2
# -*- Mode: perl; indent-tabs-mode: nil; c-basic-offset: 4 -*-
5
# The Intltool Message Extractor
7
# Copyright (C) 2000-2001, 2003 Free Software Foundation.
9
# Intltool is free software; you can redistribute it and/or
10
# modify it under the terms of the GNU General Public License as
11
# published by the Free Software Foundation; either version 2 of the
12
# License, or (at your option) any later version.
14
# Intltool is distributed in the hope that it will be useful,
15
# but WITHOUT ANY WARRANTY; without even the implied warranty of
16
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
17
# General Public License for more details.
19
# You should have received a copy of the GNU General Public License
20
# along with this program; if not, write to the Free Software
21
# Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
23
# As a special exception to the GNU General Public License, if you
24
# distribute this file as part of a program that contains a
25
# configuration script generated by Autoconf, you may include it under
26
# the same distribution terms that you use for the rest of that program.
28
# Authors: Kenneth Christiansen <kenneth@gnu.org>
29
# Darin Adler <darin@bentspoon.com>
32
## Release information
33
my $PROGRAM = "intltool-extract";
34
my $PACKAGE = "intltool";
35
my $VERSION = "0.35.0";
42
## Scalars used by the option stuff
46
my $VERSION_ARG = "0";
54
my $gettext_type = "";
64
## Use this instead of \w for XML files to handle more possible characters.
65
my $w = "[-A-Za-z0-9._:]";
72
"type=s" => \$TYPE_ARG,
73
"local|l" => \$LOCAL_ARG,
74
"help|h" => \$HELP_ARG,
75
"version|v" => \$VERSION_ARG,
76
"update" => \$UPDATE_ARG,
77
"quiet|q" => \$QUIET_ARG,
78
"srcdir=s" => \$SRCDIR_ARG,
85
## This section will check for the different options.
87
sub split_on_argument {
95
} elsif ($LOCAL_ARG) {
99
} elsif ($UPDATE_ARG) {
103
} elsif (@ARGV > 0) {
116
$OUTFILE = "$FILE.h";
121
$OUTFILE = fileparse($FILE, ());
123
system("mkdir tmp/");
125
$OUTFILE = "./tmp/$OUTFILE.h"
129
if ($TYPE_ARG =~ /^gettext\/(.*)/) {
134
## Sub for printing release information
137
${PROGRAM} (${PACKAGE}) $VERSION
138
Copyright (C) 2000, 2003 Free Software Foundation, Inc.
139
Written by Kenneth Christiansen, 2000.
141
This is free software; see the source for copying conditions. There is NO
142
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
147
## Sub for printing usage information
150
Usage: ${PROGRAM} [OPTION]... [FILENAME]
151
Generates a header file from an XML source file.
153
It grabs all strings between <_translatable_node> and its end tag in
154
XML files. Read manpage (man ${PROGRAM}) for more info.
156
--type=TYPE Specify the file type of FILENAME. Currently supports:
157
"gettext/glade", "gettext/ini", "gettext/keys"
158
"gettext/rfc822deb", "gettext/schemas",
159
"gettext/scheme", "gettext/xml"
160
-l, --local Writes output into current working directory
161
(conflicts with --update)
162
--update Writes output into the same directory the source file
163
reside (conflicts with --local)
164
--srcdir Root of the source tree
165
-v, --version Output version information and exit
166
-h, --help Display this help and exit
167
-q, --quiet Quiet mode
169
Report bugs to http://bugzilla.gnome.org/ (product name "$PACKAGE")
170
or send email to <xml-i18n-tools\@gnome.org>.
175
## Sub for printing error messages
177
print STDERR "Try `${PROGRAM} --help' for more information.\n";
182
print "Generating C format header file for translation.\n" unless $QUIET_ARG;
190
open OUT, ">$OUTFILE";
191
binmode (OUT) if $^O eq 'MSWin32';
195
print "Wrote $OUTFILE\n" unless $QUIET_ARG;
203
local $/; #slurp mode
204
open (IN, "<$SRCDIR_ARG/$FILE") || die "can't open $SRCDIR_ARG/$FILE: $!";
208
&type_ini if $gettext_type eq "ini";
209
&type_keys if $gettext_type eq "keys";
210
&type_xml if $gettext_type eq "xml";
211
&type_glade if $gettext_type eq "glade";
212
&type_scheme if $gettext_type eq "scheme";
213
&type_schemas if $gettext_type eq "schemas";
214
&type_rfc822deb if $gettext_type eq "rfc822deb";
217
sub entity_decode_minimal
243
return '\"' if $_ eq '"';
244
return '\n' if $_ eq "\n";
245
return '\\' if $_ eq '\\';
253
return join "", map &escape_char, split //, $string;
257
### For generic translatable desktop files ###
258
while ($input =~ /^_.*=(.*)$/mg) {
264
### For generic translatable mime/keys files ###
265
while ($input =~ /^\s*_\w+=(.*)$/mg) {
271
### For generic translatable XML files ###
272
my $tree = readXml($input);
278
my $vartype = ref $var;
280
if ($vartype =~ /ARRAY/) {
283
foreach my $el (@arr) {
288
} elsif ($vartype =~ /HASH/) {
291
foreach my $key (keys %hash) {
293
print_var($hash{$key});
302
# Same syntax as getAttributeString in intltool-merge.in.in, similar logic (look for ## differences comment)
303
sub getAttributeString
306
my $do_translate = shift || 1;
307
my $language = shift || "";
308
my $translate = shift;
310
foreach my $e (reverse(sort(keys %{ $sub }))) {
312
my $string = $sub->{$e};
315
$string =~ s/^[\s]+//;
316
$string =~ s/[\s]+$//;
318
if ($string =~ /^'.*'$/)
322
$string =~ s/^['"]//g;
323
$string =~ s/['"]$//g;
325
## differences from intltool-merge.in.in
327
$comments{entity_decode($string)} = $XMLCOMMENT if $XMLCOMMENT;
328
$messages{entity_decode($string)} = [];
331
## differences end here from intltool-merge.in.in
332
$result .= " $key=$quote$string$quote";
337
# Verbatim copy from intltool-merge.in.in
341
my $spacepreserve = shift || 0;
342
my @list = @{ $ref };
345
my $count = scalar(@list);
346
my $attrs = $list[0];
349
$spacepreserve = 1 if ((exists $attrs->{"xml:space"}) && ($attrs->{"xml:space"} =~ /^["']?preserve["']?$/));
350
$spacepreserve = 0 if ((exists $attrs->{"xml:space"}) && ($attrs->{"xml:space"} =~ /^["']?default["']?$/));
352
while ($index < $count) {
353
my $type = $list[$index];
354
my $content = $list[$index+1];
358
# lets strip the whitespace here, and *ONLY* here
359
$content =~ s/\s+/ /gs if (!$spacepreserve);
362
} elsif ( "$type" ne "1" ) {
363
# We've got another element
365
$result .= getAttributeString(@{$content}[0], 0); # no nested translatable elements
367
my $subresult = getXMLstring($content, $spacepreserve);
369
$result .= ">".$subresult . "</$type>";
382
# Verbatim copy from intltool-merge.in.in, except for MULTIPLE_OUTPUT handling removed
383
# Translate list of nodes if necessary
384
sub translate_subnodes
388
my $language = shift || "";
389
my $singlelang = shift || 0;
390
my $spacepreserve = shift || 0;
392
my @nodes = @{ $content };
394
my $count = scalar(@nodes);
396
while ($index < $count) {
397
my $type = $nodes[$index];
398
my $rest = $nodes[$index+1];
399
traverse($fh, $type, $rest, $language, $spacepreserve);
404
# Based on traverse() in intltool-merge.in.in
407
my $fh = shift; # unused, to allow us to sync code between -merge and -extract
408
my $nodename = shift;
410
my $language = shift || "";
411
my $spacepreserve = shift || 0;
413
if ($nodename && "$nodename" eq "1") {
414
$XMLCOMMENT = $content;
415
} elsif ($nodename) {
417
my @all = @{ $content };
418
my $attrs = shift @all;
420
my $outattr = getAttributeString($attrs, 1, $language, \$translate);
422
if ($nodename =~ /^_/) {
428
$spacepreserve = 0 if ((exists $attrs->{"xml:space"}) && ($attrs->{"xml:space"} =~ /^["']?default["']?$/));
429
$spacepreserve = 1 if ((exists $attrs->{"xml:space"}) && ($attrs->{"xml:space"} =~ /^["']?preserve["']?$/));
432
$lookup = getXMLstring($content, $spacepreserve);
433
if (!$spacepreserve) {
434
$lookup =~ s/^\s+//s;
435
$lookup =~ s/\s+$//s;
438
if ($lookup && $translate != 2) {
439
$comments{$lookup} = $XMLCOMMENT if $XMLCOMMENT;
440
$messages{$lookup} = [];
441
} elsif ($translate == 2) {
442
translate_subnodes($fh, \@all, $language, 1, $spacepreserve);
446
my $count = scalar(@all);
449
while ($index < $count) {
450
my $type = $all[$index];
451
my $rest = $all[$index+1];
452
traverse($fh, $type, $rest, $language, $spacepreserve);
462
# Verbatim copy from intltool-merge.in.in, $fh for compatibility
467
my $language = shift || "";
469
my $name = shift @{ $ref };
470
my $cont = shift @{ $ref };
472
while (!$name || "$name" eq "1") {
473
$name = shift @{ $ref };
474
$cont = shift @{ $ref };
477
my $spacepreserve = 0;
478
my $attrs = @{$cont}[0];
479
$spacepreserve = 1 if ((exists $attrs->{"xml:space"}) && ($attrs->{"xml:space"} =~ /^["']?preserve["']?$/));
481
traverse($fh, $name, $cont, $language, $spacepreserve);
484
# Verbatim copy from intltool-merge.in.in
485
sub intltool_tree_comment
488
my $data = $expat->original_string();
489
my $clist = $expat->{Curlist};
494
push @$clist, 1 => $data;
497
# Verbatim copy from intltool-merge.in.in
498
sub intltool_tree_cdatastart
501
my $clist = $expat->{Curlist};
504
push @$clist, 0 => $expat->original_string();
507
# Verbatim copy from intltool-merge.in.in
508
sub intltool_tree_cdataend
511
my $clist = $expat->{Curlist};
514
$clist->[$pos] .= $expat->original_string();
517
# Verbatim copy from intltool-merge.in.in
518
sub intltool_tree_char
522
my $clist = $expat->{Curlist};
525
# Use original_string so that we retain escaped entities
528
if ($pos > 0 and $clist->[$pos - 1] eq '0') {
529
$clist->[$pos] .= $expat->original_string();
531
push @$clist, 0 => $expat->original_string();
535
# Verbatim copy from intltool-merge.in.in
536
sub intltool_tree_start
542
# Use original_string so that we retain escaped entities
543
# in attribute values. We must convert the string to an
544
# @origlist array to conform to the structure of the Tree
547
my @original_array = split /\x/, $expat->original_string();
548
my $source = $expat->original_string();
550
# Remove leading tag.
552
$source =~ s|^\s*<\s*(\S+)||s;
554
# Grab attribute key/value pairs and push onto @origlist array.
558
if ($source =~ /^\s*([\w:-]+)\s*[=]\s*["]/)
560
$source =~ s|^\s*([\w:-]+)\s*[=]\s*["]([^"]*)["]||s;
562
push @origlist, '"' . $2 . '"';
564
elsif ($source =~ /^\s*([\w:-]+)\s*[=]\s*[']/)
566
$source =~ s|^\s*([\w:-]+)\s*[=]\s*[']([^']*)[']||s;
568
push @origlist, "'" . $2 . "'";
576
my $ol = [ { @origlist } ];
578
push @{ $expat->{Lists} }, $expat->{Curlist};
579
push @{ $expat->{Curlist} }, $tag => $ol;
580
$expat->{Curlist} = $ol;
583
# Copied from intltool-merge.in.in and added comment handler.
586
my $xmldoc = shift || return;
587
my $ret = eval 'require XML::Parser';
589
die "You must have XML::Parser installed to run $0\n\n";
591
my $xp = new XML::Parser(Style => 'Tree');
592
$xp->setHandlers(Char => \&intltool_tree_char);
593
$xp->setHandlers(Start => \&intltool_tree_start);
594
$xp->setHandlers(CdataStart => \&intltool_tree_cdatastart);
595
$xp->setHandlers(CdataEnd => \&intltool_tree_cdataend);
597
## differences from intltool-merge.in.in
598
$xp->setHandlers(Comment => \&intltool_tree_comment);
599
## differences end here from intltool-merge.in.in
601
my $tree = $xp->parse($xmldoc);
604
# <foo><!-- comment --><head id="a">Hello <em>there</em></head><bar>Howdy<ref/></bar>do</foo>
606
# [foo, [{}, 1, "comment", head, [{id => "a"}, 0, "Hello ", em, [{}, 0, "there"]], bar,
607
# [{}, 0, "Howdy", ref, [{}]], 0, "do" ] ]
613
### For schemas XML files ###
615
# FIXME: We should handle escaped < (less than)
617
<locale\ name="C">\s*
618
(<default>\s*(?:<!--([^>]*?)-->\s*)?(.*?)\s*<\/default>\s*)?
619
(<short>\s*(?:<!--([^>]*?)-->\s*)?(.*?)\s*<\/short>\s*)?
620
(<long>\s*(?:<!--([^>]*?)-->\s*)?(.*?)\s*<\/long>\s*)?
623
my @totranslate = ($3,$6,$9);
624
my @eachcomment = ($2,$5,$8);
625
foreach (@totranslate) {
626
my $currentcomment = shift @eachcomment;
629
$messages{entity_decode_minimal($_)} = [];
630
$comments{entity_decode_minimal($_)} = $currentcomment if (defined($currentcomment));
636
### For rfc822-style Debian configuration files ###
640
while ($input =~ /\G(.*?)(^|\n)(_+)([^:]+):[ \t]*(.*?)(?=\n\S|$)/sg)
642
my ($pre, $newline, $underscore, $tag, $text) = ($1, $2, $3, $4, $5);
643
while ($pre =~ m/\n/g)
647
$lineno += length($newline);
648
my @str_list = rfc822deb_split(length($underscore), $text);
649
for my $str (@str_list)
652
$messages{$str} = [];
653
$loc{$str} = $lineno;
654
$count{$str} = $strcount;
655
my $usercomment = '';
656
while($pre =~ s/(^|\n)#([^\n]*)$//s)
658
$usercomment = "\n" . $2 . $usercomment;
660
$comments{$str} = $tag . $usercomment;
662
$lineno += ($text =~ s/\n//g);
666
sub rfc822deb_split {
667
# Debian defines a special way to deal with rfc822-style files:
668
# when a value contain newlines, it consists of
669
# 1. a short form (first line)
670
# 2. a long description, all lines begin with a space,
671
# and paragraphs are separated by a single dot on a line
672
# This routine returns an array of all paragraphs, and reformat
674
# When first argument is 2, the string is a comma separated list of
678
$text =~ s/^[ \t]//mg;
679
return (split(/, */, $text, 0)) if $type ne 1;
680
return ($text) if $text !~ /\n/;
682
$text =~ s/([^\n]*)\n//;
685
for my $line (split (/\n/, $text))
688
if ($line =~ /^\.\s*$/)
695
elsif ($line =~ /^\s/)
697
# Line which must not be reformatted
698
$str .= "\n" if length ($str) && $str !~ /\n$/;
704
# Continuation line, remove newline
705
$str .= " " if length ($str) && $str !~ /\n$/;
710
push(@list, $str) if length ($str);
715
### For translatable Glade XML files ###
717
my $tags = "label|title|text|format|copyright|comments|preview_text|tooltip|message";
719
while ($input =~ /<($tags)>([^<]+)<\/($tags)>/sg) {
720
# Glade sometimes uses tags that normally mark translatable things for
721
# little bits of non-translatable content. We work around this by not
722
# translating strings that only includes something like label4 or window1.
723
$messages{entity_decode($2)} = [] unless $2 =~ /^(window|label|dialog)[0-9]+$/;
726
while ($input =~ /<items>(..[^<]*)<\/items>/sg) {
727
for my $item (split (/\n/, $1)) {
728
$messages{entity_decode($item)} = [];
732
## handle new glade files
733
while ($input =~ /<(property|atkproperty)\s+[^>]*translatable\s*=\s*"yes"(?:\s+[^>]*comments\s*=\s*"([^"]*)")?[^>]*>([^<]+)<\/\1>/sg) {
734
$messages{entity_decode($3)} = [] unless $3 =~ /^(window|label)[0-9]+$/;
735
if (defined($2) and !($3 =~ /^(window|label)[0-9]+$/)) {
736
$comments{entity_decode($3)} = entity_decode($2) ;
739
while ($input =~ /<atkaction\s+action_name="([^>]*)"\s+description="([^>]+)"\/>/sg) {
740
$messages{entity_decode_minimal($2)} = [];
745
my ($line, $i, $state, $str, $trcomment, $char);
746
for $line (split(/\n/, $input)) {
748
$state = 0; # 0 - nothing, 1 - string, 2 - translatable string
749
while ($i < length($line)) {
750
if (substr($line,$i,1) eq "\"") {
752
$comments{$str} = $trcomment if ($trcomment);
753
$messages{$str} = [];
755
$state = 0; $trcomment = "";
756
} elsif ($state == 1) {
758
$state = 0; $trcomment = "";
762
if ($i>0 && substr($line,$i-1,1) eq '_') {
767
if (substr($line,$i,1) eq ";") {
768
$trcomment = substr($line,$i+1);
769
$trcomment =~ s/^;*\s*//;
771
} elsif ($trcomment && substr($line,$i,1) !~ /\s|\(|\)|_/) {
775
if (substr($line,$i,1) eq "\\") {
776
$char = substr($line,$i+1,1);
777
if ($char ne "\"" && $char ne "\\") {
782
$str = $str . substr($line,$i,1);
793
@msgids = sort { $count{$a} <=> $count{$b} } keys %count;
797
@msgids = sort keys %messages;
799
for my $message (@msgids)
802
$offsetlines++ if $message =~ /%/;
803
if (defined ($comments{$message}))
805
while ($comments{$message} =~ m/\n/g)
810
print OUT "# ".($loc{$message} - $offsetlines). " \"$FILE\"\n"
811
if defined $loc{$message};
812
print OUT "/* ".$comments{$message}." */\n"
813
if defined $comments{$message};
814
print OUT "/* xgettext:no-c-format */\n" if $message =~ /%/;
816
my @lines = split (/\n/, $message, -1);
817
for (my $n = 0; $n < @lines; $n++)
821
print OUT "char *s = N_(\"";
828
print OUT escape($lines[$n]);