~mhammond/bzr/update-r : revision 2009.1944.1

1

=======================

2

Guide to Testing Bazaar

3

=======================

4

5

.. contents::

6

7

Testing Bazaar

8

##############

9

10

The Importance of Testing

11

=========================

12

13

Reliability is a critical success factor for any Version Control System.

14

We want Bazaar to be highly reliable across multiple platforms while

15

evolving over time to meet the needs of its community.

16

17

In a nutshell, this is what we expect and encourage:

18

19

* New functionality should have test cases. Preferably write the

20

test before writing the code.

21

22

In general, you can test at either the command-line level or the

23

internal API level. See `Writing tests`_ below for more detail.

24

25

* Try to practice Test-Driven Development: before fixing a bug, write a

26

test case so that it does not regress. Similarly for adding a new

27

feature: write a test case for a small version of the new feature before

28

starting on the code itself. Check the test fails on the old code, then

29

add the feature or fix and check it passes.

30

31

By doing these things, the Bazaar team gets increased confidence that

32

changes do what they claim to do, whether provided by the core team or

33

by community members. Equally importantly, we can be surer that changes

34

down the track do not break new features or bug fixes that you are

35

contributing today.

36

37

As of May 2008, Bazaar ships with a test suite containing over 12000 tests

38

and growing. We are proud of it and want to remain so. As community

39

members, we all benefit from it. Would you trust version control on

40

your project to a product *without* a test suite like Bazaar has?

41

42

43

Running the Test Suite

44

======================

45

46

Currently, bzr selftest is used to invoke tests.

47

You can provide a pattern argument to run a subset. For example,

48

to run just the blackbox tests, run::

49

50

./bzr selftest -v blackbox

51

52

To skip a particular test (or set of tests), use the --exclude option

53

(shorthand -x) like so::

54

55

./bzr selftest -v -x blackbox

56

57

To ensure that all tests are being run and succeeding, you can use the

58

--strict option which will fail if there are any missing features or known

59

failures, like so::

60

61

./bzr selftest --strict

62

63

To list tests without running them, use the --list-only option like so::

64

65

./bzr selftest --list-only

66

67

This option can be combined with other selftest options (like -x) and

68

filter patterns to understand their effect.

69

70

Once you understand how to create a list of tests, you can use the --load-list

71

option to run only a restricted set of tests that you kept in a file, one test

72

id by line. Keep in mind that this will never be sufficient to validate your

73

modifications, you still need to run the full test suite for that, but using it

74

can help in some cases (like running only the failed tests for some time)::

75

76

./bzr selftest -- load-list my_failing_tests

77

78

This option can also be combined with other selftest options, including

79

patterns. It has some drawbacks though, the list can become out of date pretty

80

quick when doing Test Driven Development.

81

82

To address this concern, there is another way to run a restricted set of tests:

83

the --starting-with option will run only the tests whose name starts with the

84

specified string. It will also avoid loading the other tests and as a

85

consequence starts running your tests quicker::

86

87

./bzr selftest --starting-with bzrlib.blackbox

88

89

This option can be combined with all the other selftest options including

90

--load-list. The later is rarely used but allows to run a subset of a list of

91

failing tests for example.

92

93

94

Test suite debug flags

95

----------------------

96

97

Similar to the global ``-Dfoo`` debug options, bzr selftest accepts

98

``-E=foo`` debug flags. These flags are:

99

100

:allow_debug: do *not* clear the global debug flags when running a test.

101

This can provide useful logging to help debug test failures when used

102

with e.g. ``bzr -Dhpss selftest -E=allow_debug``

103

104

105

Writing Tests

106

=============

107

108

Where should I put a new test?

109

------------------------------

110

111

Bzrlib's tests are organised by the type of test. Most of the tests in

112

bzr's test suite belong to one of these categories:

113

114

- Unit tests

115

- Blackbox (UI) tests

116

- Per-implementation tests

117

- Doctests

118

119

A quick description of these test types and where they belong in bzrlib's

120

source follows. Not all tests fall neatly into one of these categories;

121

in those cases use your judgement.

122

123

124

Unit tests

125

~~~~~~~~~~

126

127

Unit tests make up the bulk of our test suite. These are tests that are

128

focused on exercising a single, specific unit of the code as directly

129

as possible. Each unit test is generally fairly short and runs very

130

quickly.

131

132

They are found in ``bzrlib/tests/test_*.py``. So in general tests should

133

be placed in a file named test_FOO.py where FOO is the logical thing under

134

test.

135

136

For example, tests for merge3 in bzrlib belong in bzrlib/tests/test_merge3.py.

137

See bzrlib/tests/test_sampler.py for a template test script.

138

139

140

Blackbox (UI) tests

141

~~~~~~~~~~~~~~~~~~~

142

143

Tests can be written for the UI or for individual areas of the library.

144

Choose whichever is appropriate: if adding a new command, or a new command

145

option, then you should be writing a UI test. If you are both adding UI

146

functionality and library functionality, you will want to write tests for

147

both the UI and the core behaviours. We call UI tests 'blackbox' tests

148

and they belong in ``bzrlib/tests/blackbox/*.py``.

149

150

When writing blackbox tests please honour the following conventions:

151

152

1. Place the tests for the command 'name' in

153

bzrlib/tests/blackbox/test_name.py. This makes it easy for developers

154

to locate the test script for a faulty command.

155

156

2. Use the 'self.run_bzr("name")' utility function to invoke the command

157

rather than running bzr in a subprocess or invoking the

158

cmd_object.run() method directly. This is a lot faster than

159

subprocesses and generates the same logging output as running it in a

160

subprocess (which invoking the method directly does not).

161

162

3. Only test the one command in a single test script. Use the bzrlib

163

library when setting up tests and when evaluating the side-effects of

164

the command. We do this so that the library api has continual pressure

165

on it to be as functional as the command line in a simple manner, and

166

to isolate knock-on effects throughout the blackbox test suite when a

167

command changes its name or signature. Ideally only the tests for a

168

given command are affected when a given command is changed.

169

170

4. If you have a test which does actually require running bzr in a

171

subprocess you can use ``run_bzr_subprocess``. By default the spawned

172

process will not load plugins unless ``--allow-plugins`` is supplied.

173

174

175

Per-implementation tests

176

~~~~~~~~~~~~~~~~~~~~~~~~

177

178

Per-implementation tests are tests that are defined once and then run

179

against multiple implementations of an interface. For example,

180

``test_transport_implementations.py`` defines tests that all Transport

181

implementations (local filesystem, HTTP, and so on) must pass.

182

183

They are found in ``bzrlib/tests/*_implementations/test_*.py``,

184

``bzrlib/tests/per_*/*.py``, and

185

``bzrlib/tests/test_*_implementations.py``.

186

187

These are really a sub-category of unit tests, but an important one.

188

189

190

Doctests

191

~~~~~~~~

192

193

We make selective use of doctests__. In general they should provide

194

*examples* within the API documentation which can incidentally be tested. We

195

don't try to test every important case using doctests |--| regular Python

196

tests are generally a better solution. That is, we just use doctests to

197

make our documentation testable, rather than as a way to make tests.

198

199

Most of these are in ``bzrlib/doc/api``. More additions are welcome.

200

201

__ http://docs.python.org/lib/module-doctest.html

202

203

204

.. Effort tests

205

.. ~~~~~~~~~~~~

206

207

208

209

Skipping tests

210

--------------

211

212

In our enhancements to unittest we allow for some addition results beyond

213

just success or failure.

214

215

If a test can't be run, it can say that it's skipped by raising a special

216

exception. This is typically used in parameterized tests |--| for example

217

if a transport doesn't support setting permissions, we'll skip the tests

218

that relating to that. ::

219

220

try:

221

return self.branch_format.initialize(repo.bzrdir)

222

except errors.UninitializableFormat:

223

raise tests.TestSkipped('Uninitializable branch format')

224

225

Raising TestSkipped is a good idea when you want to make it clear that the

226

test was not run, rather than just returning which makes it look as if it

227

was run and passed.

228

229

Several different cases are distinguished:

230

231

TestSkipped

232

Generic skip; the only type that was present up to bzr 0.18.

233

234

TestNotApplicable

235

The test doesn't apply to the parameters with which it was run.

236

This is typically used when the test is being applied to all

237

implementations of an interface, but some aspects of the interface

238

are optional and not present in particular concrete

239

implementations. (Some tests that should raise this currently

240

either silently return or raise TestSkipped.) Another option is

241

to use more precise parameterization to avoid generating the test

242

at all.

243

244

TestPlatformLimit

245

**(Not implemented yet)**

246

The test can't be run because of an inherent limitation of the

247

environment, such as not having symlinks or not supporting

248

unicode.

249

250

UnavailableFeature

251

The test can't be run because a dependency (typically a Python

252

library) is not available in the test environment. These

253

are in general things that the person running the test could fix

254

by installing the library. It's OK if some of these occur when

255

an end user runs the tests or if we're specifically testing in a

256

limited environment, but a full test should never see them.

257

258

See `Test feature dependencies`_ below.

259

260

KnownFailure

261

The test exists but is known to fail, for example this might be

262

appropriate to raise if you've committed a test for a bug but not

263

the fix for it, or if something works on Unix but not on Windows.

264

265

Raising this allows you to distinguish these failures from the

266

ones that are not expected to fail. If the test would fail

267

because of something we don't expect or intend to fix,

268

KnownFailure is not appropriate, and TestNotApplicable might be

269

better.

270

271

KnownFailure should be used with care as we don't want a

272

proliferation of quietly broken tests.

273

274

We plan to support three modes for running the test suite to control the

275

interpretation of these results. Strict mode is for use in situations

276

like merges to the mainline and releases where we want to make sure that

277

everything that can be tested has been tested. Lax mode is for use by

278

developers who want to temporarily tolerate some known failures. The

279

default behaviour is obtained by ``bzr selftest`` with no options, and

280

also (if possible) by running under another unittest harness.

281

282

======================= ======= ======= ========

283

result strict default lax

284

======================= ======= ======= ========

285

TestSkipped pass pass pass

286

TestNotApplicable pass pass pass

287

TestPlatformLimit pass pass pass

288

TestDependencyMissing fail pass pass

289

KnownFailure fail pass pass

290

======================= ======= ======= ========

291

292

293

Test feature dependencies

294

-------------------------

295

296

Writing tests that require a feature

297

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

298

299

Rather than manually checking the environment in each test, a test class

300

can declare its dependence on some test features. The feature objects are

301

checked only once for each run of the whole test suite.

302

303

(For historical reasons, as of May 2007 many cases that should depend on

304

features currently raise TestSkipped.)

305

306

For example::

307

308

class TestStrace(TestCaseWithTransport):

309

310

_test_needs_features = [StraceFeature]

311

312

This means all tests in this class need the feature. The feature itself

313

should provide a ``_probe`` method which is called once to determine if

314

it's available.

315

316

These should generally be equivalent to either TestDependencyMissing or

317

sometimes TestPlatformLimit.

318

319

Individual tests can also require a feature using the ``requireFeature``

320

method::

321

322

self.requireFeature(StraceFeature)

323

324

Features already defined in bzrlib.tests include:

325

326

- SymlinkFeature,

327

- HardlinkFeature,

328

- OsFifoFeature,

329

- UnicodeFilenameFeature,

330

- FTPServerFeature, and

331

- CaseInsensitiveFilesystemFeature.

332

333

334

Defining a new feature that tests can require

335

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

336

337

New features for use with ``_test_needs_features`` or ``requireFeature``

338

are defined by subclassing ``bzrlib.tests.Feature`` and overriding the

339

``_probe`` and ``feature_name`` methods. For example::

340

341

class _SymlinkFeature(Feature):

342

343

def _probe(self):

344

return osutils.has_symlinks()

345

346

def feature_name(self):

347

return 'symlinks'

348

349

SymlinkFeature = _SymlinkFeature()

350

351

352

Testing exceptions and errors

353

-----------------------------

354

355

It's important to test handling of errors and exceptions. Because this

356

code is often not hit in ad-hoc testing it can often have hidden bugs --

357

it's particularly common to get NameError because the exception code

358

references a variable that has since been renamed.

359

360

.. TODO: Something about how to provoke errors in the right way?

361

362

In general we want to test errors at two levels:

363

364

1. A test in ``test_errors.py`` checking that when the exception object is

365

constructed with known parameters it produces an expected string form.

366

This guards against mistakes in writing the format string, or in the

367

``str`` representations of its parameters. There should be one for

368

each exception class.

369

370

2. Tests that when an api is called in a particular situation, it raises

371

an error of the expected class. You should typically use

372

``assertRaises``, which in the Bazaar test suite returns the exception

373

object to allow you to examine its parameters.

374

375

In some cases blackbox tests will also want to check error reporting. But

376

it can be difficult to provoke every error through the commandline

377

interface, so those tests are only done as needed |--| eg in response to a

378

particular bug or if the error is reported in an unusual way(?) Blackbox

379

tests should mostly be testing how the command-line interface works, so

380

should only test errors if there is something particular to the cli in how

381

they're displayed or handled.

382

383

384

Testing warnings

385

----------------

386

387

The Python ``warnings`` module is used to indicate a non-fatal code

388

problem. Code that's expected to raise a warning can be tested through

389

callCatchWarnings.

390

391

The test suite can be run with ``-Werror`` to check no unexpected errors

392

occur.

393

394

However, warnings should be used with discretion. It's not an appropriate

395

way to give messages to the user, because the warning is normally shown

396

only once per source line that causes the problem. You should also think

397

about whether the warning is serious enought that it should be visible to

398

users who may not be able to fix it.

399

400

401

Interface implementation testing and test scenarios

402

---------------------------------------------------

403

404

There are several cases in Bazaar of multiple implementations of a common

405

conceptual interface. ("Conceptual" because it's not necessary for all

406

the implementations to share a base class, though they often do.)

407

Examples include transports and the working tree, branch and repository

408

classes.

409

410

In these cases we want to make sure that every implementation correctly

411

fulfils the interface requirements. For example, every Transport should

412

support the ``has()`` and ``get()`` and ``clone()`` methods. We have a

413

sub-suite of tests in ``test_transport_implementations``. (Most

414

per-implementation tests are in submodules of ``bzrlib.tests``, but not

415

the transport tests at the moment.)

416

417

These tests are repeated for each registered Transport, by generating a

418

new TestCase instance for the cross product of test methods and transport

419

implementations. As each test runs, it has ``transport_class`` and

420

``transport_server`` set to the class it should test. Most tests don't

421

access these directly, but rather use ``self.get_transport`` which returns

422

a transport of the appropriate type.

423

424

The goal is to run per-implementation only the tests that relate to that

425

particular interface. Sometimes we discover a bug elsewhere that happens

426

with only one particular transport. Once it's isolated, we can consider

427

whether a test should be added for that particular implementation,

428

or for all implementations of the interface.

429

430

The multiplication of tests for different implementations is normally

431

accomplished by overriding the ``load_tests`` function used to load tests

432

from a module. This function typically loads all the tests, then applies

433

a TestProviderAdapter to them, which generates a longer suite containing

434

all the test variations.

435

436

See also `Per-implementation tests`_ (above).

437

438

439

Test scenarios

440

--------------

441

442

Some utilities are provided for generating variations of tests. This can

443

be used for per-implementation tests, or other cases where the same test

444

code needs to run several times on different scenarios.

445

446

The general approach is to define a class that provides test methods,

447

which depend on attributes of the test object being pre-set with the

448

values to which the test should be applied. The test suite should then

449

also provide a list of scenarios in which to run the tests.

450

451

Typically ``multiply_tests_from_modules`` should be called from the test

452

module's ``load_tests`` function.

453

454

455

Test support

456

------------

457

458

We have a rich collection of tools to support writing tests. Please use

459

them in preference to ad-hoc solutions as they provide portability and

460

performance benefits.

461

462

463

TestCase and its subclasses

464

~~~~~~~~~~~~~~~~~~~~~~~~~~~

465

466

The ``bzrlib.tests`` module defines many TestCase classes to help you

467

write your tests.

468

469

TestCase

470

A base TestCase that extends the Python standard library's

471

TestCase in several ways. It adds more assertion methods (e.g.

472

``assertContainsRe``), ``addCleanup``, and other features (see its API

473

docs for details). It also has a ``setUp`` that makes sure that

474

global state like registered hooks and loggers won't interfere with

475

your test. All tests should use this base class (whether directly or

476

via a subclass).

477

478

TestCaseWithMemoryTransport

479

Extends TestCase and adds methods like ``get_transport``,

480

``make_branch`` and ``make_branch_builder``. The files created are

481

stored in a MemoryTransport that is discarded at the end of the test.

482

This class is good for tests that need to make branches or use

483

transports, but that don't require storing things on disk. All tests

484

that create bzrdirs should use this base class (either directly or via

485

a subclass) as it ensures that the test won't accidentally operate on

486

real branches in your filesystem.

487

488

TestCaseInTempDir

489

Extends TestCaseWithMemoryTransport. For tests that really do need

490

files to be stored on disk, e.g. because a subprocess uses a file, or

491

for testing functionality that accesses the filesystem directly rather

492

than via the Transport layer (such as dirstate).

493

494

TestCaseWithTransport

495

Extends TestCaseInTempDir. Provides ``get_url`` and

496

``get_readonly_url`` facilities. Subclasses can control the

497

transports used by setting ``vfs_transport_factory``,

498

``transport_server`` and/or ``transport_readonly_server``.

499

500

501

See the API docs for more details.

502

503

504

BranchBuilder

505

~~~~~~~~~~~~~

506

507

When writing a test for a feature, it is often necessary to set up a

508

branch with a certain history. The ``BranchBuilder`` interface allows the

509

creation of test branches in a quick and easy manner. Here's a sample

510

session::

511

512

builder = self.make_branch_builder('relpath')

513

builder.build_commit()

514

builder.build_commit()

515

builder.build_commit()

516

branch = builder.get_branch()

517

518

``make_branch_builder`` is a method of ``TestCaseWithMemoryTransport``.

519

520

Note that many current tests create test branches by inheriting from

521

``TestCaseWithTransport`` and using the ``make_branch_and_tree`` helper to

522

give them a ``WorkingTree`` that they can commit to. However, using the

523

newer ``make_branch_builder`` helper is preferred, because it can build

524

the changes in memory, rather than on disk. Tests that are explictly

525

testing how we work with disk objects should, of course, use a real

526

``WorkingTree``.

527

528

Please see bzrlib.branchbuilder for more details.

529

530

531

TreeBuilder

532

~~~~~~~~~~~

533

534

The ``TreeBuilder`` interface allows the construction of arbitrary trees

535

with a declarative interface. A sample session might look like::

536

537

tree = self.make_branch_and_tree('path')

538

builder = TreeBuilder()

539

builder.start_tree(tree)

540

builder.build(['foo', "bar/", "bar/file"])

541

tree.commit('commit the tree')

542

builder.finish_tree()

543

544

Usually a test will create a tree using ``make_branch_and_memory_tree`` (a

545

method of ``TestCaseWithMemoryTransport``) or ``make_branch_and_tree`` (a

546

method of ``TestCaseWithTransport``).

547

548

Please see bzrlib.treebuilder for more details.

549

550

551

.. |--| unicode:: U+2014

552

553

..

554

vim: ft=rst tw=74 ai