings/jni/jni_jsobject.cpp:

22042

(JSObject::invoke):

22043

(JSObject::JSObject):

22044

22045

2004-07-16 John Sullivan <sullivan@apple.com>

22046

22047

Reviewed by Maciej.

22048

22049

- fixed <rdar://problem/3714644> REGRESSION (125.8-146): bugzilla submit link

22050

hangs browser with javascript

22051

22052

* kjs/array_object.cpp:

22053

(ArrayProtoFuncImp::call):

22054

Check for undefined type for args[0] the same way we were already checking

22055

for args[1]. In this case, args was zero-length, but we were treating

22056

args[0] like an integer anyway. Resulted in some code looping from a NAN

22057

value to 4, taking approximately forever.

22058

22059

* JavaScriptCore.pbproj/project.pbxproj:

22060

version wars

22061

22062

=== Safari-152 ===

22063

22064

2004-07-14 Maciej Stachowiak <mjs@apple.com>

22065

22066

Reviewed by John.

22067

22068

<rdar://problem/3711474>: (REGRESSION (125-146): JavaScript 'toString(16)' is broken)

22069

<rdar://problem/3644873>: (REGRESSION (125-140u): secondary list doesn't fill in at Southwest.com)

22070

22071

* kjs/number_object.cpp:

22072

(NumberProtoFuncImp::call): Initialize radix from dradix, not from itself!

22073

22074

2004-07-13 Kevin Decker <kdecker@apple.com>

22075

22076

Reviewed by kocienda.

22077

22078

- made testkjs and JavaScriptCore a subtarget of 'All'

22079

- testkjs now builds in $SYMROOTS

22080

22081

* JavaScriptCore.pbproj/project.pbxproj:

22082

22083

=== Safari-151 ===

22084

22085

2004-06-24 Chris Blumenberg <cblu@apple.com>

22086

22087

Ignore .mode1 files in JavaScriptCore.pbproj

22088

22089

Reviewed by kocienda.

22090

22091

* JavaScriptCore.pbproj/.cvsignore:

22092

22093

2004-06-23 Richard Williamson <rjw@apple.com>

22094

22095

Implemented changes for latest npruntime.h.

22096

22097

Reviewed by Chris.

22098

22099

* JavaScriptCore.pbproj/project.pbxproj:

22100

* bindings/NP_jsobject.cpp:

22101

(listFromVariantArgs):

22102

(identiferFromNPIdentifier):

22103

(_NPN_CreateScriptObject):

22104

(NPN_Call):

22105

(NPN_Evaluate):

22106

(NPN_GetProperty):

22107

(NPN_SetProperty):

22108

(NPN_RemoveProperty):

22109

* bindings/NP_jsobject.h:

22110

* bindings/c/c_class.cpp:

22111

(CClass::methodsNamed):

22112

(CClass::fieldNamed):

22113

* bindings/c/c_instance.cpp:

22114

(CInstance::invokeMethod):

22115

* bindings/c/c_utility.cpp:

22116

(convertNPVariantToValue):

22117

* bindings/c/c_utility.h:

22118

* bindings/npruntime.cpp:

22119

(stringIdentifierEqual):

22120

(stringIdentifierHash):

22121

(getStringIdentifierDictionary):

22122

(intIdentifierEqual):

22123

(intIdentifierHash):

22124

(getIntIdentifierDictionary):

22125

(NPN_GetStringIdentifier):

22126

(NPN_GetStringIdentifiers):

22127

(NPN_GetIntIdentifier):

22128

(NPN_IdentifierIsString):

22129

(NPN_UTF8FromIdentifier):

22130

(NPN_VariantToInt32):

22131

(NPN_VariantToDouble):

22132

(NPN_SetException):

22133

* bindings/npruntime.h:

22134

* bindings/objc/WebScriptObject.mm:

22135

(+[WebScriptObject _convertValueToObjcValue:KJS::root:Bindings::]):

22136

* bindings/runtime_object.cpp:

22137

(RuntimeObjectImp::~RuntimeObjectImp):

22138

* bindings/runtime_root.cpp:

22139

(KJS::Bindings::rootForInterpreter):

22140

* bindings/testbindings.cpp:

22141

(initializeIdentifiers):

22142

(logMessage):

22143

(setDoubleValue):

22144

(setIntValue):

22145

(setBooleanValue):

22146

22147

=== JavaScriptCore-146.1 ===

22148

22149

2004-06-16 Richard Williamson <rjw@apple.com>

22150

22151

Fixed <rdar://problem/3702287> Crash returning nil from bound ObjC

22152

22153

This turned out to be a show stopper for Dashboard. Accessing a nil

22154

ObjC property from JS caused a crash. Similar to the problem

22155

3696112 fixed below.

22156

22157

Reviewed by Trey.

22158

22159

* bindings/objc/objc_runtime.mm:

22160

(KJS::Bindings::ObjcField::valueFromInstance):

22161

22162

=== Safari-146 ===

22163

22164

2004-06-16 Richard Williamson <rjw@apple.com>

22165

22166

Fixed <rdar://problem/3696112>: nil from an Objective-C class seems to get wrapped as a JavaScript proxy that will not print.

22167

22168

This turned out to be a show stopper for Dashboard. We now

22169

return Undefined() when nil is returned from a ObjC method

22170

that returns an object type.

22171

22172

Reviewed by Maciej.

22173

22174

* bindings/objc/objc_utility.mm:

22175

(KJS::Bindings::convertObjcValueToValue):

22176

22177

=== Safari-145 ===

22178

22179

2004-06-15 Richard Williamson <rjw@apple.com>

22180

22181

Fixed <rdar://problem/3695875>: Objective-C instances that are exported to JavaScript are too promiscuous

22182

22183

No longer need to check respondsToSelector: for

22184

isSelectorExcludedFromWebScript: and isKeyExcludedFromWebScript:

22185

because these now have a default implementation on NSObject.

22186

22187

Reviewed by Trey.

22188

22189

* bindings/objc/objc_class.mm:

22190

(ObjcClass::methodsNamed):

22191

(ObjcClass::fieldNamed):

22192

22193

2004-06-14 Darin Adler <darin@apple.com>

22194

22195

Reviewed by Maciej.

22196

22197

- fixed some things for GC that Patrick missed, or that happened after the branch

22198

22199

* bindings/objc/WebScriptObject.mm:

22200

(-[WebScriptObject dealloc]): Moved removeNativeReference call here from private object.

22201

(-[WebScriptObject finalize]): Added.

22202

22203

- added some missing nil checks

22204

22205

* bindings/objc/objc_instance.mm:

22206

(ObjcInstance::ObjcInstance): Check for nil.

22207

(ObjcInstance::~ObjcInstance): Check for nil.

22208

(ObjcInstance::operator=): Check for nil.

22209

22210

2004-06-14 Darin Adler <darin@apple.com>

22211

22212

Reviewed by me, code changes by Patrick Beard.

22213

22214

- fixed <rdar://problem/3671507>: (WebKit should adopt GC changes and compile with GC enabled)

22215

22216

* bindings/objc/objc_instance.mm:

22217

(ObjcInstance::ObjcInstance): Use CFRetain instead of retain.

22218

(ObjcInstance::~ObjcInstance): Use CFRelease instead of release.

22219

(ObjcInstance::operator=): More of the same.

22220

(ObjcInstance::end): Use [pool drain] if compiling on Tiger.

22221

22222

* bindings/objc/objc_runtime.mm:

22223

(ObjcArray::ObjcArray): Use CFRetain instead of retain.

22224

(ObjcArray::~ObjcArray): Use CFRelease instead of release.

22225

(ObjcArray::operator=): More of the same.

22226

22227

* bindings/testbindings.mm: Fixed incorrect license.

22228

(main): Use [pool drain] if compiling on Tiger.

22229

22230

=== Safari-144 ===

22231

22232

2004-06-10 Kevin Decker <kdecker@apple.com>

22233

22234

Reviewed by John.

22235

22236

* kjs/lexer.cpp:

22237

(Lexer::setCode):

22238

- fixed <rdar://problem/3682398>: (error console line numbers are offset by 1)

22239

* kjs/lexer.h:

22240

(KJS::Lexer::lineNo):

22241

- fixed <rdar://problem/3682398>: (error console line numbers are offset by 1)

22242

22243

=== JavaScriptCore-143.2 ===

22244

22245

2004-06-07 Darin Adler <darin@apple.com>

22246

22247

- fixed <rdar://problem/3682489>: (JavaScriptGlue no longer compiles because Interpreter::evaluate parameters changed)

22248

22249

* kjs/interpreter.h: Added an overload to make JavaScriptGlue compile.

22250

* kjs/interpreter.cpp: (KJS::Interpreter::evaluate): Implemented the overload.

22251

22252

=== JavaScriptCore-143.1 ===

22253

22254

2004-06-04 Kevin Decker <kdecker@apple.com>

22255

22256

Reviewed by Darin

22257

22258

- fixed <rdar://problem/3680594>

22259

22260

* kjs/object.cpp:

22261

(KJS::Error::create):

22262

22263

=== Safari-143 ===

22264

22265

2004-06-04 Darin Adler <darin@apple.com>

22266

22267

* kjs/testkjs.cpp: (main): Fix build breakage by adding URL and line number parameters.

22268

22269

2004-06-04 Kevin Decker <kdecker@apple.com>

22270

22271

Reviewed by Dave.

22272

22273

- ObjC bindings do not (yet) pass along sourceurl or line numbers

22274

- we don't have a way as of yet to accomidate line numbers and urls for dynamic javascript

22275

- changed the wording of an error message

22276

- the lexer, parser, and interpreter have been made "sourceURL aware"

22277

- stored the url into Error

22278

22279

* bindings/NP_jsobject.cpp:

22280

(NPN_Evaluate):

22281

* bindings/jni/jni_jsobject.cpp:

22282

(JSObject::eval):

22283

* bindings/objc/WebScriptObject.mm:

22284

(-[WebScriptObject evaluateWebScript:]):

22285

* kjs/function.cpp:

22286

(GlobalFuncImp::call):

22287

* kjs/function_object.cpp:

22288

(FunctionObjectImp::construct):

22289

* kjs/internal.cpp:

22290

(Parser::parse):

22291

(InterpreterImp::checkSyntax):

22292

(InterpreterImp::evaluate):

22293

* kjs/internal.h:

22294

* kjs/interpreter.cpp:

22295

(Interpreter::evaluate):

22296

* kjs/interpreter.h:

22297

* kjs/lexer.cpp:

22298

(Lexer::setCode):

22299

* kjs/lexer.h:

22300

(KJS::Lexer::sourceURL):

22301

* kjs/nodes.cpp:

22302

(Node::Node):

22303

(Node::throwError):

22304

(FunctionCallNode::evaluate):

22305

* kjs/nodes.h:

22306

* kjs/object.cpp:

22307

(KJS::Error::create):

22308

* kjs/object.h:

22309

22310

2004-06-04 Richard Williamson <rjw@apple.com>

22311

22312

Fixed crash when attempting to access properties on nil

22313

object.

22314

22315

Reviewed by John.

22316

22317

* bindings/objc/objc_instance.mm:

22318

(ObjcInstance::getClass):

22319

* bindings/runtime_object.cpp:

22320

(RuntimeObjectImp::get):

22321

* bindings/testM.js:

22322

* bindings/testbindings.mm:

22323

(-[MyFirstInterface getString]):

22324

22325

2004-05-27 Kevin Decker <kdecker@apple.com>

22326

22327

Reviewed by Ken.

22328

22329

-revised generated error message content

22330

22331

* kjs/error_object.cpp:

22332

(ErrorProtoFuncImp::call):

22333

* kjs/internal.cpp:

22334

(Parser::parse):

22335

* kjs/object.cpp:

22336

(KJS::Error::create):

22337

22338

=== Safari-142 ===

22339

22340

2004-05-27 Richard Williamson <rjw@apple.com>

22341

22342

Renamed WebScriptMethods to WebScripting based on feedback from Nancy.

22343

22344

Reviewed by Chris.

22345

22346

* bindings/objc/WebScriptObject.h:

22347

22348

2004-05-27 Darin Adler <darin@apple.com>

22349

22350

Reviewed by Maciej.

22351

22352

- moved to new symlink technique for embedding frameworks

22353

22354

* JavaScriptCore.pbproj/project.pbxproj: Get rid of embed-frameworks build step

22355

because we don't need it any more.

22356

22357

2004-05-24 Richard Williamson <rjw@apple.com>

22358

22359

Changed RuntimeArrayImp to inherit from ArrayInstanceImp and

22360

fixed ClassInfo to correctly reflect inheritance. This is required

22361

because of the runtime checks in JSC for arrays, i.e. in

22362

the Function objects apply method.

22363

22364

Reviewed by Ken.

22365

22366

* bindings/jni/jni_runtime.cpp:

22367

(JavaArray::convertJObjectToArray):

22368

* bindings/objc/objc_utility.mm:

22369

(KJS::Bindings::convertObjcValueToValue):

22370

* bindings/runtime_array.cpp:

22371

(RuntimeArrayImp::RuntimeArrayImp):

22372

* bindings/runtime_array.h:

22373

* bindings/testM.js: Added.

22374

* bindings/testbindings.mm:

22375

(+[MyFirstInterface webScriptNameForSelector:]):

22376

(-[MyFirstInterface logMessages:]):

22377

(-[MyFirstInterface logMessage:prefix:]):

22378

(-[MyFirstInterface callJSObject::]):

22379

22380

2004-05-22 Darin Adler <darin@apple.com>

22381

22382

Reviewed by Maciej.

22383

22384

- fixed <rdar://problem/3664260>: (JS needs to listen to timezone change notifications)

22385

22386

* kjs/date_object.cpp: (CopyLocalTimeZone): As per Chris Kane and Jordan Hubbard, use <notify.h>

22387

with a hardcoded string of "com.apple.system.timezone", and do CFTimeZoneResetSystem since

22388

CoreFoundation doesn't do this itself. Turns out this affects the default time zone as long as

22389

it hasn't been set explicitly.

22390

22391

=== Safari-141 ===

22392

22393

2004-05-20 Richard Williamson <rjw@apple.com>

22394

22395

Implemented WebScriptObject/DOM wrapper voodoo. DOM wrappers

22396

can now be referenced like any other WebScriptObject, meaning

22397

you can do JS operations on them.

22398

22399

All added implementation of finalizeForWebScript.

22400

22401

Reviewed by Ken.

22402

22403

* bindings/objc/WebScriptObject.h:

22404

* bindings/objc/WebScriptObject.mm:

22405

(-[WebScriptObject _initializeWithObjectImp:KJS::root:Bindings::]):

22406

(-[WebScriptObject _initWithObjectImp:KJS::root:Bindings::]):

22407

(-[WebScriptObject KJS::]):

22408

(-[WebScriptObject dealloc]):

22409

(-[WebScriptObject callWebScriptMethod:withArguments:]):

22410

(-[WebScriptObject evaluateWebScript:]):

22411

(-[WebScriptObject setValue:forKey:]):

22412

(-[WebScriptObject valueForKey:]):

22413

(-[WebScriptObject stringRepresentation]):

22414

* bindings/objc/WebScriptObjectPrivate.h:

22415

* bindings/objc/objc_instance.mm:

22416

(ObjcInstance::~ObjcInstance):

22417

22418

2004-05-19 Richard Williamson <rjw@apple.com>

22419

22420

Removed extraneous tabs that were added (by XCode?).

22421

22422

* bindings/objc/WebScriptObject.h:

22423

22424

2004-05-19 Darin Adler <darin@apple.com>

22425

22426

- fixed headers with licenses mangled by Xcode auto-indenting

22427

22428

* bindings/jni/jni_jsobject.cpp:

22429

* bindings/jni/jni_jsobject.h:

22430

* bindings/runtime_array.h:

22431

* bindings/runtime_root.cpp:

22432

* bindings/runtime_root.h:

22433

22434

2004-05-18 Richard Williamson <rjw@apple.com>

22435

22436

Added exception logging. Also check for exception and

22437

set results as appropriate.

22438

22439

Reviewed by Maciej (partially reviewed).

22440

22441

* bindings/objc/WebScriptObject.mm:

22442

(-[WebScriptObject callWebScriptMethod:withArguments:]):

22443

(-[WebScriptObject evaluateWebScript:]):

22444

(-[WebScriptObject setValue:forKey:]):

22445

(-[WebScriptObject valueForKey:]):

22446

22447

2004-05-18 Richard Williamson <rjw@apple.com>

22448

22449

Finsished implementing support for windowScriptObject.

22450

Had to make WebScriptObjectPrivate.h accessible from

22451

WebCore.

22452

22453

Reviewed by Maciej.

22454

22455

* JavaScriptCore.pbproj/project.pbxproj:

22456

* bindings/objc/WebScriptObjectPrivate.h:

22457

22458

2004-05-18 Richard Williamson <rjw@apple.com>

22459

22460

Use KVC to set/get values instead of directly accessing

22461

ivars.

22462

22463

Reviewed by Maciej.

22464

22465

* bindings/objc/WebScriptObject.mm:

22466

(-[WebScriptObject callWebScriptMethod:withArguments:]):

22467

(+[WebScriptObject _convertValueToObjcValue:KJS::root:Bindings::]):

22468

* bindings/objc/objc_runtime.mm:

22469

(ObjcField::valueFromInstance):

22470

(convertValueToObjcObject):

22471

(ObjcField::setValueToInstance):

22472

22473

2004-05-17 Richard Williamson <rjw@apple.com>

22474

22475

Implemented new API for WebScriptObject.

22476

22477

Fixed <rdar://problem/3657145>: (objc to javascript method calls do not cause updates.)

22478

Fixed <rdar://problem/3654887>: (Update to JSC to refer to new JSObject LiveConnect object) (w/ help from Vicki)

22479

22480

Reviewed by Hyatt.

22481

22482

* JavaScriptCore.pbproj/project.pbxproj:

22483

* bindings/c/c_instance.cpp:

22484

(CInstance::invokeMethod):

22485

* bindings/jni/jni_instance.cpp:

22486

(JavaInstance::invokeMethod):

22487

* bindings/jni/jni_jsobject.cpp:

22488

(JSObject::convertValueToJObject):

22489

* bindings/jni/jni_utility.cpp:

22490

(KJS::Bindings::getJNIField):

22491

* bindings/objc/WebScriptObject.mm:

22492

(_didExecute):

22493

(-[WebScriptObject _initWithObjectImp:KJS::root:Bindings::]):

22494

(-[WebScriptObject KJS::]):

22495

(-[WebScriptObject dealloc]):

22496

(+[WebScriptObject throwException:]):

22497

(listFromNSArray):

22498

(-[WebScriptObject callWebScriptMethod:withArguments:]):

22499

(-[WebScriptObject evaluateWebScript:]):

22500

(-[WebScriptObject setValue:forKey:]):

22501

(-[WebScriptObject valueForKey:]):

22502

(-[WebScriptObject stringRepresentation]):

22503

(+[WebScriptObject _convertValueToObjcValue:KJS::root:Bindings::]):

22504

(+[WebUndefined undefined]):

22505

(-[WebUndefined initWithCoder:]):

22506

(-[WebUndefined encodeWithCoder:]):

22507

(-[WebUndefined copyWithZone:]):

22508

(-[WebUndefined retain]):

22509

(-[WebUndefined release]):

22510

(-[WebUndefined retainCount]):

22511

(-[WebUndefined autorelease]):

22512

(-[WebUndefined dealloc]):

22513

(-[WebUndefined copy]):

22514

(-[WebUndefined replacementObjectForPortCoder:]):

22515

* bindings/objc/WebScriptObjectPrivate.h: Added.

22516

* bindings/objc/objc_class.mm:

22517

(ObjcClass::methodsNamed):

22518

(ObjcClass::fieldNamed):

22519

* bindings/objc/objc_instance.mm:

22520

(ObjcInstance::invokeMethod):

22521

* bindings/objc/objc_jsobject.h:

22522

* bindings/objc/objc_jsobject.mm:

22523

* bindings/objc/objc_runtime.mm:

22524

(ObjcField::valueFromInstance):

22525

* bindings/objc/objc_utility.mm:

22526

(KJS::Bindings::JSMethodNameToObjCMethodName):

22527

(KJS::Bindings::convertValueToObjcValue):

22528

(KJS::Bindings::convertObjcValueToValue):

22529

* bindings/runtime.cpp:

22530

(Instance::setDidExecuteFunction):

22531

(Instance::didExecuteFunction):

22532

(Instance::setValueOfField):

22533

* bindings/runtime.h:

22534

* bindings/testbindings.mm:

22535

(+[MyFirstInterface webScriptNameForSelector:]):

22536

(-[MyFirstInterface callJSObject::]):

22537

22538

2004-05-14 Vicki Murley <vicki@apple.com>

22539

22540

Reviewed by mjs.

22541

22542

<rdar://problem/3642427>: framework marketing number should be 2.0 for DoubleBarrel release

22543

22544

* JavaScriptCore.pbproj/project.pbxproj: change CFBundleShortVersionString to 2.0

22545

22546

=== Safari-140 ===

22547

22548

2004-05-13 Richard Williamson <rjw@apple.com>

22549

22550

Fixed indentation.

22551

22552

Reviewed by Chris.

22553

22554

* ChangeLog:

22555

* bindings/objc/WebScriptObject.h:

22556

22557

2004-05-13 Richard Williamson <rjw@apple.com>

22558

22559

Approved API changes. Currently unimplemented.

22560

22561

Reviewed by Chris.

22562

22563

* ChangeLog:

22564

* JavaScriptCore.pbproj/project.pbxproj:

22565

* bindings/objc/WebScriptObject.h: Added.

22566

* bindings/objc/WebScriptObject.mm: Added.

22567

(+[WebScriptObject throwException:]):

22568

(-[WebScriptObject callWebScriptMethod:withArguments:]):

22569

(-[WebScriptObject evaluateWebScript:]):

22570

(-[WebScriptObject stringRepresentation]):

22571

(+[WebUndefined undefined]):

22572

(-[WebUndefined initWithCoder:]):

22573

(-[WebUndefined encodeWithCoder:]):

22574

(-[WebUndefined copyWithZone:]):

22575

22576

2004-05-07 Vicki Murley <vicki@apple.com>

22577

22578

Reviewed by darin.

22579

22580

Turn off GC since it uses ppc only instructions (which breaks

22581

the B&I build).

22582

22583

* kjs/value.h: set USE_CONSERVATIVE_GC to 0

22584

22585

=== Safari-139 ===

22586

22587

2004-05-07 Maciej Stachowiak <mjs@apple.com>

22588

22589

Reviewed by Darin.

22590

22591

- add -funroll-loops=16 compiler option for approx .5% speedup on

22592

HTML iBench and .5-1% speedup on JS iBench.

22593

22594

* JavaScriptCore.pbproj/project.pbxproj:

22595

22596

2004-04-25 Maciej Stachowiak <mjs@apple.com>

22597

22598

Reviewed by Darin.

22599

22600

Enable full conservative GC mode in addition to test mode. When

22601

conservative GC is enabled, we now get an 11% speed improvement on

22602

the iBench. Also fix some spots I missed before.

22603

22604

Specific noteworth changes:

22605

22606

* kjs/collector.cpp:

22607

(KJS::Collector::markStackObjectsConservatively): Check possible

22608

cell pointers for 8-byte aligment and verify they are not 0.

22609

22610

* kjs/protected_values.cpp:

22611

(KJS::ProtectedValues::increaseProtectCount): Move null-tolerance from here...

22612

(KJS::ProtectedValues::decreaseProtectCount): ...and here...

22613

* kjs/protect.h:

22614

(KJS::gcProtectNullTolerant): ...to here...

22615

(KJS::gcUnprotectNullTolerant): ...and here, because not all callers need the null

22616

tolerance, and doing the check is expensive.

22617

22618

* kjs/protected_values.cpp:

22619

(KJS::ProtectedValues::computeHash): Replace hash function with a much faster one

22620

that is still very good.

22621

22622

* kjs/protect.h:

22623

(KJS::gcProtect):

22624

(KJS::gcUnprotect):

22625

(KJS::ProtectedValue::ProtectedValue):

22626

(KJS::ProtectedValue::~ProtectedValue):

22627

(KJS::ProtectedValue::operator=):

22628

(KJS::ProtectedObject::ProtectedObject):

22629

(KJS::ProtectedObject::~ProtectedObject):

22630

(KJS::ProtectedObject::operator=):

22631

(KJS::ProtectedReference::ProtectedReference):

22632

(KJS::ProtectedReference::~ProtectedReference):

22633

(KJS::ProtectedReference::operator=):

22634

* kjs/protected_values.cpp:

22635

(KJS::ProtectedValues::getProtectCount):

22636

(KJS::ProtectedValues::increaseProtectCount):

22637

(KJS::ProtectedValues::decreaseProtectCount):

22638

(KJS::ProtectedValues::computeHash):

22639

* bindings/runtime_root.cpp:

22640

(KJS::Bindings::addNativeReference):

22641

(KJS::Bindings::removeNativeReference):

22642

(RootObject::removeAllNativeReferences):

22643

* bindings/runtime_root.h:

22644

(KJS::Bindings::RootObject::~RootObject):

22645

(KJS::Bindings::RootObject::setRootObjectImp):

22646

* kjs/collector.cpp:

22647

(KJS::Collector::allocate):

22648

(KJS::Collector::collect):

22649

* kjs/collector.h:

22650

* kjs/internal.cpp:

22651

(NumberImp::create):

22652

(InterpreterImp::globalInit):

22653

(InterpreterImp::globalClear):

22654

(InterpreterImp::mark):

22655

* kjs/list.cpp:

22656

(KJS::List::derefValues):

22657

(KJS::List::refValues):

22658

(KJS::List::append):

22659

* kjs/object.cpp:

22660

(KJS::ObjectImp::setInternalValue):

22661

(KJS::ObjectImp::putDirect):

22662

* kjs/value.cpp:

22663

(ValueImp::mark):

22664

(ValueImp::marked):

22665

* kjs/value.h:

22666

(KJS::ValueImp::ValueImp):

22667

(KJS::ValueImp::~ValueImp):

22668

(KJS::ValueImp::):

22669

(KJS::Value::Value):

22670

(KJS::Value::~Value):

22671

(KJS::Value::operator=):

22672

22673

2004-04-30 Richard Williamson <rjw@apple.com>

22674

22675

Asking an NSInvocation for it's return value when return type

22676

is void throws an exception. Added check for void return types

22677

to avoid this exception.

22678

22679

Reviewed by Ken.

22680

22681

* bindings/objc/objc_instance.mm:

22682

(ObjcInstance::invokeMethod):

22683

22684

2004-04-29 Richard Williamson <rjw@apple.com>

22685

22686

Fixed several bad problems with the ObjC bindings. In particular, conversion

22687

to/from JavaScriptObject (soon to be WebScriptObject) was completely broken.

22688

22689

Reviewed by Chris.

22690

22691

* bindings/objc/objc_jsobject.h:

22692

* bindings/objc/objc_jsobject.mm:

22693

(-[JavaScriptObject initWithObjectImp:KJS::root:Bindings::]):

22694

(-[JavaScriptObject KJS::]):

22695

(+[JavaScriptObject _convertValueToObjcValue:KJS::root:Bindings::]):

22696

(-[JavaScriptObject call:arguments:]):

22697

(-[JavaScriptObject evaluate:]):

22698

(-[JavaScriptObject getMember:]):

22699

(-[JavaScriptObject getSlot:]):

22700

* bindings/objc/objc_runtime.mm:

22701

(ObjcField::valueFromInstance):

22702

(ObjcField::setValueToInstance):

22703

* bindings/objc/objc_utility.mm:

22704

(KJS::Bindings::convertValueToObjcValue):

22705

(KJS::Bindings::convertObjcValueToValue):

22706

* bindings/runtime.h:

22707

* bindings/runtime_root.cpp:

22708

(KJS::Bindings::rootForInterpreter):

22709

(KJS::Bindings::addNativeReference):

22710

(KJS::Bindings::removeNativeReference):

22711

* bindings/runtime_root.h:

22712

* bindings/testbindings.mm:

22713

(-[MyFirstInterface logMessage:]):

22714

(-[MyFirstInterface setJSObject:]):

22715

(-[MyFirstInterface callJSObject::]):

22716

22717

2004-04-24 Darin Adler <darin@apple.com>

22718

22719

Reviewed by Dave.

22720

22721

* kjs/ustring.cpp: (KJS::UString::append): Fix one case that was allocating a buffer

22722

that is 2x too big.

22723

22724

2004-04-23 Maciej Stachowiak <mjs@apple.com>

22725

22726

Reviewed by Darin.

22727

22728

Implementation of conservative GC, based partly on code from

22729

Darin. It's turned off for now, so it shouldn't have any effect on

22730

the normal build.

22731

22732

* JavaScriptCore.pbproj/project.pbxproj:

22733

* kjs/collector.cpp:

22734

(KJS::Collector::markStackObjectsConservatively):

22735

(KJS::Collector::markProtectedObjects):

22736

(KJS::Collector::collect):

22737

* kjs/collector.h:

22738

* kjs/protect.h:

22739

(KJS::gcProtect):

22740

(KJS::gcUnprotect):

22741

* kjs/protected_values.cpp: Added.

22742

(KJS::ProtectedValues::getProtectCount):

22743

(KJS::ProtectedValues::increaseProtectCount):

22744

(KJS::ProtectedValues::insert):

22745

(KJS::ProtectedValues::decreaseProtectCount):

22746

(KJS::ProtectedValues::expand):

22747

(KJS::ProtectedValues::shrink):

22748

(KJS::ProtectedValues::rehash):

22749

(KJS::ProtectedValues::computeHash):

22750

* kjs/protected_values.h: Added.

22751

* kjs/value.cpp:

22752

(ValueImp::useConservativeMark):

22753

(ValueImp::mark):

22754

(ValueImp::marked):

22755

* kjs/value.h:

22756

(KJS::ValueImp::):

22757

22758

=== Safari-138 ===

22759

22760

2004-04-22 Richard Williamson <rjw@apple.com>

22761

22762

Fixed build snafu (re-declaration of NPBool in npruntime.h and

22763

npapi.h).

22764

22765

* bindings/npruntime.h:

22766

22767

2004-04-22 Richard Williamson <rjw@apple.com>

22768

22769

Updated plugin binding API to reflect latest revision from

22770

working group.

22771

22772

Biggest change is the introduction of NPVariant used to represent

22773

value types. NPVariant replaces the use of NPObject for the

22774

exchange of values between scripting environment and native code.

22775

22776

Reviewed by John.

22777

22778

* JavaScriptCore.pbproj/project.pbxproj:

22779

* bindings/NP_jsobject.cpp:

22780

(identiferFromNPIdentifier):

22781

(NPN_Call):

22782

(NPN_Evaluate):

22783

(NPN_GetProperty):

22784

(NPN_SetProperty):

22785

(NPN_ToString):

22786

(NPN_GetPropertyAtIndex):

22787

(NPN_SetPropertyAtIndex):

22788

* bindings/c/c_class.cpp:

22789

(CClass::methodsNamed):

22790

(CClass::fieldNamed):

22791

* bindings/c/c_instance.cpp:

22792

(CInstance::invokeMethod):

22793

(CInstance::defaultValue):

22794

* bindings/c/c_runtime.cpp:

22795

(CField::valueFromInstance):

22796

(CField::setValueToInstance):

22797

* bindings/c/c_utility.cpp:

22798

(convertNPStringToUTF16):

22799

(convertUTF8ToUTF16):

22800

(coerceValueToNPVariantStringType):

22801

(convertValueToNPVariant):

22802

(convertNPVariantToValue):

22803

* bindings/c/c_utility.h:

22804

* bindings/npruntime.cpp:

22805

(NPN_GetIdentifier):

22806

(NPN_GetIdentifiers):

22807

(NPN_UTF8FromIdentifier):

22808

(NPN_VariantIsVoid):

22809

(NPN_VariantIsNull):

22810

(NPN_VariantIsUndefined):

22811

(NPN_VariantIsBool):

22812

(NPN_VariantIsInt32):

22813

(NPN_VariantIsDouble):

22814

(NPN_VariantIsString):

22815

(NPN_VariantIsObject):

22816

(NPN_VariantToBool):

22817

(NPN_VariantToString):

22818

(NPN_VariantToInt32):

22819

(NPN_VariantToDouble):

22820

(NPN_VariantToObject):

22821

(NPN_InitializeVariantAsVoid):

22822

(NPN_InitializeVariantAsNull):

22823

(NPN_InitializeVariantAsUndefined):

22824

(NPN_InitializeVariantWithBool):

22825

(NPN_InitializeVariantWithInt32):

22826

(NPN_InitializeVariantWithDouble):

22827

(NPN_InitializeVariantWithString):

22828

(NPN_InitializeVariantWithStringCopy):

22829

(NPN_InitializeVariantWithObject):

22830

(NPN_InitializeVariantWithVariant):

22831

(NPN_ReleaseVariantValue):

22832

(NPN_CreateObject):

22833

(NPN_RetainObject):

22834

(NPN_ReleaseObject):

22835

(NPN_IsKindOfClass):

22836

(NPN_SetExceptionWithUTF8):

22837

(NPN_SetException):

22838

* bindings/npruntime.h:

22839

(_NPString::):

22840

(_NPString::_NPVariant::):

22841

* bindings/testbindings.cpp:

22842

(logMessage):

22843

(setDoubleValue):

22844

(setIntValue):

22845

(setStringValue):

22846

(setBooleanValue):

22847

(getDoubleValue):

22848

(getIntValue):

22849

(getStringValue):

22850

(getBooleanValue):

22851

(myGetProperty):

22852

(mySetProperty):

22853

(myInvoke):

22854

(myAllocate):

22855

22856

2004-04-22 Darin Adler <darin@apple.com>

22857

22858

Reviewed by Maciej.

22859

22860

- fixed <rdar://problem/3627473>: "REGRESSION (125-137): memory trasher in UString::append, causing many different crashes"

22861

22862

* kjs/ustring.cpp:

22863

(KJS::UString::expandCapacity): Fix sizeof(UChar *) that should be sizeof(UChar).

22864

Was resulting in a buffer 2x the needed size.

22865

(KJS::UString::expandPreCapacity): Ditto.

22866

(KJS::UString::append): Fix malloc that is missing a sizeof(UChar).

22867

22868

2004-04-21 Maciej Stachowiak <mjs@apple.com>

22869

22870

Reviewed by Darin.

22871

22872

Preliminary change for conservative GC. Create "protected"

22873

subclasses to GC-protect objects when on heap, since we will soon

22874

remove the built-in refcounting of the normal wrapper classes. Use

22875

them where needed.

22876

22877

* JavaScriptCore.pbproj/project.pbxproj:

22878

* kjs/context.h:

22879

* kjs/internal.h:

22880

(KJS::InterpreterImp::globalObject):

22881

* kjs/interpreter.h:

22882

* kjs/property_map.cpp:

22883

* kjs/reference.h:

22884

* kjs/reference_list.cpp:

22885

22886

2004-04-19 Maciej Stachowiak <mjs@apple.com>

22887

22888

Reviewed by Dave.

22889

22890

Optimize prepend using the shared substring optimization. Also,

22891

limit the applicability of shared append and shared prepend. If

22892

you overdo it, it does more harm than good, because you create a

22893

bunch of strings that are disqualified from future shared

22894

append/prepend, for not much immediate savings in allocate/copy

22895

expense.

22896

22897

* kjs/ustring.cpp:

22898

(KJS::):

22899

(KJS::UString::Rep::create):

22900

(KJS::UString::expandedSize):

22901

(KJS::UString::usedPreCapacity):

22902

(KJS::UString::expandCapacity):

22903

(KJS::UString::expandPreCapacity):

22904

(KJS::UString::UString):

22905

(KJS::UString::append):

22906

(KJS::UString::operator=):

22907

* kjs/ustring.h:

22908

(KJS::UString::Rep::data):

22909

22910

2004-04-16 Maciej Stachowiak <mjs@apple.com>

22911

Reviewed by Richard.

22912

22913

No more need for Completion or Reference to privately inherit from

22914

Value, none of the superclass functionality is used.

22915

22916

* kjs/completion.h:

22917

* kjs/reference.h:

22918

22919

=== Safari-137 ===

22920

22921

2004-04-16 Richard Williamson <rjw@apple.com>

22922

22923

Added interpreter lock protection around object creation.

22924

22925

Reviewed by Chris.

22926

22927

* bindings/runtime.cpp:

22928

(Instance::createRuntimeObject):

22929

22930

2004-04-16 Maciej Stachowiak <mjs@apple.com>

22931

22932

Reviewed by Ken.

22933

22934

Another JavaScript speed improvement: use the mechanism from

22935

string append optimization to make taking a substring fast, again

22936

sharing the buffer.

22937

22938

A further 22% improvement on the 24fun string speed test.

22939

22940

* kjs/ustring.cpp:

22941

(KJS::):

22942

(KJS::UString::Rep::create):

22943

(KJS::UString::UString):

22944

(KJS::UString::append):

22945

(KJS::UString::operator=):

22946

(KJS::UString::substr):

22947

* kjs/ustring.h:

22948

(KJS::UString::Rep::data):

22949

22950

2004-04-13 Maciej Stachowiak <mjs@apple.com>

22951

22952

Reviewed by Darin.

22953

22954

- fixed <rdar://problem/3600695>: String manipulation in JavaScript 24fun test is very slow (slow)

22955

- fixed <rdar://problem/3600691>: Table generation test is really slow

22956

- fixed <rdar://problem/3600661>: 24fun date test is really slow

22957

22958

80% speedup on the string test, lesser speedups on the other two.

22959

22960

Two different optimizations here:

22961

22962

1) Avoid large overhead of scanning strings to see if they are all

22963

ASCII before numeric conversion.

22964

22965

* kjs/nodes.cpp:

22966

(AssignNode::evaluate): Don't convert to integer until we know for

22967

sure the operation will need it. Attempting to convert strings to

22968

numbers is a waste when they are being appended with +=.

22969

22970

2) Avoid huge cost of appending strings.

22971

22972

This is done by allowing multiple strings to share a buffer but

22973

actually use different ranges of it. The first time a string is

22974

appended to, we start leaving at least 10% extra space in the

22975

buffer, so doing N appends to the same string takes O(log N)

22976

mallocs instead of O(N).

22977

22978

* kjs/identifier.cpp:

22979

(KJS::Identifier::equal):

22980

(KJS::Identifier::add):

22981

* kjs/ustring.cpp:

22982

(KJS::):

22983

(KJS::UCharReference::operator=):

22984

(KJS::UCharReference::ref):

22985

(KJS::UString::Rep::create):

22986

(KJS::UString::Rep::destroy):

22987

(KJS::UString::expandedSize):

22988

(KJS::UString::usedCapacity):

22989

(KJS::UString::expandCapacity):

22990

(KJS::UString::UString):

22991

(KJS::UString::null):

22992

(KJS::UString::append):

22993

(KJS::UString::operator=):

22994

(KJS::UString::toStrictUInt32):

22995

(KJS::UString::detach):

22996

(KJS::KJS::operator==):

22997

* kjs/ustring.h:

22998

(KJS::UString::Rep::data):

22999

(KJS::UString::Rep::hash):

23000

23001

2004-04-09 Maciej Stachowiak <mjs@apple.com>

23002

23003

Reviewed by John.

23004

23005

- fix deployment build by avoiding deployment-only warning.

23006

23007

* kjs/scope_chain.cpp:

23008

(KJS::ScopeChain::bottom):

23009

23010

2004-04-09 Maciej Stachowiak <mjs@apple.com>

23011

23012

Reviewed by John.

23013

23014

Changed things so that newly created objects get a prototype based

23015

on the scope chain of the current function, rather than the

23016

interpreter that started execution. This fixes the following bugs:

23017

23018

<rdar://problem/3368523>: ARCH: wrong prototype used to create new objects (hang on lookup.atomica.com)

23019

<rdar://problem/3559173>: ARCH: Cannot scan using a HP Jetdirect product (JS object prototypes bind incorrectly)

23020

23021

* JavaScriptCore.pbproj/project.pbxproj:

23022

* kjs/array_object.cpp:

23023

(CompareWithCompareFunctionArguments::CompareWithCompareFunctionArguments):

23024

(ArrayProtoFuncImp::ArrayProtoFuncImp):

23025

(ArrayProtoFuncImp::call):

23026

(ArrayObjectImp::construct):

23027

* kjs/bool_object.cpp:

23028

(BooleanObjectImp::construct):

23029

* kjs/date_object.cpp:

23030

(DateProtoFuncImp::DateProtoFuncImp):

23031

(DateProtoFuncImp::call):

23032

(DateObjectImp::construct):

23033

* kjs/error_object.cpp:

23034

(ErrorObjectImp::construct):

23035

* kjs/function.cpp:

23036

(FunctionImp::FunctionImp):

23037

(FunctionImp::call):

23038

(DeclaredFunctionImp::construct):

23039

(ArgumentsImp::ArgumentsImp):

23040

(GlobalFuncImp::call):

23041

* kjs/function_object.cpp:

23042

(FunctionProtoFuncImp::call):

23043

(FunctionObjectImp::construct):

23044

* kjs/internal.cpp:

23045

(BooleanImp::toObject):

23046

(StringImp::toObject):

23047

(NumberImp::toObject):

23048

(InterpreterImp::InterpreterImp):

23049

(InterpreterImp::clear):

23050

(InterpreterImp::interpreterWithGlobalObject):

23051

* kjs/internal.h:

23052

* kjs/interpreter.cpp:

23053

(ExecState::lexicalInterpreter):

23054

* kjs/interpreter.h:

23055

(KJS::ExecState::dynamicInterpreter):

23056

(KJS::ExecState::interpreter):

23057

* kjs/math_object.cpp:

23058

(MathFuncImp::MathFuncImp):

23059

* kjs/nodes.cpp:

23060

(StatementNode::hitStatement):

23061

(StatementNode::abortStatement):

23062

(RegExpNode::evaluate):

23063

(ElementNode::evaluate):

23064

(ArrayNode::evaluate):

23065

(ObjectLiteralNode::evaluate):

23066

(PropertyValueNode::evaluate):

23067

(FunctionCallNode::evaluate):

23068

(FuncDeclNode::processFuncDecl):

23069

(FuncExprNode::evaluate):

23070

* kjs/number_object.cpp:

23071

(NumberObjectImp::construct):

23072

* kjs/object.cpp:

23073

(KJS::ObjectImp::defaultValue):

23074

(KJS::Error::create):

23075

* kjs/object_object.cpp:

23076

(ObjectObjectImp::construct):

23077

* kjs/reference.cpp:

23078

(Reference::putValue):

23079

* kjs/regexp_object.cpp:

23080

(RegExpProtoFuncImp::call):

23081

(RegExpObjectImp::arrayOfMatches):

23082

(RegExpObjectImp::construct):

23083

* kjs/scope_chain.cpp:

23084

(KJS::ScopeChain::bottom):

23085

* kjs/scope_chain.h:

23086

* kjs/string_object.cpp:

23087

(StringProtoFuncImp::StringProtoFuncImp):

23088

(StringProtoFuncImp::call):

23089

(StringObjectImp::construct):

23090

23091

=== Safari-136 ===

23092

23093

=== Safari-135 ===

23094

23095

2004-03-31 Richard Williamson <rjw@apple.com>

23096

23097

Tedious renames based on feedback from plugin-futures list.

23098

NP_ functions are renamed with NPN_ prefix.

23099

Types prefix renamed from NP_ to NP.

23100

NPN_CreateStringWithUTF8 and NPN_SetExceptionWithUTF8 now take a length, optionally -1 if string is null terminated.

23101

23102

No review because this was just a renaming patch.

23103

23104

* bindings/NP_jsobject.cpp:

23105

(listFromNPArray):

23106

(jsAllocate):

23107

(identiferFromNPIdentifier):

23108

(NPN_Call):

23109

(NPN_Evaluate):

23110

(NPN_GetProperty):

23111

(NPN_SetProperty):

23112

(NPN_RemoveProperty):

23113

(NPN_ToString):

23114

(NPN_GetPropertyAtIndex):

23115

(NPN_SetPropertyAtIndex):

23116

* bindings/NP_jsobject.h:

23117

* bindings/c/c_class.cpp:

23118

(CClass::_commonInit):

23119

(CClass::classForIsA):

23120

(CClass::CClass):

23121

(CClass::methodsNamed):

23122

(CClass::fieldNamed):

23123

* bindings/c/c_class.h:

23124

* bindings/c/c_instance.cpp:

23125

(CInstance::CInstance):

23126

(CInstance::~CInstance):

23127

(CInstance::operator=):

23128

(CInstance::invokeMethod):

23129

(CInstance::defaultValue):

23130

* bindings/c/c_instance.h:

23131

(KJS::Bindings::CInstance::getObject):

23132

* bindings/c/c_runtime.cpp:

23133

(CField::valueFromInstance):

23134

(CField::setValueToInstance):

23135

* bindings/c/c_runtime.h:

23136

(KJS::Bindings::CField::CField):

23137

(KJS::Bindings::CField::name):

23138

(KJS::Bindings::CMethod::CMethod):

23139

(KJS::Bindings::CMethod::name):

23140

* bindings/c/c_utility.cpp:

23141

(coerceValueToNPString):

23142

(convertValueToNPValueType):

23143

(convertNPValueTypeToValue):

23144

* bindings/c/c_utility.h:

23145

* bindings/npruntime.cpp:

23146

(NPN_IdentifierFromUTF8):

23147

(NPN_IsValidIdentifier):

23148

(NPN_GetIdentifiers):

23149

(NPN_UTF8FromIdentifier):

23150

(NPN_CreateObject):

23151

(NPN_RetainObject):

23152

(NPN_ReleaseObject):

23153

(NPN_IsKindOfClass):

23154

(NPN_SetExceptionWithUTF8):

23155

(NPN_SetException):

23156

(numberAllocate):

23157

(NPN_CreateNumberWithInt):

23158

(NPN_CreateNumberWithFloat):

23159

(NPN_CreateNumberWithDouble):

23160

(NPN_IntFromNumber):

23161

(NPN_FloatFromNumber):

23162

(NPN_DoubleFromNumber):

23163

(stringAllocate):

23164

(NPN_CreateStringWithUTF8):

23165

(NPN_CreateStringWithUTF16):

23166

(NPN_DeallocateUTF8):

23167

(NPN_UTF8FromString):

23168

(NPN_UTF16FromString):

23169

(NPN_StringLength):

23170

(booleanAllocate):

23171

(NPN_CreateBoolean):

23172

(NPN_BoolFromBoolean):

23173

(nullAllocate):

23174

(NPN_GetNull):

23175

(undefinedAllocate):

23176

(NPN_GetUndefined):

23177

(arrayAllocate):

23178

(arrayDeallocate):

23179

(NPN_CreateArray):

23180

(NPN_CreateArrayV):

23181

(NPN_ObjectAtIndex):

23182

* bindings/npruntime.h:

23183

* bindings/runtime.cpp:

23184

(Instance::createBindingForLanguageInstance):

23185

* bindings/testbindings.cpp:

23186

(initializeIdentifiers):

23187

(myHasProperty):

23188

(myHasMethod):

23189

(myGetProperty):

23190

(mySetProperty):

23191

(logMessage):

23192

(setDoubleValue):

23193

(setIntValue):

23194

(setStringValue):

23195

(setBooleanValue):

23196

(getDoubleValue):

23197

(getIntValue):

23198

(getStringValue):

23199

(getBooleanValue):

23200

(myInvoke):

23201

(myAllocate):

23202

(myInvalidate):

23203

(myDeallocate):

23204

(main):

23205

23206

2004-03-31 Richard Williamson <rjw@apple.com>

23207

23208

Changed references to NP_runtime.h to npruntime.h

23209

23210

* JavaScriptCore.pbproj/project.pbxproj:

23211

* bindings/NP_jsobject.h:

23212

* bindings/c/c_class.h:

23213

* bindings/c/c_instance.h:

23214

* bindings/c/c_runtime.h:

23215

* bindings/c/c_utility.h:

23216

* bindings/npruntime.cpp:

23217

23218

2004-03-31 Richard Williamson <rjw@apple.com>

23219

23220

Renamed NP_runtime.h to npruntime.h to match Netscape SDK.

23221

23222

* JavaScriptCore.pbproj/project.pbxproj:

23223

* bindings/NP_jsobject.h:

23224

* bindings/npruntime.cpp:

23225

23226

=== Safari-134 ===

23227

23228

2004-03-23 Richard Williamson <rjw@apple.com>

23229

23230

Added implementation of KJS::Value <-> NP_Object conversion functions.

23231

Augmented test program for 'C' bindings.

23232

Added asserts and parameter checking to all public API.

23233

23234

Reviewed by Ken.

23235

23236

* JavaScriptCore.pbproj/project.pbxproj:

23237

* bindings/NP_jsobject.cpp:

23238

(NP_ToString):

23239

* bindings/NP_jsobject.h: Added.

23240

* bindings/NP_runtime.cpp:

23241

(NP_IdentifierFromUTF8):

23242

(NP_IsValidIdentifier):

23243

(NP_GetIdentifiers):

23244

(NP_CreateObject):

23245

(NP_RetainObject):

23246

(NP_ReleaseObject):

23247

(NP_IsKindOfClass):

23248

(NP_SetExceptionWithUTF8):

23249

(NP_SetException):

23250

(NP_IntFromNumber):

23251

(NP_FloatFromNumber):

23252

(NP_DoubleFromNumber):

23253

(NP_CreateStringWithUTF8):

23254

(NP_CreateStringWithUTF16):

23255

(NP_DeallocateUTF8):

23256

(NP_UTF8FromString):

23257

(NP_UTF16FromString):

23258

(NP_StringLength):

23259

(NP_BoolFromBoolean):

23260

* bindings/NP_runtime.h:

23261

* bindings/c/c_instance.cpp:

23262

(CInstance::invokeMethod):

23263

* bindings/c/c_utility.cpp:

23264

(coerceValueToNPString):

23265

(convertValueToNPValueType):

23266

(convertNPValueTypeToValue):

23267

* bindings/c/c_utility.h:

23268

* bindings/test.js:

23269

* bindings/testC.js: Added.

23270

* bindings/testbindings.cpp:

23271

(logMessage):

23272

(setDoubleValue):

23273

(setIntValue):

23274

(setStringValue):

23275

(setBooleanValue):

23276

(getDoubleValue):

23277

(getIntValue):

23278

(getStringValue):

23279

(getBooleanValue):

23280

(myInterfaceInvoke):

23281

(myInterfaceAllocate):

23282

23283

=== Safari-133 ===

23284

23285

2004-03-19 Darin Adler <darin@apple.com>

23286

23287

Reviewed by Ken.

23288

23289

- fixed problem with methods like setUTCHour

23290

23291

* kjs/date_object.cpp: (DateProtoFuncImp::call): Fix conversion back to time_t to use the appropriate

23292

GMT vs. local time function based on the utc flag.

23293

23294

2004-03-17 Richard Williamson <rjw@apple.com>

23295

23296

Added a context parameter to result callbacks use by JavaScriptObject functions. This was a change requested by Eric Carlson on the QT plugin team.

23297

23298

Reviewed by Ken.

23299

23300

* bindings/NP_jsobject.cpp:

23301

(NP_Call):

23302

(NP_Evaluate):

23303

(NP_GetProperty):

23304

(NP_ToString):

23305

(NP_GetPropertyAtIndex):

23306

* bindings/NP_runtime.h:

23307

23308

2004-03-16 Richard Williamson <rjw@apple.com>

23309

23310

Fixed 3590169. Regression (crash) caused by the switch to MethodLists. Crash when attempting to invoke a method from JavaScript to Java that is not implemented.

23311

23312

Reviewed by John.

23313

23314

* bindings/jni/jni_class.cpp:

23315

(JavaClass::methodsNamed):

23316

23317

2004-03-15 Richard Williamson <rjw@apple.com>

23318

23319

Fixed 3570854. Don't attempt to convert Null to strings. We

23320

were incorrectly converting to "Null".

23321

23322

Actually fixed by Scott Kovatch.

23323

23324

Reviewed by Richard.

23325

23326

* bindings/jni/jni_utility.cpp:

23327

(KJS::Bindings::convertValueToJValue):

23328

23329

=== Safari-132 ===

23330

23331

2004-03-11 Richard Williamson <rjw@apple.com>

23332

23333

Stitched together the NP stuff to our language independent

23334

JavaScript binding stuff. Very close to being done.

23335

23336

Added program to test C bindings (and NP stuff). Just tests

23337

properties. Will add methods and JavaScript access, etc.

23338

23339

Updated Makefile.am to account for new bindings/c directory.

23340

23341

Change NP_UTF8 from "const char *" to "char" to allow for

23342

declarations like "const NP_UTF8 *" and "NP_UTF8 *". Ditto

23343

for NP_UTF16.

23344

23345

Added NP_IsValidIdentifier().

23346

23347

Reviewed by Chris.

23348

23349

* JavaScriptCore.pbproj/project.pbxproj:

23350

* Makefile.am:

23351

* bindings/NP_jsobject.cpp:

23352

(identiferFromNPIdentifier):

23353

(NP_Evaluate):

23354

* bindings/NP_runtime.cpp:

23355

(NP_IdentifierFromUTF8):

23356

(NP_IsValidIdentifier):

23357

(NP_GetIdentifiers):

23358

(NP_UTF8FromIdentifier):

23359

(NP_SetExceptionWithUTF8):

23360

(NP_SetException):

23361

(NP_CreateStringWithUTF8):

23362

(NP_CreateStringWithUTF16):

23363

(NP_UTF8FromString):

23364

(NP_UTF16FromString):

23365

* bindings/NP_runtime.h:

23366

* bindings/c/c_class.cpp: Added.

23367

(CClass::_commonDelete):

23368

(CClass::_commonCopy):

23369

(CClass::_commonInit):

23370

(_createClassesByIsAIfNecessary):

23371

(CClass::classForIsA):

23372

(CClass::CClass):

23373

(CClass::name):

23374

(CClass::methodsNamed):

23375

(CClass::fieldNamed):

23376

* bindings/c/c_class.h: Added.

23377

(KJS::Bindings::CClass::~CClass):

23378

(KJS::Bindings::CClass::CClass):

23379

(KJS::Bindings::CClass::operator=):

23380

(KJS::Bindings::CClass::constructorAt):

23381

(KJS::Bindings::CClass::numConstructors):

23382

* bindings/c/c_instance.cpp: Added.

23383

(CInstance::CInstance):

23384

(CInstance::~CInstance):

23385

(CInstance::operator=):

23386

(CInstance::getClass):

23387

(CInstance::begin):

23388

(CInstance::end):

23389

(CInstance::invokeMethod):

23390

(CInstance::defaultValue):

23391

(CInstance::stringValue):

23392

(CInstance::numberValue):

23393

(CInstance::booleanValue):

23394

(CInstance::valueOf):

23395

* bindings/c/c_instance.h: Added.

23396

(KJS::Bindings::CInstance::getObject):

23397

* bindings/c/c_runtime.cpp: Added.

23398

(CField::valueFromInstance):

23399

(CField::setValueToInstance):

23400

* bindings/c/c_runtime.h: Added.

23401

(KJS::Bindings::CField::CField):

23402

(KJS::Bindings::CField::name):

23403

(KJS::Bindings::CField::type):

23404

(KJS::Bindings::CMethod::CMethod):

23405

(KJS::Bindings::CMethod::name):

23406

(KJS::Bindings::CMethod::numParameters):

23407

* bindings/c/c_utility.cpp: Added.

23408

(coerceValueToNPValueType):

23409

(convertValueToNPValueType):

23410

(convertNPValueTypeToValue):

23411

* bindings/c/c_utility.h: Added.

23412

* bindings/make_testbindings:

23413

* bindings/runtime.cpp:

23414

(Instance::createBindingForLanguageInstance):

23415

* bindings/runtime.h:

23416

(KJS::Bindings::Instance::):

23417

* bindings/testbindings.cpp: Added.

23418

(initializeIdentifiers):

23419

(myInterfaceHasProperty):

23420

(myInterfaceHasMethod):

23421

(myInterfaceGetProperty):

23422

(myInterfaceSetProperty):

23423

(myInterfaceInvoke):

23424

(myInterfaceAllocate):

23425

(myInterfaceInvalidate):

23426

(myInterfaceDeallocate):

23427

(GlobalImp::className):

23428

(readJavaScriptFromFile):

23429

(main):

23430

23431

2004-03-10 Richard Williamson <rjw@apple.com>

23432

23433

Made changes to support new asychronous approach to calls from

23434

plugin to JavaScript

23435

23436

Reviewed by Chris.

23437

23438

* bindings/NP_jsobject.cpp:

23439

(NP_Call):

23440

(NP_Evaluate):

23441

(NP_GetProperty):

23442

(NP_ToString):

23443

(NP_GetPropertyAtIndex):

23444

* bindings/NP_runtime.h:

23445

* bindings/make_testbindings:

23446

* bindings/runtime.cpp:

23447

(Instance::createBindingForLanguageInstance):

23448

23449

2004-03-10 Richard Williamson <rjw@apple.com>

23450

23451

Updated header to include proposed changes from

23452

plugin-futures list. Calls from plugin to JavaScript

23453

are now asynchronous.

23454

23455

Reviewed by Chris.

23456

23457

* bindings/NP_runtime.h:

23458

23459

=== Safari-131 ===

23460

23461

2004-03-04 Richard Williamson <rjw@apple.com>

23462

23463

Implementation of NP_JavaScriptObject. This is the 'C' class

23464

that wraps a JavaScript object.

23465

23466

Reviewed by Chris.

23467

23468

* JavaScriptCore.pbproj/project.pbxproj:

23469

* bindings/NP_jsobject.cpp: Added.

23470

(coerceValueToNPValueType):

23471

(convertValueToNPValueType):

23472

(convertNPValueTypeToValue):

23473

(listFromNPArray):

23474

(jsAllocate):

23475

(jsDeallocate):

23476

(identiferFromNPIdentifier):

23477

(NP_Call):

23478

(NP_Evaluate):

23479

(NP_GetProperty):

23480

(NP_SetProperty):

23481

(NP_RemoveProperty):

23482

(NP_ToString):

23483

(NP_GetPropertyAtIndex):

23484

(NP_SetPropertyAtIndex):

23485

* bindings/NP_runtime.cpp:

23486

(NP_ObjectAtIndex):

23487

* bindings/NP_runtime.h:

23488

* bindings/runtime_object.h:

23489

23490

2004-03-04 Richard Williamson <rjw@apple.com>

23491

23492

Added NP_Array implementation.

23493

23494

Changed NP_Boolean to just depend on two static instances, no

23495

space is required for values.

23496

23497

Reviewed by Chris.

23498

23499

* bindings/NP_runtime.cpp:

23500

(NP_CreateBoolean):

23501

(NP_BoolFromBoolean):

23502

(arrayAllocate):

23503

(arrayDeallocate):

23504

(NP_CreateArray):

23505

(NP_CreateArrayV):

23506

(NP_ObjectAtIndex):

23507

* bindings/NP_runtime.h:

23508

23509

2004-03-03 Darin Adler <darin@apple.com>

23510

23511

Reviewed by Vicki.

23512

23513

* English.lproj/InfoPlist.strings: Removed. No need to localize the version and

23514

23515

* JavaScriptCore.pbproj/project.pbxproj: Removed InfoPlist.strings from build.

23516

23517

2004-03-03 Richard Williamson <rjw@apple.com>

23518

23519

More 'C' binding implementation. Fleshed out all the

23520

'primitive' data types.

23521

23522

Reviewed by Chris.

23523

23524

* bindings/NP_runtime.cpp:

23525

(NP_ReleaseObject):

23526

(numberAllocate):

23527

(stringAllocate):

23528

(stringDeallocate):

23529

(NP_CreateStringWithUTF8):

23530

(NP_CreateStringWithUTF16):

23531

(NP_UTF8FromString):

23532

(NP_UTF16FromString):

23533

(NP_StringLength):

23534

(booleanAllocate):

23535

(booleanDeallocate):

23536

(NP_CreateBoolean):

23537

(NP_BoolFromBoolean):

23538

(nullAllocate):

23539

(nullDeallocate):

23540

(NP_GetNull):

23541

(undefinedAllocate):

23542

(undefinedDeallocate):

23543

(NP_GetUndefined):

23544

* bindings/NP_runtime.h:

23545

23546

2004-03-03 Richard Williamson <rjw@apple.com>

23547

23548

More 'C' binding implementation.

23549

23550

Reviewed by Chris.

23551

23552

* bindings/NP_runtime.cpp:

23553

(identifierEqual):

23554

(identifierHash):

23555

(getIdentifierDictionary):

23556

(NP_IdentifierFromUTF8):

23557

(NP_UTF8FromIdentifier):

23558

(NP_CreateObject):

23559

(NP_ReleaseObject):

23560

(NP_IsKindOfClass):

23561

(numberCreate):

23562

(NP_CreateNumberWithInt):

23563

(NP_CreateNumberWithFloat):

23564

(NP_CreateNumberWithDouble):

23565

(NP_IntFromNumber):

23566

(NP_FloatFromNumber):

23567

(NP_DoubleFromNumber):

23568

* bindings/NP_runtime.h:

23569

23570

2004-03-02 Richard Williamson <rjw@apple.com>

23571

23572

Removed retain/release from NP_Class. Classes will not be allowed to implement their

23573

own customer retain/release scheme.

23574

23575

Reviewed by Chris.

23576

23577

* bindings/NP_runtime.cpp:

23578

(NP_RetainObject):

23579

(NP_ReleaseObject):

23580

* bindings/NP_runtime.h:

23581

23582

2004-03-02 Richard Williamson <rjw@apple.com>

23583

23584

C binding API. Partial implementation.

23585

23586

Completed ObjectiveC bindings (not based on the C API). These will re-implemented over the C binding API, but I wanted to get this code in the tree.

23587

23588

Factored root object reference counting scheme. It is now useful independent

23589

of LiveConnect.

23590

23591

Reviewed by Chris.

23592

23593

* JavaScriptCore.pbproj/project.pbxproj:

23594

* bindings/NP_runtime.cpp: Added.

23595

(NP_IdentifierFromUTF8):

23596

(NP_GetIdentifiers):

23597

(NP_UTF8FromIdentifier):

23598

(NP_CreateObject):

23599

(NP_RetainObject):

23600

(NP_ReleaseObject):

23601

(NP_IsKindOfClass):

23602

(NP_SetException):

23603

(NP_Call):

23604

(NP_Evaluate):

23605

(NP_GetProperty):

23606

(NP_SetProperty):

23607

(NP_RemoveProperty):

23608

(NP_ToString):

23609

(NP_GetPropertyAtIndex):

23610

(NP_SetPropertyAtIndex):

23611

(NP_CreateNumberWithInt):

23612

(NP_CreateNumberWithFloat):

23613

(NP_CreateNumberWithDouble):

23614

(NP_IntFromNumber):

23615

(NP_FloatFromNumber):

23616

(NP_DoubleFromNumber):

23617

(NP_CreateStringWithUTF8):

23618

(NP_CreateStringWithUTF16):

23619

(NP_UTF8FromString):

23620

(NP_UTF16FromString):

23621

(NP_CreateBoolean):

23622

(NP_BoolFromBoolean):

23623

(NP_GetNull):

23624

(NP_GetUndefined):

23625

(NP_CreateArray):

23626

(NP_CreateArrayV):

23627

(NP_ObjectAtIndex):

23628

* bindings/NP_runtime.h: Added.

23629

* bindings/jni/jni_jsobject.cpp:

23630

(JSObject::invoke):

23631

(JSObject::finalize):

23632

(JSObject::createNative):

23633

(JSObject::convertValueToJObject):

23634

* bindings/jni/jni_jsobject.h:

23635

* bindings/objc/objc_jsobject.h:

23636

* bindings/objc/objc_jsobject.mm:

23637

(rootForView):

23638

(windowJavaScriptObject):

23639

(-[JavaScriptObject initWithObjectImp:KJS::root:Bindings::]):

23640

(-[JavaScriptObject dealloc]):

23641

(-[JavaScriptObject _convertValueToObjcValue:KJS::]):

23642

(-[JavaScriptObject call:arguments:]):

23643

(-[JavaScriptObject evaluate:]):

23644

(-[JavaScriptObject getMember:]):

23645

(-[JavaScriptObject setMember:value:]):

23646

(-[JavaScriptObject removeMember:]):

23647

(-[JavaScriptObject toString]):

23648

(-[JavaScriptObject getSlot:]):

23649

(-[JavaScriptObject setSlot:value:]):

23650

* bindings/objc/objc_utility.h:

23651

* bindings/objc/objc_utility.mm:

23652

(KJS::Bindings::convertValueToObjcValue):

23653

* bindings/runtime_root.cpp: Added.

23654

(getReferencesByRootDictionary):

23655

(getReferencesDictionary):

23656

(KJS::Bindings::findReferenceDictionary):

23657

(KJS::Bindings::rootForImp):

23658

(KJS::Bindings::addNativeReference):

23659

(KJS::Bindings::removeNativeReference):

23660

(completedJavaScriptAccess):

23661

(initializeJavaScriptAccessLock):

23662

(lockJavaScriptAccess):

23663

(unlockJavaScriptAccess):

23664

(RootObject::dispatchToJavaScriptThread):

23665

(performJavaScriptAccess):

23666

(RootObject::setFindRootObjectForNativeHandleFunction):

23667

(RootObject::removeAllNativeReferences):

23668

* bindings/runtime_root.h: Added.

23669

(KJS::Bindings::RootObject::RootObject):

23670

(KJS::Bindings::RootObject::~RootObject):

23671

(KJS::Bindings::RootObject::setRootObjectImp):

23672

(KJS::Bindings::RootObject::rootObjectImp):

23673

(KJS::Bindings::RootObject::setInterpreter):

23674

(KJS::Bindings::RootObject::interpreter):

23675

(KJS::Bindings::RootObject::findRootObjectForNativeHandleFunction):

23676

(KJS::Bindings::RootObject::runLoop):

23677

(KJS::Bindings::RootObject::performJavaScriptSource):

23678

23679

=== Safari-130 ===

23680

23681

=== Safari-129 ===

23682

23683

2004-02-18 Richard Williamson <rjw@apple.com>

23684

23685

Added NSNumber/Number conversion.

23686

23687

Removed some unnecessary KJS:: namespace specifiers.

23688

23689

Reviewed by Ken.

23690

23691

* bindings/objc/objc_utility.mm:

23692

(KJS::Bindings::convertValueToObjcValue):

23693

(KJS::Bindings::convertObjcValueToValue):

23694

* bindings/runtime_array.h:

23695

23696

2004-02-18 Richard Williamson <rjw@apple.com>

23697

23698

Added support for export NSArrays.

23699

23700

Updated valueAt() to take an ExecState so we can throw

23701

JS exceptions.

23702

23703

Implemented excludeSelectorFromJavaScript: in ObjcClass. This allows

23704

ObjectiveC classes to control the visibility of their methods in

23705

JavaScript.

23706

23707

Reviewed by Ken.

23708

23709

* bindings/jni/jni_runtime.cpp:

23710

(JavaField::valueFromInstance):

23711

(JavaArray::valueAt):

23712

* bindings/jni/jni_runtime.h:

23713

* bindings/objc/objc_class.mm:

23714

(ObjcClass::methodsNamed):

23715

* bindings/objc/objc_runtime.h:

23716

(KJS::Bindings::ObjcArray::getObjcArray):

23717

* bindings/objc/objc_runtime.mm:

23718

(ObjcField::valueFromInstance):

23719

(ObjcField::setValueToInstance):

23720

(ObjcArray::ObjcArray):

23721

(ObjcArray::~ObjcArray):

23722

(ObjcArray::operator=):

23723

(ObjcArray::setValueAt):

23724

(ObjcArray::valueAt):

23725

(ObjcArray::getLength):

23726

* bindings/objc/objc_utility.mm:

23727

(KJS::Bindings::convertValueToObjcValue):

23728

(KJS::Bindings::convertObjcValueToValue):

23729

* bindings/runtime.cpp:

23730

(Instance::getValueOfField):

23731

* bindings/runtime.h:

23732

* bindings/runtime_array.cpp:

23733

(RuntimeArrayImp::get):

23734

* bindings/runtime_object.cpp:

23735

(RuntimeObjectImp::get):

23736

23737

2004-02-17 Richard Williamson <rjw@apple.com>

23738

23739

Added String <-> NSString conversion.

23740

Added tests of String <-> NSString conversion to test program.

23741

23742

Reviewed by Chris.

23743

23744

* bindings/objc/objc_utility.mm:

23745

(KJS::Bindings::convertValueToObjcValue):

23746

(KJS::Bindings::convertObjcValueToValue):

23747

* bindings/test.js:

23748

* bindings/testbindings.mm:

23749

(-[MyFirstInterface getString]):

23750

23751

2004-02-15 Darin Adler <darin@apple.com>

23752

23753

Reviewed by Dave.

23754

23755

* JavaScriptCore.pbproj/project.pbxproj: Tweak build styles a bit, fixing OptimizedWithSymbols,

23756

and removing redundant settings of things that match defaults in other build styles.

23757

23758

2004-02-13 Richard Williamson <rjw@apple.com>

23759

23760

Work towards the JavaScript ObjC bindings. The bindings now work for

23761

simple scalar types. testbindings.mm is an illustration of how the

23762

bindings work.

23763

23764

Reviewed by Ken.

23765

23766

* JavaScriptCore.pbproj/project.pbxproj:

23767

* Makefile.am:

23768

* bindings/jni/jni_class.cpp:

23769

(JavaClass::methodsNamed):

23770

* bindings/jni/jni_class.h:

23771

* bindings/jni/jni_instance.cpp:

23772

(JavaInstance::invokeMethod):

23773

* bindings/jni/jni_instance.h:

23774

* bindings/jni/jni_runtime.h:

23775

(KJS::Bindings::JavaMethod::returnType):

23776

* bindings/make_testbindings: Added.

23777

* bindings/objc/objc_class.h: Added.

23778

(KJS::Bindings::ObjcClass::~ObjcClass):

23779

(KJS::Bindings::ObjcClass::ObjcClass):

23780

(KJS::Bindings::ObjcClass::operator=):

23781

(KJS::Bindings::ObjcClass::constructorAt):

23782

(KJS::Bindings::ObjcClass::numConstructors):

23783

* bindings/objc/objc_class.mm: Added.

23784

(ObjcClass::_commonDelete):

23785

(ObjcClass::_commonCopy):

23786

(ObjcClass::_commonInit):

23787

(_createClassesByIsAIfNecessary):

23788

(ObjcClass::classForIsA):

23789

(ObjcClass::ObjcClass):

23790

(ObjcClass::name):

23791

(ObjcClass::methodsNamed):

23792

(ObjcClass::fieldNamed):

23793

* bindings/objc/objc_header.h: Added.

23794

* bindings/objc/objc_instance.h: Added.

23795

(KJS::Bindings::ObjcInstance::getObject):

23796

* bindings/objc/objc_instance.mm: Added.

23797

(ObjcInstance::ObjcInstance):

23798

(ObjcInstance::~ObjcInstance):

23799

(ObjcInstance::operator=):

23800

(ObjcInstance::begin):

23801

(ObjcInstance::end):

23802

(ObjcInstance::getClass):

23803

(ObjcInstance::invokeMethod):

23804

(ObjcInstance::defaultValue):

23805

(ObjcInstance::stringValue):

23806

(ObjcInstance::numberValue):

23807

(ObjcInstance::booleanValue):

23808

(ObjcInstance::valueOf):

23809

* bindings/objc/objc_jsobject.h: Added.

23810

* bindings/objc/objc_jsobject.mm: Added.

23811

* bindings/objc/objc_runtime.h:

23812

(KJS::Bindings::ObjcField::~ObjcField):

23813

(KJS::Bindings::ObjcField::ObjcField):

23814

(KJS::Bindings::ObjcField::operator=):

23815

(KJS::Bindings::ObjcMethod::ObjcMethod):

23816

(KJS::Bindings::ObjcMethod::~ObjcMethod):

23817

(KJS::Bindings::ObjcMethod::operator=):

23818

* bindings/objc/objc_runtime.mm: Added.

23819

(ObjcMethod::ObjcMethod):

23820

(ObjcMethod::name):

23821

(ObjcMethod::numParameters):

23822

(ObjcMethod::getMethodSignature):

23823

(ObjcField::ObjcField):

23824

(ObjcField::name):

23825

(ObjcField::type):

23826

(ObjcField::valueFromInstance):

23827

(ObjcField::setValueToInstance):

23828

* bindings/objc/objc_utility.h: Added.

23829

(KJS::Bindings::):

23830

* bindings/objc/objc_utility.mm: Added.

23831

(KJS::Bindings::JSMethodNameToObjCMethodName):

23832

(KJS::Bindings::convertValueToObjcValue):

23833

(KJS::Bindings::convertObjcValueToValue):

23834

(KJS::Bindings::objcValueTypeForType):

23835

* bindings/runtime.cpp:

23836

(MethodList::MethodList):

23837

(MethodList::operator=):

23838

(Instance::setValueOfField):

23839

(Instance::createBindingForLanguageInstance):

23840

(Instance::createRuntimeObject):

23841

* bindings/runtime.h:

23842

* bindings/runtime_method.cpp:

23843

(RuntimeMethodImp::RuntimeMethodImp):

23844

(RuntimeMethodImp::get):

23845

(RuntimeMethodImp::call):

23846

* bindings/runtime_method.h:

23847

* bindings/runtime_object.cpp:

23848

(RuntimeObjectImp::get):

23849

(RuntimeObjectImp::hasProperty):

23850

* bindings/test.js: Added.

23851

* bindings/testbindings.mm: Added.

23852

(-[MySecondInterface init]):

23853

(-[MyFirstInterface init]):

23854

(-[MyFirstInterface dealloc]):

23855

(+[MyFirstInterface JavaScriptNameForSelector:]):

23856

(-[MyFirstInterface getInt]):

23857

(-[MyFirstInterface setInt:]):

23858

(-[MyFirstInterface getMySecondInterface]):

23859

(-[MyFirstInterface logMessage:]):

23860

(GlobalImp::className):

23861

(readJavaScriptFromFile):

23862

(main):

23863

23864

=== Safari-128 ===

23865

23866

2004-02-08 Darin Adler <darin@apple.com>

23867

23868

Reviewed by Dave.

23869

23870

- fixed things seen in the profile, for a total speedup of 4% on cvs-base (including changes across all projects)

23871

23872

* JavaScriptCorePrefix.h: Add a workaround for a bug in our system headers that prevents the <ctype.h>

23873

macros from working right in C++ code that uses the <cctype> header.

23874

23875

* kjs/ustring.cpp:

23876

(KJS::inlineUTF8SequenceLengthNonASCII): Added.

23877

(KJS::UTF8SequenceLengthNonASCII): Added.

23878

(KJS::inlineUTF8SequenceLength): Added.

23879

(KJS::UTF8SequenceLength): Calls inlineUTF8SequenceLengthNonASCII now.

23880

(KJS::decodeUTF8Sequence): Use new inlineUTF8SequenceLengthNonASCII; faster for ASCII.

23881

(KJS::createSortedOffsetsArray): Add special case for 1, 2, and 3 offsets, so we don't do qsort for those.

23882

(KJS::convertUTF16OffsetsToUTF8Offsets): Use new inlineUTF8SequenceLengthNonASCII; faster for ASCII.

23883

(KJS::convertUTF8OffsetsToUTF16Offsets): Use new inlineUTF8SequenceLengthNonASCII; faster for ASCII.

23884

23885

- fixed the test program so it won't hit the interpreter lock assertion

23886

23887

* kjs/testkjs.cpp: (main): Just lock around the whole thing, since the test is singly threaded.

23888

23889

=== Safari-127 ===

23890

23891

2004-02-06 Richard Williamson <rjw@apple.com>

23892

23893

Fixed 3550242 and 3546977. The first diff prevents an assert from firing. The second diff prevents a JavaScript exception, caused be an invalid conversion, which has a downstream consequence of preventing a valid conversion.

23894

23895

Reviewed by John.

23896

23897

* bindings/jni/jni_jsobject.cpp:

23898

(JSObject::toString):

23899

* bindings/jni/jni_utility.cpp:

23900

(KJS::Bindings::convertValueToJValue):

23901

23902

2004-02-02 Darin Adler <darin@apple.com>

23903

23904

Reviewed by Maciej.

23905

23906

- fixed <rdar://problem/3546613>: array of negative size leads to crash (test page at oscar.the-rileys.net)

23907

23908

* kjs/array_object.cpp:

23909

(ArrayInstanceImp::ArrayInstanceImp): If the length is greater than 10,000, don't allocate an array until

23910

we start putting values in. This prevents new Array(2147483647) from causing trouble.

23911

(ArrayObjectImp::construct): Check number as described in specification, and raise a range error if the

23912

number is out of range. This prevents new Array(-1) from causing trouble.

23913

23914

- fixed <rdar://problem/3545756>: Math.round screws up on numbers bigger than 2^31 (incorrect results on HP-35 calculator page)

23915

23916

* kjs/math_object.cpp: (MathFuncImp::call): Change implementation to be much simpler and not involve

23917

casting to int. Results now match those in other browsers.

23918

23919

2004-02-02 Darin Adler <darin@apple.com>

23920

23921

Reviewed by Maciej.

23922

23923

- fixed <rdar://problem/3519285>: integer operations on large negative numbers yield bad results (discovered with "HTMLCrypt")

23924

- fixed other related overflow issues

23925

23926

* kjs/value.h: Changed return types of toInteger, toInt32, toUInt32, and toUInt16.

23927

* kjs/value.cpp:

23928

(ValueImp::toInteger): Change to return a double, since this operation, from the ECMA specification,

23929

must not restrict values to the range of a particular integer type.

23930

(ValueImp::toInt32): Used a sized integer type for the result of this function, and also added

23931

proper handling for negative results from fmod.

23932

(ValueImp::toUInt32): Ditto.

23933

(ValueImp::toUInt16): Ditto.

23934

(ValueImp::dispatchToUInt32): Changed result type from unsigned to uint32_t.

23935

23936

* kjs/array_object.cpp: (ArrayProtoFuncImp::call): Use a double instead of an int to handle

23937

out-of-integer-range values better in the slice function.

23938

* kjs/internal.cpp: (KJS::roundValue): Streamline the function, handling NAN and infinity properly.

23939

* kjs/number_object.cpp: (NumberProtoFuncImp::call): Use a double instead of an int to handle

23940

out-of-integer-range values better in the toString function.

23941

* kjs/string_object.cpp: (StringProtoFuncImp::call): Use a double instead of an int to handle

23942

out-of-integer-range values better in the charAt, charCodeAt, indexOf, lastIndexOf, slice,

23943

and substr functions.

23944

23945

=== Safari-126 ===

23946

23947

2004-01-30 Richard Williamson <rjw@apple.com>

23948

23949

Fixed 3542044. Create KJS::String using UString constructor instead of passing UTF8 string to char* constructor.

23950

23951

Reviewed by Darin.

23952

23953

* bindings/jni/jni_instance.cpp:

23954

(JavaInstance::stringValue):

23955

23956

2004-01-26 Darin Adler <darin@apple.com>

23957

23958

* Makefile.am: Switch from pbxbuild to xcodebuild.

23959

23960

2004-01-22 Richard Williamson <rjw@apple.com>

23961

23962

Added stubs for ObjC language binding to JavaScript.

23963

23964

* JavaScriptCore.pbproj/project.pbxproj:

23965

* bindings/jni/jni_runtime.h:

23966

* bindings/objc/objc_runtime.h: Added.

23967

(KJS::Bindings::ObjcParameter::ObjcParameter):

23968

(KJS::Bindings::ObjcParameter::~ObjcParameter):

23969

(KJS::Bindings::ObjcParameter::operator=):

23970

(KJS::Bindings::ObjcParameter::type):

23971

(KJS::Bindings::ObjcConstructor::ObjcConstructor):

23972

(KJS::Bindings::ObjcConstructor::~ObjcConstructor):

23973

(KJS::Bindings::ObjcConstructor::_commonCopy):

23974

(KJS::Bindings::ObjcConstructor::operator=):

23975

(KJS::Bindings::ObjcConstructor::value):

23976

(KJS::Bindings::ObjcConstructor::parameterAt):

23977

(KJS::Bindings::ObjcConstructor::numParameters):

23978

(KJS::Bindings::ObjcField::ObjcField):

23979

(KJS::Bindings::ObjcField::~ObjcField):

23980

* bindings/runtime.h:

23981

23982

2004-01-22 Richard Williamson <rjw@apple.com>

23983

23984

Simplified JavaString by using UString as backing store. This

23985

revealed a bug in CString's assignment operator which I fixed.

23986

23987

Removed some dead code.

23988

23989

Reviewed by John.

23990

23991

* bindings/jni/jni_runtime.h:

23992

(KJS::Bindings::JavaString::JavaString):

23993

(KJS::Bindings::JavaString::_commonInit):

23994

(KJS::Bindings::JavaString::UTF8String):

23995

(KJS::Bindings::JavaString::uchars):

23996

(KJS::Bindings::JavaString::length):

23997

(KJS::Bindings::JavaString::ustring):

23998

* bindings/runtime_object.cpp:

23999

(RuntimeObjectImp::RuntimeObjectImp):

24000

* bindings/runtime_object.h:

24001

* kjs/ustring.cpp:

24002

(KJS::CString::CString):

24003

(KJS::CString::operator=):

24004

24005

=== Safari-125 ===

24006

24007

=== Safari-124 ===

24008

24009

2004-01-16 Richard Williamson <rjw@apple.com>

24010

24011

Fixed 3525853. We weren't handling mapping to overloaded Java

24012

methods very well. Even though this is undefined the other

24013

browsers support it. Also fixed a bug with returning arrays

24014

from Java functions.

24015

24016

Reviewed by John.

24017

24018

* bindings/jni/jni_class.cpp:

24019

(JavaClass::_commonInit):

24020

(JavaClass::methodsNamed):

24021

* bindings/jni/jni_class.h:

24022

* bindings/jni/jni_instance.cpp:

24023

(JavaInstance::invokeMethod):

24024

* bindings/jni/jni_instance.h:

24025

* bindings/jni/jni_runtime.cpp:

24026

(JavaArray::convertJObjectToArray):

24027

(JavaField::valueFromInstance):

24028

(JavaMethod::signature):

24029

(JavaArray::valueAt):

24030

* bindings/jni/jni_runtime.h:

24031

* bindings/jni_jsobject.cpp:

24032

(JSObject::call):

24033

(JSObject::convertJObjectToValue):

24034

* bindings/runtime.cpp:

24035

(MethodList::addMethod):

24036

(MethodList::length):

24037

(MethodList::methodAt):

24038

(MethodList::~MethodList):

24039

* bindings/runtime.h:

24040

(KJS::Bindings::MethodList::MethodList):

24041

* bindings/runtime_method.cpp:

24042

(RuntimeMethodImp::RuntimeMethodImp):

24043

(RuntimeMethodImp::get):

24044

(RuntimeMethodImp::call):

24045

* bindings/runtime_method.h:

24046

* bindings/runtime_object.cpp:

24047

(RuntimeObjectImp::get):

24048

(RuntimeObjectImp::hasProperty):

24049

24050

2004-01-16 Richard Williamson <rjw@apple.com>

24051

24052

Fixed 3531229. Another place that needs the Push/PopLocalFrame

24053

protection implemented for 3530401.

24054

24055

Reviewed by John.

24056

24057

* bindings/runtime_method.cpp:

24058

(RuntimeMethodImp::call):

24059

24060

2004-01-15 Richard Williamson <rjw@apple.com>

24061

24062

Fixed 3530401. JNI doesn't cleanup local refs created on the

24063

main thread. IMO this is a bad bug in our JMI implementation.

24064

24065

To work-around the problem I explicitly delete all local refs.

24066

Further, I've added Push/PopLocalFrame calls to catch any refs

24067

that I may have missed. This will guarantee that we don't leak

24068

any Java references.

24069

24070

Reviewed by John.

24071

24072

* bindings/jni/jni_class.cpp:

24073

(JavaClass::_commonInit):

24074

(JavaClass::JavaClass):

24075

* bindings/jni/jni_instance.cpp:

24076

(JavaInstance::begin):

24077

(JavaInstance::end):

24078

* bindings/jni/jni_instance.h:

24079

* bindings/jni/jni_runtime.cpp:

24080

(JavaConstructor::JavaConstructor):

24081

(JavaMethod::JavaMethod):

24082

* bindings/jni_jsobject.cpp:

24083

(JSObject::listFromJArray):

24084

* bindings/runtime.h:

24085

(KJS::Bindings::Instance::begin):

24086

(KJS::Bindings::Instance::end):

24087

* bindings/runtime_object.cpp:

24088

(RuntimeObjectImp::get):

24089

(RuntimeObjectImp::put):

24090

(RuntimeObjectImp::canPut):

24091

(RuntimeObjectImp::hasProperty):

24092

(RuntimeObjectImp::defaultValue):

24093

24094

2004-01-15 Vicki Murley <vicki@apple.com>

24095

24096

Reviewed by Darin.

24097

24098

24099

24100

2004-01-14 Richard Williamson <rjw@apple.com>

24101

24102

Fixed 3529466. With recent changes to Java plugin we must no

24103

longer call DeleteLocalRef(). Not a problem, it was an optimization anyway.

24104

24105

Reviewed by John.

24106

24107

* bindings/jni/jni_instance.cpp:

24108

(JObjectWrapper::JObjectWrapper):

24109

24110

=== Safari-122 ===

24111

24112

2004-01-14 Richard Williamson <rjw@apple.com>

24113

24114

Fixed 3529010.

24115

24116

Finalize may be called on an JSObject after we've already remove all our references. The assert in this case is firing because we've received a finalize call from Java for an instance that we no longer know about. The fix is to check in finalize that we're getting a call on an instance that we still care about.

24117

24118

Reviewed by John.

24119

24120

* bindings/jni_jsobject.cpp:

24121

(addJavaReference):

24122

(removeJavaReference):

24123

(RootObject::removeAllJavaReferencesForRoot):

24124

(JSObject::invoke):

24125

24126

2004-01-13 Richard Williamson <rjw@apple.com>

24127

24128

Fixed 3528324.

24129

24130

The run loop that is used to execute JavaScript (in practice, always the main run loop) is held in a class variable. It is set and retained once and should not be released. Unfortunately is it being released when the 'root' object on a LiveConnect applet is released. This has the symptom of eventually causing an deallocation of the main run loop! Usually after about 5 instantiations/destructions of a LiveConnect applet. The CFRelease of the run loop was removed.

24131

24132

Reviewed by Hyatt.

24133

24134

* bindings/jni_jsobject.h:

24135

(KJS::Bindings::RootObject::~RootObject):

24136

24137

=== Safari-121 ===

24138

24139

=== Safari-120 ===

24140

24141

2004-01-06 Richard Williamson <rjw@apple.com>

24142

24143

Fixed 3521814. Finalize messages weren't being dispatched!

24144

24145

Reviewed by John.

24146

24147

* bindings/jni_jsobject.cpp:

24148

(JSObject::invoke):

24149

24150

2004-01-05 Richard Williamson <rjw@apple.com>

24151

24152

Added cache of JNI method IDs to minimize allocations. This mitigates the problem

24153

described by 3515579.

24154

24155

Also cleanup up logging of Java exceptions.

24156

24157

Reviewed by John.

24158

24159

* bindings/jni/jni_class.cpp:

24160

(JavaClass::classForInstance):

24161

* bindings/jni/jni_instance.cpp:

24162

(JavaInstance::JavaInstance):

24163

(JavaInstance::getClass):

24164

(JavaInstance::invokeMethod):

24165

(JObjectWrapper::JObjectWrapper):

24166

(JObjectWrapper::~JObjectWrapper):

24167

* bindings/jni/jni_instance.h:

24168

(KJS::Bindings::JavaInstance::operator=):

24169

* bindings/jni/jni_runtime.cpp:

24170

(JavaMethod::JavaMethod):

24171

(JavaMethod::methodID):

24172

* bindings/jni/jni_runtime.h:

24173

(KJS::Bindings::JavaMethod::JavaMethod):

24174

* bindings/jni/jni_utility.cpp:

24175

(callJNIMethod):

24176

(callJNIMethodIDA):

24177

(callJNIMethodA):

24178

(KJS::Bindings::getMethodID):

24179

(KJS::Bindings::callJNIVoidMethodIDA):

24180

(KJS::Bindings::callJNIObjectMethodIDA):

24181

(KJS::Bindings::callJNIByteMethodIDA):

24182

(KJS::Bindings::callJNICharMethodIDA):

24183

(KJS::Bindings::callJNIShortMethodIDA):

24184

(KJS::Bindings::callJNIIntMethodIDA):

24185

(KJS::Bindings::callJNILongMethodIDA):

24186

(KJS::Bindings::callJNIFloatMethodIDA):

24187

(KJS::Bindings::callJNIDoubleMethodIDA):

24188

(KJS::Bindings::callJNIBooleanMethodIDA):

24189

(KJS::Bindings::getCharactersFromJStringInEnv):

24190

(KJS::Bindings::getUCharactersFromJStringInEnv):

24191

(KJS::Bindings::getJNIField):

24192

* bindings/jni/jni_utility.h:

24193

24194

l2003-12-23 John Sullivan <sullivan@apple.com>

24195

24196

* JavaScriptCore.pbproj/project.pbxproj:

24197

Xcode version wars, harmless

24198

24199

2003-12-23 Darin Adler <darin@apple.com>

24200

24201

Reviewed by John (concept, not code, which is just the old code coming back).

24202

24203

- fixed 3518092: REGRESSION (100-119): getting NaN instead of HH:MM times

24204

24205

* kjs/date_object.cpp: Added back our CF-based implementations of gmtime, localtime,

24206

mktime, timegm, and time, because mktime, at least, won't handle a year of 0.

24207

24208

2003-12-19 Richard Williamson <rjw@apple.com>

24209

24210

Fixed 3515597. When an error occurs we need

24211

to make sure result values are zeroed.

24212

24213

Cleaned up logs by adding a newline.

24214

24215

Reviewed by John.

24216

24217

* bindings/jni/jni_utility.cpp:

24218

(KJS::Bindings::getJavaVM):

24219

(KJS::Bindings::getJNIEnv):

24220

(callJNIMethod):

24221

(callJNIMethodA):

24222

(KJS::Bindings::getJNIField):

24223

* bindings/jni_jsobject.cpp:

24224

(JSObject::convertValueToJObject):

24225

24226

=== Safari-119 ===

24227

24228

2003-12-17 Richard Williamson <rjw@apple.com>

24229

24230

Ensure that all the symbols we export are in the KJS

24231

namespace (3512245).

24232

24233

Also renamed JavaString.characters() to JavaString.UTF8String()

24234

for enhanced clarity.

24235

24236

Added some sanity checking to constructor of JObjectWrapper.

24237

24238

Reviewed by Dave.

24239

24240

* ChangeLog:

24241

* bindings/jni/jni_class.cpp:

24242

* bindings/jni/jni_class.h:

24243

* bindings/jni/jni_instance.cpp:

24244

(JavaInstance::invokeMethod):

24245

(JObjectWrapper::JObjectWrapper):

24246

* bindings/jni/jni_instance.h:

24247

* bindings/jni/jni_runtime.cpp:

24248

(JavaParameter::JavaParameter):

24249

(JavaField::JavaField):

24250

(JavaMethod::JavaMethod):

24251

(JavaMethod::signature):

24252

* bindings/jni/jni_runtime.h:

24253

(KJS::Bindings::JavaString::ascii):

24254

(KJS::Bindings::JavaString::UTF8String):

24255

(KJS::Bindings::JavaString::JavaString):

24256

(KJS::Bindings::JavaString::_commonInit):

24257

(KJS::Bindings::JavaString::uchars):

24258

(KJS::Bindings::JavaString::length):

24259

(KJS::Bindings::JavaString::ustring):

24260

(KJS::Bindings::JavaParameter::type):

24261

(KJS::Bindings::JavaField::name):

24262

(KJS::Bindings::JavaField::type):

24263

(KJS::Bindings::JavaMethod::name):

24264

(KJS::Bindings::JavaMethod::returnType):

24265

* bindings/jni/jni_utility.cpp:

24266

(KJS::Bindings::getJavaVM):

24267

(KJS::Bindings::getJNIEnv):

24268

(KJS::Bindings::callJNIVoidMethod):

24269

(KJS::Bindings::callJNIObjectMethod):

24270

(KJS::Bindings::callJNIBooleanMethod):

24271

(KJS::Bindings::callJNIByteMethod):

24272

(KJS::Bindings::callJNICharMethod):

24273

(KJS::Bindings::callJNIShortMethod):

24274

(KJS::Bindings::callJNIIntMethod):

24275

(KJS::Bindings::callJNILongMethod):

24276

(KJS::Bindings::callJNIFloatMethod):

24277

(KJS::Bindings::callJNIDoubleMethod):

24278

(KJS::Bindings::callJNIVoidMethodA):

24279

(KJS::Bindings::callJNIObjectMethodA):

24280

(KJS::Bindings::callJNIByteMethodA):

24281

(KJS::Bindings::callJNICharMethodA):

24282

(KJS::Bindings::callJNIShortMethodA):

24283

(KJS::Bindings::callJNIIntMethodA):

24284

(KJS::Bindings::callJNILongMethodA):

24285

(KJS::Bindings::callJNIFloatMethodA):

24286

(KJS::Bindings::callJNIDoubleMethodA):

24287

(KJS::Bindings::callJNIBooleanMethodA):

24288

(KJS::Bindings::getCharactersFromJString):

24289

(KJS::Bindings::releaseCharactersForJString):

24290

(KJS::Bindings::getCharactersFromJStringInEnv):

24291

(KJS::Bindings::releaseCharactersForJStringInEnv):

24292

(KJS::Bindings::getUCharactersFromJStringInEnv):

24293

(KJS::Bindings::releaseUCharactersForJStringInEnv):

24294

(KJS::Bindings::JNITypeFromClassName):

24295

(KJS::Bindings::signatureFromPrimitiveType):

24296

(KJS::Bindings::JNITypeFromPrimitiveType):

24297

(KJS::Bindings::getJNIField):

24298

(KJS::Bindings::convertValueToJValue):

24299

* bindings/jni/jni_utility.h:

24300

* bindings/jni_jsobject.cpp:

24301

(KJS::Bindings::JSObject::invoke):

24302

(KJS::Bindings::JSObject::JSObject):

24303

(KJS::Bindings::JSObject::call):

24304

(KJS::Bindings::JSObject::eval):

24305

(KJS::Bindings::JSObject::getMember):

24306

(KJS::Bindings::JSObject::setMember):

24307

(KJS::Bindings::JSObject::removeMember):

24308

(KJS::Bindings::JSObject::getSlot):

24309

(KJS::Bindings::JSObject::setSlot):

24310

(KJS::Bindings::JSObject::toString):

24311

(KJS::Bindings::JSObject::finalize):

24312

(KJS::Bindings::JSObject::createNative):

24313

(KJS::Bindings::JSObject::convertValueToJObject):

24314

(KJS::Bindings::JSObject::convertJObjectToValue):

24315

(KJS::Bindings::JSObject::listFromJArray):

24316

* bindings/jni_jsobject.h:

24317

* bindings/runtime.cpp:

24318

* bindings/runtime.h:

24319

* bindings/runtime_method.cpp:

24320

* bindings/runtime_method.h:

24321

24322

=== Safari-118 ===

24323

24324

2003-12-16 Richard Williamson <rjw@apple.com>

24325

24326

Ack! More assertions. Lock ALL entry points into the interpreter!

24327

(3511733).

24328

24329

Reviewed by Ken.

24330

24331

* bindings/jni_jsobject.cpp:

24332

(Bindings::JSObject::call):

24333

(Bindings::JSObject::eval):

24334

(Bindings::JSObject::getMember):

24335

(Bindings::JSObject::setMember):

24336

(Bindings::JSObject::removeMember):

24337

(Bindings::JSObject::getSlot):

24338

(Bindings::JSObject::setSlot):

24339

(Bindings::JSObject::convertJObjectToValue):

24340

24341

2003-12-15 Richard Williamson <rjw@apple.com>

24342

24343

Fixed a couple of snafus and removed some logging.

24344

24345

Reviewed by Maciej.

24346

24347

* bindings/jni_jsobject.cpp:

24348

(Bindings::performJavaScriptAccess):

24349

(Bindings::completedJavaScriptAccess):

24350

(Bindings::dispatchToJavaScriptThread):

24351

Removed some annoying JS_LOG clutter.

24352

24353

(Bindings::RootObject::removeAllJavaReferencesForRoot):

24354

Fixed allocation of key buffer that was called after it was needed.

24355

24356

(Bindings::JSObject::invoke):

24357

(Bindings::JSObject::JSObject):

24358

(Bindings::JSObject::getMember):

24359

(Bindings::JSObject::getSlot):

24360

Added additional interpreter locks around getMember and getSlot.

24361

These functions may cause allocation of JS impls.

24362

24363

2003-12-15 Richard Williamson <rjw@apple.com>

24364

24365

args weren't passed to 'call' invocation. d'oh.

24366

lock interpreter when we create instances of JS impls.

24367

24368

Reviewed by Maciej.

24369

24370

* bindings/jni_jsobject.cpp:

24371

(Bindings::JSObject::call):

24372

(Bindings::JSObject::eval):

24373

(Bindings::JSObject::getMember):

24374

(Bindings::JSObject::setMember):

24375

(Bindings::JSObject::getSlot):

24376

(Bindings::JSObject::convertValueToJObject):

24377

(Bindings::JSObject::convertJObjectToValue):

24378

(Bindings::JSObject::listFromJArray):

24379

* bindings/jni_jsobject.h:

24380

24381

2003-12-15 Richard Williamson <rjw@apple.com>

24382

24383

Last piece of LiveConnect! This checkin adds implementation

24384

of the Java to JavaScript object conversion functions.

24385

24386

Reviewed by John.

24387

24388

* bindings/jni/jni_instance.cpp:

24389

(JavaInstance::invokeMethod):

24390

* bindings/jni/jni_utility.cpp:

24391

* bindings/jni/jni_utility.h:

24392

* bindings/jni_jsobject.cpp:

24393

(Bindings::JSObject::invoke):

24394

(Bindings::JSObject::call):

24395

(Bindings::JSObject::eval):

24396

(Bindings::JSObject::getMember):

24397

(Bindings::JSObject::setMember):

24398

(Bindings::JSObject::getSlot):

24399

(Bindings::JSObject::setSlot):

24400

(Bindings::JSObject::createNative):

24401

(Bindings::JSObject::convertValueToJObject):

24402

(Bindings::JSObject::convertJObjectToValue):

24403

(Bindings::JSObject::listFromJArray):

24404

* bindings/jni_jsobject.h:

24405

(Bindings::):

24406

* bindings/runtime_method.cpp:

24407

(RuntimeMethodImp::get):

24408

(RuntimeMethodImp::codeType):

24409

(RuntimeMethodImp::execute):

24410

24411

2003-12-12 Richard Williamson <rjw@apple.com>

24412

24413

Added implementation of stubs in JSObject. All that

24414

remains is a couple of simple conversion functions stubs and

24415

we're done with LiveConnect. Also, changed string passing to

24416

JS to use uchars instead of chars.

24417

24418

Reviewed by Maciej.

24419

24420

* bindings/jni/jni_runtime.h:

24421

(Bindings::JavaString::JavaString):

24422

(Bindings::JavaString::_commonInit):

24423

(Bindings::JavaString::_commonCopy):

24424

(Bindings::JavaString::_commonDelete):

24425

(Bindings::JavaString::~JavaString):

24426

(Bindings::JavaString::operator=):

24427

(Bindings::JavaString::uchars):

24428

(Bindings::JavaString::length):

24429

(Bindings::JavaString::ustring):

24430

* bindings/jni/jni_utility.cpp:

24431

(getUCharactersFromJStringInEnv):

24432

(releaseUCharactersForJStringInEnv):

24433

(convertValueToJObject):

24434

(convertJObjectToValue):

24435

* bindings/jni/jni_utility.h:

24436

* bindings/jni_jsobject.cpp:

24437

(Bindings::JSObject::invoke):

24438

(Bindings::JSObject::call):

24439

(Bindings::JSObject::eval):

24440

(Bindings::JSObject::getMember):

24441

(Bindings::JSObject::setMember):

24442

(Bindings::JSObject::removeMember):

24443

(Bindings::JSObject::getSlot):

24444

(Bindings::JSObject::setSlot):

24445

* bindings/jni_jsobject.h:

24446

24447

2003-12-12 Richard Williamson <rjw@apple.com>

24448

24449

Ensure that all calls from Java into JavaScript are

24450

performed on a designated thread (the main thread).

24451

24452

Reviewed by Ken.

24453

24454

* bindings/jni_jsobject.cpp:

24455

(isJavaScriptThread):

24456

(rootForImp):

24457

(Bindings::performJavaScriptAccess):

24458

(Bindings::completedJavaScriptAccess):

24459

(Bindings::initializeJavaScriptAccessLock):

24460

(Bindings::lockJavaScriptAccess):

24461

(Bindings::unlockJavaScriptAccess):

24462

(Bindings::dispatchToJavaScriptThread):

24463

(Bindings::RootObject::setFindRootObjectForNativeHandleFunction):

24464

(Bindings::RootObject::removeAllJavaReferencesForRoot):

24465

(Bindings::JSObject::invoke):

24466

(Bindings::JSObject::JSObject):

24467

(Bindings::JSObject::call):

24468

(Bindings::JSObject::eval):

24469

(Bindings::JSObject::getMember):

24470

(Bindings::JSObject::setMember):

24471

(Bindings::JSObject::removeMember):

24472

(Bindings::JSObject::getSlot):

24473

(Bindings::JSObject::setSlot):

24474

(Bindings::JSObject::toString):

24475

(Bindings::JSObject::finalize):

24476

(Bindings::JSObject::getWindow):

24477

* bindings/jni_jsobject.h:

24478

(Bindings::RootObject::~RootObject):

24479

(Bindings::RootObject::findRootObjectForNativeHandleFunction):

24480

(Bindings::RootObject::runLoop):

24481

(Bindings::RootObject::performJavaScriptSource):

24482

(Bindings::):

24483

24484

2003-12-11 Richard Williamson <rjw@apple.com>

24485

24486

Added support for calling a JavaScript function from

24487

Java. Right now this only works for void func(void)

24488

functions, but the conversion of args and return values

24489

will come shortly.

24490

24491

Cleaned up and verified reference counting scheme, and

24492

dereferencing of vended JavaScript objects when applet is

24493

destroyed (actually when part is destroyed).

24494

24495

Removed link hack for testkjs now that the Java folks think

24496

they have a solution for the 1.4.2 JavaVM link problem. Although

24497

Greg B. thinks his solution may cause problems for the 1.3.1

24498

version of the VM!?!

24499

24500

Reviewed by Ken.

24501

24502

* Makefile.am:

24503

* bindings/jni/jni_runtime.h:

24504

(Bindings::JavaString::JavaString):

24505

* bindings/jni/jni_utility.cpp:

24506

(convertValueToJValue):

24507

(convertValueToJObject):

24508

(listFromJArray):

24509

* bindings/jni/jni_utility.h:

24510

* bindings/jni_jsobject.cpp:

24511

(KJS_setFindRootObjectForNativeHandleFunction):

24512

(KJS_findRootObjectForNativeHandleFunction):

24513

(getReferencesByRootDictionary):

24514

(getReferencesDictionary):

24515

(findReferenceDictionary):

24516

(rootForImp):

24517

(addJavaReference):

24518

(removeJavaReference):

24519

* bindings/jni_jsobject.h:

24520

(Bindings::RootObject::RootObject):

24521

(Bindings::RootObject::~RootObject):

24522

(Bindings::RootObject::setRootObjectImp):

24523

(Bindings::RootObject::rootObjectImp):

24524

(Bindings::RootObject::setInterpreter):

24525

(Bindings::RootObject::interpreter):

24526

24527

=== Safari-117 ===

24528

24529

2003-12-10 Darin Adler <darin@apple.com>

24530

24531

Reviewed by Maciej.

24532

24533

- fixed regression in JavaScript tests reported by the KDE guys

24534

- fixed 3506345: REGRESSION (115-116): VIP: chordfind.com no longer displays chords

24535

24536

* kjs/ustring.h: Add tolerateEmptyString parameter to toDouble and toULong.

24537

* kjs/ustring.cpp:

24538

(KJS::UString::toDouble): Separate the "tolerant" parameter into two separate ones:

24539

tolerateTrailingJunk and tolerateEmptyString. Add new overloads; better for code size

24540

and binary compatibility than default parameter values.

24541

(KJS::UString::toULong): Pass tolerateEmptyString down to toDouble. Add new overload.

24542

24543

* kjs/string_object.cpp: (StringProtoFuncImp::call): Pass false for the new

24544

"tolerate empty string" parameter.

24545

24546

2003-12-10 Richard Williamson <rjw@apple.com>

24547

24548

Added code to manage reference counting of JavaScript

24549

objects passed to Java. Also added implementation of

24550

KJS_JSCreateNativeJSObject. This is the function that

24551

provides the root object to Java (KJS::Window).

24552

24553

Reviewed by Hyatt.

24554

24555

* JavaScriptCore.pbproj/project.pbxproj:

24556

* bindings/jni_jsobject.cpp:

24557

(KJS_setFindObjectForNativeHandleFunction):

24558

(KJS_findObjectForNativeHandleFunction):

24559

(getReferencesByOwnerDictionary):

24560

(getReferencesDictionary):

24561

(findReferenceDictionary):

24562

(addJavaReference):

24563

(removeJavaReference):

24564

(removeAllJavaReferencesForOwner):

24565

* bindings/jni_jsobject.h:

24566

24567

2003-12-09 Richard Williamson <rjw@apple.com>

24568

24569

LiveConnect stubs that correspond to the native methods

24570

on JSObject. These will be called from the new Java plugin

24571

when an instance of JSObject is instantiated and messaged.

24572

When these are implemented the Java will be able to originate

24573

calls into JavaScript.

24574

24575

Also a temporary work-around added to Makefile.am to solve

24576

a link problem. The 1.4.2 JavaVM accidentally links against

24577

libobjc. This call a failure linking testkjs. Mike Hay is

24578

working with someone to fix the problem (3505587).

24579

24580

Reviewed by Chris.

24581

24582

* JavaScriptCore.pbproj/project.pbxproj:

24583

* Makefile.am:

24584

* bindings/jni_jsobject.cpp: Added.

24585

(KJS_JSCreateNativeJSObject):

24586

(KJS_JSObject_JSFinalize):

24587

(KJS_JSObject_JSObjectCall):

24588

(KJS_JSObject_JSObjectEval):

24589

(KJS_JSObject_JSObjectGetMember):

24590

(KJS_JSObject_JSObjectSetMember):

24591

(KJS_JSObject_JSObjectRemoveMember):

24592

(KJS_JSObject_JSObjectGetSlot):

24593

(KJS_JSObject_JSObjectSetSlot):

24594

(KJS_JSObject_JSObjectToString):

24595

* bindings/jni_jsobject.h: Added.

24596

24597

2003-12-09 Maciej Stachowiak <mjs@apple.com>

24598

24599

Reviewed by John.

24600

24601

<rdar://problem/3505183>: JavaScriptCore should assert that interpreter is locked in collector

24602

24603

* kjs/collector.cpp:

24604

(KJS::Collector::allocate): Assert that interpreter lock count is not 0.

24605

(KJS::Collector::collect): likewise

24606

24607

2003-12-08 Richard Williamson <rjw@apple.com>

24608

24609

LiveConnect: The last piece of the JavaScript side of the

24610

LiveConnect implementation. This change adds support for

24611

setting/getting values from Java arrays in JavaScript.

24612

24613

Reviewed by John.

24614

24615

* bindings/jni/jni_instance.h:

24616

* bindings/jni/jni_runtime.cpp:

24617

(JavaField::JavaField):

24618

(convertJObjectToArray):

24619

(JavaArray::JavaArray):

24620

(JavaArray::~JavaArray):

24621

(JavaArray::setValueAt):

24622

(JavaArray::valueAt):

24623

(JavaArray::getLength):

24624

* bindings/jni/jni_runtime.h:

24625

(Bindings::JavaArray::operator=):

24626

(Bindings::JavaArray::javaArray):

24627

* bindings/jni/jni_utility.cpp:

24628

(JNITypeFromPrimitiveType):

24629

(convertValueToJValue):

24630

* bindings/jni/jni_utility.h:

24631

* bindings/runtime.h:

24632

* bindings/runtime_array.cpp:

24633

(RuntimeArrayImp::RuntimeArrayImp):

24634

(RuntimeArrayImp::~RuntimeArrayImp):

24635

(RuntimeArrayImp::get):

24636

(RuntimeArrayImp::put):

24637

(RuntimeArrayImp::hasProperty):

24638

* bindings/runtime_array.h:

24639

(KJS::RuntimeArrayImp::getLength):

24640

(KJS::RuntimeArrayImp::getConcreteArray):

24641

* bindings/runtime_object.cpp:

24642

(RuntimeObjectImp::get):

24643

(RuntimeObjectImp::canPut):

24644

(RuntimeObjectImp::hasProperty):

24645

24646

2003-12-05 Richard Williamson <rjw@apple.com>

24647

24648

LiveConnect: Part 1 of supporting JS bindings to

24649

native language arrays.

24650

24651

Reviewed by Chris.

24652

24653

* JavaScriptCore.pbproj/project.pbxproj:

24654

* bindings/jni/jni_runtime.cpp:

24655

(JavaField::JavaField):

24656

(convertJObjectToArray):

24657

(JavaField::valueFromInstance):

24658

(JavaField::setValueToInstance):

24659

* bindings/jni/jni_runtime.h:

24660

* bindings/runtime.cpp:

24661

(Instance::setValueOfField):

24662

* bindings/runtime.h:

24663

(Bindings::Array::~Array):

24664

24665

2003-12-04 Richard Williamson <rjw@apple.com>

24666

24667

LiveConnect: Moved defaultValue into concrete implementation because

24668

more intelligent conversion can be perform with knowledge

24669

of the class of the original instance.

24670

24671

Reviewed by Chris.

24672

24673

* bindings/jni/jni_class.cpp:

24674

(JavaClass::isNumberClass):

24675

(JavaClass::isBooleanClass):

24676

(JavaClass::isStringClass):

24677

* bindings/jni/jni_class.h:

24678

* bindings/jni/jni_instance.cpp:

24679

(JavaInstance::defaultValue):

24680

(JavaInstance::valueOf):

24681

* bindings/jni/jni_instance.h:

24682

(Bindings::JavaInstance::javaInstance):

24683

* bindings/runtime.h:

24684

* bindings/runtime_object.cpp:

24685

(RuntimeObjectImp::defaultValue):

24686

24687

2003-12-04 Richard Williamson <rjw@apple.com>

24688

24689

LiveConnect: Added support for setting the value of Java

24690

fields.

24691

24692

Reviewed by Chris.

24693

24694

* bindings/jni/jni_instance.cpp:

24695

(JavaInstance::invokeMethod):

24696

* bindings/jni/jni_runtime.cpp:

24697

(JavaParameter::JavaParameter):

24698

(JavaField::JavaField):

24699

(JavaField::valueFromInstance):

24700

(JavaField::setValueToInstance):

24701

(JavaMethod::JavaMethod):

24702

* bindings/jni/jni_runtime.h:

24703

(Bindings::JavaField::getJNIType):

24704

* bindings/jni/jni_utility.cpp:

24705

(JNITypeFromClassName):

24706

(convertValueToJValue):

24707

* bindings/jni/jni_utility.h:

24708

* bindings/runtime.cpp:

24709

(Instance::setValueOfField):

24710

* bindings/runtime.h:

24711

* bindings/runtime_object.cpp:

24712

(RuntimeObjectImp::get):

24713

(RuntimeObjectImp::put):

24714

(RuntimeObjectImp::defaultValue):

24715

24716

2003-12-04 Richard Williamson <rjw@apple.com>

24717

24718

Added support for string conversions.

24719

Changed various JavaString member variables to be inline.

24720

Implemented defaultValue for context relevant type coercion.

24721

24722

Reviewed by Chris.

24723

24724

* bindings/jni/jni_class.cpp:

24725

(JavaClass::JavaClass):

24726

(JavaClass::setClassName):

24727

(JavaClass::classForInstance):

24728

* bindings/jni/jni_class.h:

24729

* bindings/jni/jni_instance.cpp:

24730

(JavaInstance::stringValue):

24731

(JavaInstance::numberValue):

24732

(JavaInstance::booleanValue):

24733

(JavaInstance::invokeMethod):

24734

* bindings/jni/jni_instance.h:

24735

* bindings/jni/jni_runtime.cpp:

24736

(JavaParameter::JavaParameter):

24737

(JavaField::JavaField):

24738

(JavaMethod::JavaMethod):

24739

(appendClassName):

24740

(JavaMethod::signature):

24741

* bindings/jni/jni_runtime.h:

24742

(Bindings::JavaString::JavaString):

24743

(Bindings::JavaString::~JavaString):

24744

(Bindings::JavaString::operator=):

24745

(Bindings::JavaString::characters):

24746

(Bindings::JavaParameter::JavaParameter):

24747

(Bindings::JavaParameter::~JavaParameter):

24748

(Bindings::JavaParameter::operator=):

24749

(Bindings::JavaParameter::type):

24750

(Bindings::JavaField::JavaField):

24751

(Bindings::JavaField::~JavaField):

24752

(Bindings::JavaField::operator=):

24753

(Bindings::JavaField::name):

24754

(Bindings::JavaField::type):

24755

(Bindings::JavaMethod::JavaMethod):

24756

(Bindings::JavaMethod::_commonDelete):

24757

(Bindings::JavaMethod::name):

24758

(Bindings::JavaMethod::returnType):

24759

* bindings/jni/jni_utility.cpp:

24760

(convertValueToJValue):

24761

* bindings/runtime.h:

24762

(Bindings::Instance::valueOf):

24763

* bindings/runtime_method.cpp:

24764

(RuntimeMethodImp::call):

24765

* bindings/runtime_object.cpp:

24766

(RuntimeObjectImp::RuntimeObjectImp):

24767

(RuntimeObjectImp::get):

24768

(RuntimeObjectImp::defaultValue):

24769

* bindings/runtime_object.h:

24770

(KJS::RuntimeObjectImp::classInfo):

24771

24772

=== Safari-116 ===

24773

24774

2003-12-03 Richard Williamson <rjw@apple.com>

24775

24776

LiveConnect: Added support for parameter passing to Java and conversion

24777

of return values.

24778

24779

Reviewed by Chris.

24780

24781

* bindings/jni/jni_instance.cpp:

24782

(JavaInstance::invokeMethod):

24783

* bindings/jni/jni_instance.h:

24784

* bindings/jni/jni_runtime.cpp:

24785

(JavaParameter::JavaParameter):

24786

(JavaMethod::JavaMethod):

24787

(JavaMethod::signature):

24788

* bindings/jni/jni_runtime.h:

24789

(Bindings::JavaParameter::JavaParameter):

24790

(Bindings::JavaParameter::operator=):

24791

(Bindings::JavaParameter::getJNIType):

24792

* bindings/jni/jni_utility.cpp:

24793

(callJNIBooleanMethodA):

24794

(convertValueToJValue):

24795

* bindings/jni/jni_utility.h:

24796

* bindings/runtime.h:

24797

* bindings/runtime_method.cpp:

24798

(RuntimeMethodImp::call):

24799

* bindings/runtime_object.cpp:

24800

(RuntimeObjectImp::get):

24801

24802

2003-12-02 Richard Williamson <rjw@apple.com>

24803

24804

Added support for calling simple methods in Java from JavaScript.

24805

(void return and no parameters). Yay, LiveConnect lives.

24806

24807

Still need write argument and return value conversion code.

24808

24809

Reviewed by Chris.

24810

24811

* JavaScriptCore.pbproj/project.pbxproj:

24812

* bindings/jni/jni_instance.cpp:

24813

(JavaInstance::getClass):

24814

(JavaInstance::invokeMethod):

24815

* bindings/jni/jni_instance.h:

24816

* bindings/jni/jni_runtime.cpp:

24817

(JavaMethod::JavaMethod):

24818

(JavaMethod::signature):

24819

(JavaMethod::JNIReturnType):

24820

* bindings/jni/jni_runtime.h:

24821

(Bindings::JavaMethod::_commonDelete):

24822

(Bindings::JavaMethod::_commonCopy):

24823

(Bindings::JavaMethod::name):

24824

* bindings/jni/jni_utility.cpp:

24825

(signatureFromPrimitiveType):

24826

* bindings/jni/jni_utility.h:

24827

* bindings/runtime.h:

24828

* bindings/runtime_method.cpp: Added.

24829

(RuntimeMethodImp::RuntimeMethodImp):

24830

(RuntimeMethodImp::~RuntimeMethodImp):

24831

(RuntimeMethodImp::get):

24832

(RuntimeMethodImp::implementsCall):

24833

(RuntimeMethodImp::call):

24834

(RuntimeMethodImp::codeType):

24835

(RuntimeMethodImp::execute):

24836

* bindings/runtime_method.h: Added.

24837

* bindings/runtime_object.cpp:

24838

(RuntimeObjectImp::RuntimeObjectImp):

24839

(RuntimeObjectImp::get):

24840

* bindings/runtime_object.h:

24841

* kjs/function.cpp:

24842

(FunctionImp::FunctionImp):

24843

* kjs/interpreter.h:

24844

24845

2003-12-01 Darin Adler <darin@apple.com>

24846

24847

Reviewed by Maciej.

24848

24849

- fixed 3493799: JavaScript string.replace expands $ if it's the last character in replacement string

24850

24851

* kjs/ustring.cpp: (KJS::UString::toDouble): Fix backwards handling of the "tolerant" boolean.

24852

This indirectly caused the string.replace bug.

24853

24854

2003-12-02 Maciej Stachowiak <mjs@apple.com>

24855

24856

Merged patches from Harri Porten and David Faure to fix:

24857

24858

<rdar://problem/3497643>: reproducible crash printing self-referential array

24859

24860

* kjs/array_object.cpp:

24861

(ArrayProtoFuncImp::call): Break out of the loop if an exception was thrown.

24862

* kjs/nodes.cpp:

24863

(FunctionCallNode::evaluate): Move function call depth check from here...

24864

* kjs/object.cpp:

24865

(KJS::Object::call): ...to here.

24866

* kjs/object.h: Un-inline Object::call now that it does more.

24867

24868

2003-12-01 Richard Williamson <rjw@apple.com>

24869

24870

Fixed mistake in method signatures used to get boolean and integer fields.

24871

24872

Reviewed by Chris.

24873

24874

* bindings/jni/jni_runtime.cpp:

24875

(JavaField::valueFromInstance):

24876

24877

2003-12-01 Richard Williamson <rjw@apple.com>

24878

24879

Fixed parameter passing to applet. Child elements are NOT valid in setStyle(). So we now create the widget before needed with createWidgetIfNecessary. This either happens when doing the first layout, or when JavaScript first references the applet element.

24880

24881

Fixed early delete of the the main applet instance. When the JS collector cleaned up the last JS object referring to the applet instance we were deleting the java instance. This caused the applet instance cached on the applet element to be invalid. The applet instance is the only Java object not to be cleaned up by the JS collector.

24882

24883

Added support for getting at Java object fields.

24884

24885

Reviewed by Chris.

24886

24887

* JavaScriptCore.pbproj/project.pbxproj:

24888

* Makefile.am:

24889

* bindings/jni/jni_instance.cpp:

24890

(JObjectWrapper::JObjectWrapper):

24891

* bindings/jni/jni_instance.h:

24892

(Bindings::JObjectWrapper::~JObjectWrapper):

24893

* bindings/jni/jni_runtime.cpp:

24894

(JavaField::valueFromInstance):

24895

* bindings/runtime_object.cpp:

24896

(RuntimeObjectImp::~RuntimeObjectImp):

24897

(RuntimeObjectImp::RuntimeObjectImp):

24898

(RuntimeObjectImp::get):

24899

(RuntimeObjectImp::deleteProperty):

24900

* bindings/runtime_object.h:

24901

24902

=== Safari-115 ===

24903

24904

2003-11-21 Maciej Stachowiak <mjs@apple.com>

24905

24906

Patch from Harri Porten, reviewed by me.

24907

24908

- fixed 3491712 - String slice with negative arguments does not offset from end of string

24909

24910

* kjs/string_object.cpp:

24911

(StringProtoFuncImp::call): Handle negative arguments as offsets from end by

24912

adding length and clamping to [0,length-1].

24913

24914

2003-11-21 Maciej Stachowiak <mjs@apple.com>

24915

24916

Patch from Harri Porten, reviewed by me.

24917

24918

- fixed 3491709 - using Function.apply with a primitive type as the arg list causes crash

24919

24920

* kjs/function_object.cpp:

24921

(FunctionProtoFuncImp::call): Nest parentheses properly.

24922

24923

2003-11-20 Richard Williamson <rjw@apple.com>

24924

24925

More LiveConnect stuff. Primitive Java fields are now

24926

accessible from JavaScript! Yay!

24927

24928

Reviewed by Maciej.

24929

24930

* bindings/jni/jni_class.cpp:

24931

(JavaClass::methodNamed):

24932

(JavaClass::fieldNamed):

24933

* bindings/jni/jni_class.h:

24934

(Bindings::JavaClass::_commonDelete):

24935

* bindings/jni/jni_instance.cpp:

24936

(JavaInstance::JavaInstance):

24937

(JavaInstance::~JavaInstance):

24938

(JavaInstance::getClass):

24939

* bindings/jni/jni_instance.h:

24940

(Bindings::JavaInstance::javaInstance):

24941

* bindings/jni/jni_runtime.cpp:

24942

(JavaField::JavaField):

24943

(JavaField::valueFromInstance):

24944

* bindings/jni/jni_runtime.h:

24945

(Bindings::JavaField::JavaField):

24946

(Bindings::JavaField::~JavaField):

24947

(Bindings::JavaField::operator=):

24948

* bindings/jni/jni_utility.cpp:

24949

(callJNIMethod):

24950

(callJNIMethodA):

24951

(callJNIVoidMethod):

24952

(callJNIObjectMethod):

24953

(callJNIBooleanMethod):

24954

(callJNIByteMethod):

24955

(callJNICharMethod):

24956

(callJNIShortMethod):

24957

(callJNIIntMethod):

24958

(callJNILongMethod):

24959

(callJNIFloatMethod):

24960

(callJNIDoubleMethod):

24961

(callJNIVoidMethodA):

24962

(callJNIObjectMethodA):

24963

(callJNIByteMethodA):

24964

(callJNICharMethodA):

24965

(callJNIShortMethodA):

24966

(callJNIIntMethodA):

24967

(callJNILongMethodA):

24968

(callJNIFloatMethodA):

24969

(callJNIDoubleMethodA):

24970

(releaseCharactersForJStringInEnv):

24971

(primitiveTypeFromClassName):

24972

(getJNIField):

24973

* bindings/jni/jni_utility.h:

24974

* bindings/runtime.cpp:

24975

(Instance::createBindingForLanguageInstance):

24976

(Instance::getValueOfField):

24977

* bindings/runtime.h:

24978

* bindings/runtime_object.cpp:

24979

(RuntimeObjectImp::get):

24980

24981

2003-11-20 Richard Williamson <rjw@apple.com>

24982

24983

More LiveConnect stuff.

24984

24985

Reviewed by Chris.

24986

24987

* bindings/jni/jni_class.cpp:

24988

(JavaClass::classForName):

24989

(JavaClass::classForInstance):

24990

* bindings/jni/jni_instance.cpp:

24991

(JavaInstance::getValueOfField):

24992

* bindings/jni/jni_instance.h:

24993

(Bindings::JObjectWrapper::JObjectWrapper):

24994

* bindings/jni/jni_runtime.h:

24995

(Bindings::JavaConstructor::~JavaConstructor):

24996

(Bindings::JavaConstructor::operator=):

24997

(Bindings::JavaMethod::JavaMethod):

24998

(Bindings::JavaMethod::_commonDelete):

24999

(Bindings::JavaMethod::signature):

25000

* bindings/jni/jni_utility.cpp:

25001

(getJNIEnv):

25002

(attachToJavaVM):

25003

* bindings/jni/jni_utility.h:

25004

* bindings/runtime.h:

25005

* bindings/runtime_object.cpp:

25006

(RuntimeObjectImp::~RuntimeObjectImp):

25007

(RuntimeObjectImp::get):

25008

* bindings/runtime_object.h:

25009

25010

2003-11-19 Richard Williamson <rjw@apple.com>

25011

25012

More LiveConnect stuff.

25013

25014

Reviewed by Ken.

25015

25016

* JavaScriptCore.pbproj/project.pbxproj:

25017

* bindings/jni/jni_class.cpp: Added.

25018

(JavaClass::_commonInit):

25019

(JavaClass::JavaClass):

25020

(_createClassesByNameIfNecessary):

25021

(JavaClass::classForName):

25022

(JavaClass::classForInstance):

25023

(JavaClass::methodNamed):

25024

(JavaClass::fieldNamed):

25025

* bindings/jni/jni_class.h: Added.

25026

(Bindings::JavaClass::_commonDelete):

25027

(Bindings::JavaClass::~JavaClass):

25028

(Bindings::JavaClass::_commonCopy):

25029

(Bindings::JavaClass::JavaClass):

25030

(Bindings::JavaClass::operator=):

25031

(Bindings::JavaClass::name):

25032

(Bindings::JavaClass::constructorAt):

25033

(Bindings::JavaClass::numConstructors):

25034

* bindings/jni/jni_instance.cpp: Added.

25035

(JavaInstance::JavaInstance):

25036

(JavaInstance::~JavaInstance):

25037

* bindings/jni/jni_instance.h: Added.

25038

(Bindings::JObjectWrapper::JObjectWrapper):

25039

(Bindings::JObjectWrapper::~JObjectWrapper):

25040

(Bindings::JObjectWrapper::ref):

25041

(Bindings::JObjectWrapper::deref):

25042

(Bindings::JavaInstance::getClass):

25043

(Bindings::JavaInstance::operator=):

25044

* bindings/jni/jni_runtime.cpp:

25045

(JavaMethod::JavaMethod):

25046

* bindings/jni/jni_runtime.h:

25047

(Bindings::JavaString::JavaString):

25048

(Bindings::JavaString::~JavaString):

25049

(Bindings::JavaString::operator=):

25050

* bindings/jni/jni_utility.cpp:

25051

(getJavaVM):

25052

(getJNIEnv):

25053

(getCharactersFromJString):

25054

(releaseCharactersForJString):

25055

(getCharactersFromJStringInEnv):

25056

(releaseCharactersForJStringInEnv):

25057

* bindings/jni/jni_utility.h:

25058

* bindings/runtime.cpp:

25059

(Instance::createBindingForLanguageInstance):

25060

* bindings/runtime.h:

25061

(Bindings::Instance::):

25062

25063

2003-11-18 Richard Williamson <rjw@apple.com>

25064

25065

More live connect stubs. We're getting close.

25066

25067

Reviewed by Chris.

25068

25069

* JavaScriptCore.pbproj/project.pbxproj:

25070

* bindings/jni/jni_runtime.cpp:

25071

(JavaClass::JavaClass):

25072

(JavaInstance::JavaInstance):

25073

(JavaInstance::~JavaInstance):

25074

* bindings/jni/jni_runtime.h:

25075

(Bindings::JavaConstructor::value):

25076

(Bindings::JavaField::value):

25077

(Bindings::JavaMethod::value):

25078

(Bindings::JavaClass::_commonDelete):

25079

(Bindings::JavaClass::_commonCopy):

25080

(Bindings::JavaClass::methodNamed):

25081

(Bindings::JavaClass::fieldNamed):

25082

(Bindings::JavaInstance::getClass):

25083

* bindings/runtime.cpp: Added.

25084

* bindings/runtime.h:

25085

(Bindings::Instance::~Instance):

25086

* bindings/runtime_object.cpp: Added.

25087

(RuntimeObjectImp::classInfo):

25088

(RuntimeObjectImp::RuntimeObjectImp):

25089

(RuntimeObjectImp::get):

25090

(RuntimeObjectImp::put):

25091

(RuntimeObjectImp::canPut):

25092

(RuntimeObjectImp::hasProperty):

25093

(RuntimeObjectImp::deleteProperty):

25094

(RuntimeObjectImp::defaultValue):

25095

(RuntimeObjectImp::_initializeClassInfoFromInstance):

25096

* bindings/runtime_object.h: Added.

25097

(KJS::RuntimeObjectImp::setInternalInstance):

25098

(KJS::RuntimeObjectImp::getInternalInstance):

25099

* kjs/object.cpp:

25100

(KJS::ObjectImp::get):

25101

(KJS::ObjectImp::hasProperty):

25102

* kjs/value.h:

25103

(KJS::):

25104

25105

2003-11-17 Maciej Stachowiak <mjs@apple.com>

25106

25107

Patch from Harri, reviewed by me.

25108

25109

- fixed 3487375 - backwards array slice causes infinite loop

25110

25111

* kjs/array_object.cpp:

25112

(ArrayProtoFuncImp::call):

25113

25114

2003-11-17 Maciej Stachowiak <mjs@apple.com>

25115

25116

Patch from Harri Porten reviewed by me.

25117

25118

- fixed 3487371 - operator precedence for bitwise or, xor and and is wrong

25119

25120

* kjs/grammar.y: Correct the precedence.

25121

25122

2003-11-16 Maciej Stachowiak <mjs@apple.com>

25123

25124

Reviewed by John.

25125

25126

- fixed 3483829 - JavaScriptCore needs workaround to compile on Merlot

25127

25128

* JavaScriptCore.pbproj/project.pbxproj: Add -Wno-long-double to

25129

warning flags.

25130

25131

=== Safari-114 ===

25132

25133

2003-11-13 Richard Williamson <rjw@apple.com>

25134

25135

Factored common code between copy constructor and assignment operator.

25136

25137

Reviewed by Chris.

25138

25139

* ChangeLog:

25140

* bindings/jni/jni_runtime.h:

25141

(Bindings::JavaConstructor::_commonCopy):

25142

(Bindings::JavaConstructor::JavaConstructor):

25143

(Bindings::JavaConstructor::operator=):

25144

(Bindings::JavaField::type):

25145

* bindings/runtime.h:

25146

25147

2003-11-13 Richard Williamson <rjw@apple.com>

25148

25149

More LiveConnect stuff. This checkin adds abstract classes to model

25150

language runtimes and a JNI based set of concrete implementations for

25151

Java.

25152

25153

Reviewed by Chris.

25154

25155

* JavaScriptCore.pbproj/project.pbxproj:

25156

* Makefile.am:

25157

* bindings/Makefile.am: Removed.

25158

* bindings/jni/Makefile.am: Removed.

25159

* bindings/jni/jni_runtime.cpp: Added.

25160

(JavaField::JavaField):

25161

(JavaConstructor::JavaConstructor):

25162

(JavaMethod::JavaMethod):

25163

(JavaClass::JavaClass):

25164

* bindings/jni/jni_runtime.h: Added.

25165

(Bindings::JavaString::JavaString):

25166

(Bindings::JavaString::~JavaString):

25167

(Bindings::JavaString::operator=):

25168

(Bindings::JavaString::characters):

25169

(Bindings::JavaParameter::JavaParameter):

25170

(Bindings::JavaParameter::~JavaParameter):

25171

(Bindings::JavaParameter::operator=):

25172

(Bindings::JavaParameter::type):

25173

(Bindings::JavaConstructor::JavaConstructor):

25174

(Bindings::JavaConstructor::~JavaConstructor):

25175

(Bindings::JavaConstructor::operator=):

25176

(Bindings::JavaConstructor::parameterAt):

25177

(Bindings::JavaConstructor::numParameters):

25178

(Bindings::JavaField::JavaField):

25179

(Bindings::JavaField::~JavaField):

25180

(Bindings::JavaField::operator=):

25181

(Bindings::JavaField::name):

25182

(Bindings::JavaField::type):

25183

(Bindings::JavaMethod::JavaMethod):

25184

(Bindings::JavaMethod::_commonDelete):

25185

(Bindings::JavaMethod::~JavaMethod):

25186

(Bindings::JavaMethod::_commonCopy):

25187

(Bindings::JavaMethod::operator=):

25188

(Bindings::JavaMethod::name):

25189

(Bindings::JavaMethod::returnType):

25190

(Bindings::JavaMethod::parameterAt):

25191

(Bindings::JavaMethod::numParameters):

25192

(Bindings::JavaClass::_commonDelete):

25193

(Bindings::JavaClass::~JavaClass):

25194

(Bindings::JavaClass::_commonCopy):

25195

(Bindings::JavaClass::JavaClass):

25196

(Bindings::JavaClass::operator=):

25197

(Bindings::JavaClass::name):

25198

(Bindings::JavaClass::methodAt):

25199

(Bindings::JavaClass::numMethods):

25200

(Bindings::JavaClass::constructorAt):

25201

(Bindings::JavaClass::numConstructors):

25202

(Bindings::JavaClass::fieldAt):

25203

(Bindings::JavaClass::numFields):

25204

* bindings/jni/jni_utility.cpp:

25205

(callJNIMethod):

25206

(callJNIMethodA):

25207

(callJNIObjectMethod):

25208

(callJNIByteMethod):

25209

(callJNICharMethod):

25210

(callJNIShortMethod):

25211

(callJNIIntMethod):

25212

(callJNILongMethod):

25213

(callJNIFloatMethod):

25214

(callJNIDoubleMethod):

25215

(callJNIVoidMethodA):

25216

(callJNIObjectMethodA):

25217

(callJNIByteMethodA):

25218

(callJNICharMethodA):

25219

(callJNIShortMethodA):

25220

(callJNIIntMethodA):

25221

(callJNILongMethodA):

25222

(callJNIFloatMethodA):

25223

(callJNIDoubleMethodA):

25224

(getCharactersFromJString):

25225

(releaseCharactersForJString):

25226

* bindings/jni/jni_utility.h:

25227

* bindings/objc/Makefile.am: Removed.

25228

* bindings/runtime.h: Added.

25229

(Bindings::Parameter::~Parameter):

25230

(Bindings::Constructor::~Constructor):

25231

(Bindings::Field::~Field):

25232

(Bindings::Method::~Method):

25233

(Bindings::Class::~Class):

25234

25235

2003-11-13 Maciej Stachowiak <mjs@apple.com>

25236

25237

Reviewed by John.

25238

25239

- fixed 3472562 - Null or Undefined variables passed to IN operator cause javascript exceptions

25240

25241

* kjs/nodes.cpp:

25242

(ForInNode::execute): If the in value is null or undefined, bail

25243

out early, since attempting to iterate its properties will throw

25244

an exception.

25245

25246

2003-11-12 Darin Adler <darin@apple.com>

25247

25248

- fixed the build

25249

25250

* Makefile.am: Fix the build by removing the bindings directory from SUBDIRS.

25251

Later, we can either add this back and add the Makefile.am files to the top

25252

level configure.in or leave it out and remove the Makefile.am files.

25253

25254

2003-11-12 Richard Williamson <rjw@apple.com>

25255

25256

Added utility functions for calling JNI methods.

25257

25258

Reviewed by Chris.

25259

25260

* JavaScriptCore.pbproj/project.pbxproj:

25261

* Makefile.am:

25262

* bindings/Makefile.am: Added.

25263

* bindings/jni/Makefile.am: Added.

25264

* bindings/jni/jni_utility.cpp: Added.

25265

(attachToJavaVM):

25266

(callJNIMethod):

25267

(callJNIVoidMethod):

25268

(callJNIObjectMethod):

25269

(callJNIByteMethod):

25270

(callJNICharMethod):

25271

(callJNIShortMethod):

25272

(callJNIIntMethod):

25273

(callJNILongMethod):

25274

(callJNIFloatMethod):

25275

(callJNIDoubleMethod):

25276

* bindings/jni/jni_utility.h: Added.

25277

* bindings/objc/Makefile.am: Added.

25278

25279

2003-11-08 Darin Adler <darin@apple.com>

25280

25281

Reviewed by John.

25282

25283

- fixed 3477528 -- array.sort(function) fails if the function returns a non-zero value that rounds to zero

25284

25285

* kjs/array_object.cpp:

25286

(compareByStringForQSort): Added checks for undefined values to match what the specification calls for.

25287

(compareWithCompareFunctionForQSort): Added checks for undefined values as above, and also changed the

25288

code that looks at the compare function result to look at the number returned without rounding to an integer.

25289

(ArrayProtoFuncImp::call): Changed the code that looks at the compare function result to look at the number

25290

returned without rounding to an integer.

25291

25292

=== Safari-113 ===

25293

25294

2003-11-03 Vicki Murley <vicki@apple.com>

25295

25296

Reviewed by kocienda.

25297

25298

- fixed <rdar://problem/3471096>: non-B&I builds should not use order files, because they cause false "regressions" in perf.

25299

25300

* JavaScriptCore.pbproj/project.pbxproj: added empty SECTORDER_FLAGS variables to the Development and Deployment build styles

25301

25302

2003-11-02 Darin Adler <darin@apple.com>

25303

25304

Reviewed by Maciej.

25305

25306

- changed list manipulation to use Harri Porten's idea of a circular

25307

linked list that is built from head to tail rather than building the

25308

list backwards and reversing the list when done

25309

25310

* kjs/grammar.y: Handle CatchNode and FinallyNode in a type-safe way.

25311

Change many places that passed 0L to pass nothing at all, or to pass 0.

25312

25313

* kjs/nodes.h:

25314

(KJS::ElementNode::ElementNode): Build a circular list instead of a 0-terminated

25315

backwards list.

25316

(KJS::ArrayNode::ArrayNode): Break the circular list instead of reversing the list.

25317

(KJS::PropertyValueNode::PropertyValueNode): Moved before ObjectLiteralNode so the

25318

inline code in ObjectLiteralNode works. Build a circular list instead of a 0-terminated

25319

backwards list. Made the case for the first node separate so we don't need a nil check.

25320

(KJS::ObjectLiteralNode::ObjectLiteralNode): Break the circular list instead of

25321

reversing the list.

25322

(KJS::ArgumentListNode::ArgumentListNode): Build a circular list instead of a 0-terminated

25323

backwards list. Also, made the constructors inline (moved here from .cpp file).

25324

(KJS::ArgumentsNode::ArgumentsNode): Break the circular list instead of

25325

reversing the list.

25326

(KJS::NewExprNode::NewExprNode): Changed a 0L to 0.

25327

(KJS::StatListNode::StatListNode): Make this constructor no longer inline (moved into

25328

.cpp file). The one in the .cpp file builds a circular list instead of a 0-terminated

25329

backwards list.

25330

(KJS::VarDeclListNode::VarDeclListNode): Build a circular list instead of a 0-terminated

25331

backwards list.

25332

(KJS::VarStatementNode::VarStatementNode): Break the circular list instead of reversing

25333

the list.

25334

(KJS::BlockNode::BlockNode): Make this constructor no longer inline (moved into .cpp file).

25335

The one in the .cpp file breaks the list instead of reversing it.

25336

(KJS::ForNode::ForNode): Break the circular list instead of reversing the list.

25337

(KJS::CaseClauseNode::CaseClauseNode): Break the circular list instead of reversing the

25338

list.

25339

(KJS::ClauseListNode::ClauseListNode): Build a circular list instead of a 0-terminated

25340

backwards list.

25341

(KJS::CaseBlockNode::CaseBlockNode): Make this constructor no longer inline (moved into

25342

.cpp file). The one in the .cpp file breaks the list instead of reversing it.

25343

(KJS::TryNode::TryNode): Changed constructor to take typed parameters for the catch and

25344

finally nodes rather than just Node.

25345

(KJS::ParameterNode::ParameterNode): Build a circular list instead of a 0-terminated

25346

backwards list.

25347

(KJS::FuncDeclNode::FuncDeclNode): Break the circular list instead of reversing the

25348

list.

25349

(KJS::FuncExprNode::FuncExprNode): Break the circular list instead of reversing the

25350

list.

25351

25352

* kjs/nodes.cpp:

25353

(StatListNode::StatListNode): Moved this constructor here, no longer inline.

25354

Did the "break circular list" thing instead of the "reverse list" thing.

25355

Added setLoc calls to match KJS in the KDE tree; since we don't currently

25356

use the JavaScript debugging support, it's unclear whether there's any benefit, but

25357

later we might be using it and it's good to be as close as possible.

25358

(BlockNode::BlockNode): Moved this constructor here, no longer inline.

25359

Did the "break circular list" thing instead of the "reverse list" thing.

25360

Added setLoc calls.

25361

(CaseBlockNode::CaseBlockNode): Moved this constructor here, no longer inline.

25362

Did the "break circular list" thing instead of the "reverse list" thing.

25363

(SourceElementsNode::SourceElementsNode): Moved this constructor here, no longer inline.

25364

Did the "break circular list" thing instead of the "reverse list" thing.

25365

Added setLoc calls.

25366

25367

* kjs/grammar.cpp: Regenerated.

25368

* kjs/grammar.cpp.h: Regenerated.

25369

* kjs/grammar.h: Regenerated.

25370

25371

=== Safari-112 ===

25372

25373

2003-10-30 Maciej Stachowiak <mjs@apple.com>

25374

25375

Reviewed by Ken.

25376

25377

- fixed 3427069 - browsing mp3.com causes leaks (KJS)

25378

25379

* kjs/string_object.cpp:

25380

(StringProtoFuncImp::call): Don't do an early return, since that

25381

could leak a temporary regexp.

25382

25383

2003-10-29 Maciej Stachowiak <mjs@apple.com>

25384

25385

Reviewed by Darin.

25386

25387

- fixed 3426076 - Leak of JS lexer data visiting http://www.ebay.com

25388

25389

* kjs/grammar.cpp:

25390

(yyerror): Updated the commented code.

25391

* kjs/grammar.y: Don't delete string and identifier tokens when done

25392

with them any more, they'll get cleaned up by the lexer now.

25393

* kjs/internal.cpp:

25394

(Parser::parse): Tell lexer when done parsing.

25395

* kjs/lexer.cpp:

25396

(Lexer::Lexer): Initialize new data members.

25397

(Lexer::lex): Use new methods to make strings and identifiers, and

25398

save them.

25399

(Lexer::makeIdentifier): Make a new Identifier and save it in an

25400

auto-growing array.

25401

(Lexer::makeUString): Likewise for UStrings.

25402

(Lexer::doneParsing): Clean up arrays of Ifentifiers and UStrings.

25403

* kjs/lexer.h:

25404

25405

2003-10-28 Maciej Stachowiak <mjs@apple.com>

25406

25407

Reviewed by Ken.

25408

25409

- fixed 3413962 - malicious web pages can kill all future JavaScript execution by breaking recursion limit check

25410

25411

* kjs/nodes.cpp:

25412

(FunctionCallNode::evaluate): If we're going to return early due

25413

to breaking the recursion limit, make sure to lower it again, or

25414

it will creep up by one each time it's exceeded.

25415

25416

2003-10-26 Darin Adler <darin@apple.com>

25417

25418

* JavaScriptCorePrefix.h: Added a C case to the NULL definition since we use C as well

25419

as C++ in this project.

25420

25421

2003-10-26 Darin Adler <darin@apple.com>

25422

25423

- rolled in some CString changes Harri Porten did on the KDE side

25424

25425

* kjs/ustring.cpp:

25426

(KJS::CString::CString): Use memcpy instead of strcpy for speed. Fix an off by one error

25427

in the copy constructor.

25428

(KJS::CString::operator=): Use memcpy instead of strcpy for speed.

25429

25430

* JavaScriptCorePrefix.h: Add a definition of NULL here that takes advantage of the GNU

25431

__null feature even if the system C library doesn't.

25432

25433

== Rolled over to ChangeLog-2003-10-25 ==

Older »