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 = "0.36.2";
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";
118
my $dirname = dirname ($OUTFILE);
119
if (! -d "$dirname" && $dirname ne "") {
120
system ("mkdir -p $dirname");
126
$OUTFILE = fileparse($FILE, ());
128
system("mkdir tmp/");
130
$OUTFILE = "./tmp/$OUTFILE.h"
134
if ($TYPE_ARG =~ /^gettext\/(.*)/) {
139
## Sub for printing release information
142
${PROGRAM} (${PACKAGE}) $VERSION
143
Copyright (C) 2000, 2003 Free Software Foundation, Inc.
144
Written by Kenneth Christiansen, 2000.
146
This is free software; see the source for copying conditions. There is NO
147
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
152
## Sub for printing usage information
155
Usage: ${PROGRAM} [OPTION]... [FILENAME]
156
Generates a header file from an XML source file.
158
It grabs all strings between <_translatable_node> and its end tag in
159
XML files. Read manpage (man ${PROGRAM}) for more info.
161
--type=TYPE Specify the file type of FILENAME. Currently supports:
162
"gettext/glade", "gettext/ini", "gettext/keys"
163
"gettext/rfc822deb", "gettext/schemas",
164
"gettext/scheme", "gettext/xml", "gettext/quoted"
165
-l, --local Writes output into current working directory
166
(conflicts with --update)
167
--update Writes output into the same directory the source file
168
reside (conflicts with --local)
169
--srcdir Root of the source tree
170
-v, --version Output version information and exit
171
-h, --help Display this help and exit
172
-q, --quiet Quiet mode
174
Report bugs to http://bugzilla.gnome.org/ (product name "$PACKAGE")
175
or send email to <xml-i18n-tools\@gnome.org>.
180
## Sub for printing error messages
182
print STDERR "Try `${PROGRAM} --help' for more information.\n";
187
print "Generating C format header file for translation.\n" unless $QUIET_ARG;
195
open OUT, ">$OUTFILE";
196
binmode (OUT) if $^O eq 'MSWin32';
200
print "Wrote $OUTFILE\n" unless $QUIET_ARG;
208
local $/; #slurp mode
209
open (IN, "<$SRCDIR_ARG/$FILE") || die "can't open $SRCDIR_ARG/$FILE: $!";
213
&type_ini if $gettext_type eq "ini";
214
&type_keys if $gettext_type eq "keys";
215
&type_xml if $gettext_type eq "xml";
216
&type_glade if $gettext_type eq "glade";
217
&type_scheme if $gettext_type eq "scheme";
218
&type_schemas if $gettext_type eq "schemas";
219
&type_rfc822deb if $gettext_type eq "rfc822deb";
220
&type_quoted if $gettext_type eq "quoted";
223
sub entity_decode_minimal
249
return '\"' if $_ eq '"';
250
return '\n' if $_ eq "\n";
251
return '\\\\' if $_ eq '\\';
259
return join "", map &escape_char, split //, $string;
263
### For generic translatable desktop files ###
264
while ($input =~ /^(#(.+)\n)?^_.*=(.*)$/mg) {
273
### For generic translatable mime/keys files ###
274
while ($input =~ /^\s*_\w+=(.*)$/mg) {
280
### For generic translatable XML files ###
281
my $tree = readXml($input);
287
my $vartype = ref $var;
289
if ($vartype =~ /ARRAY/) {
292
foreach my $el (@arr) {
297
} elsif ($vartype =~ /HASH/) {
300
foreach my $key (keys %hash) {
302
print_var($hash{$key});
311
# Same syntax as getAttributeString in intltool-merge.in.in, similar logic (look for ## differences comment)
312
sub getAttributeString
315
my $do_translate = shift || 1;
316
my $language = shift || "";
317
my $translate = shift;
319
foreach my $e (reverse(sort(keys %{ $sub }))) {
321
my $string = $sub->{$e};
324
$string =~ s/^[\s]+//;
325
$string =~ s/[\s]+$//;
327
if ($string =~ /^'.*'$/)
331
$string =~ s/^['"]//g;
332
$string =~ s/['"]$//g;
334
## differences from intltool-merge.in.in
336
$comments{entity_decode($string)} = $XMLCOMMENT if $XMLCOMMENT;
337
$messages{entity_decode($string)} = [];
340
## differences end here from intltool-merge.in.in
341
$result .= " $key=$quote$string$quote";
346
# Verbatim copy from intltool-merge.in.in
350
my $spacepreserve = shift || 0;
351
my @list = @{ $ref };
354
my $count = scalar(@list);
355
my $attrs = $list[0];
358
$spacepreserve = 1 if ((exists $attrs->{"xml:space"}) && ($attrs->{"xml:space"} =~ /^["']?preserve["']?$/));
359
$spacepreserve = 0 if ((exists $attrs->{"xml:space"}) && ($attrs->{"xml:space"} =~ /^["']?default["']?$/));
361
while ($index < $count) {
362
my $type = $list[$index];
363
my $content = $list[$index+1];
367
# lets strip the whitespace here, and *ONLY* here
368
$content =~ s/\s+/ /gs if (!$spacepreserve);
371
} elsif ( "$type" ne "1" ) {
372
# We've got another element
374
$result .= getAttributeString(@{$content}[0], 0); # no nested translatable elements
376
my $subresult = getXMLstring($content, $spacepreserve);
378
$result .= ">".$subresult . "</$type>";
391
# Verbatim copy from intltool-merge.in.in, except for MULTIPLE_OUTPUT handling removed
392
# Translate list of nodes if necessary
393
sub translate_subnodes
397
my $language = shift || "";
398
my $singlelang = shift || 0;
399
my $spacepreserve = shift || 0;
401
my @nodes = @{ $content };
403
my $count = scalar(@nodes);
405
while ($index < $count) {
406
my $type = $nodes[$index];
407
my $rest = $nodes[$index+1];
408
traverse($fh, $type, $rest, $language, $spacepreserve);
413
# Based on traverse() in intltool-merge.in.in
416
my $fh = shift; # unused, to allow us to sync code between -merge and -extract
417
my $nodename = shift;
419
my $language = shift || "";
420
my $spacepreserve = shift || 0;
422
if ($nodename && "$nodename" eq "1") {
423
$XMLCOMMENT = $content;
424
} elsif ($nodename) {
426
my @all = @{ $content };
427
my $attrs = shift @all;
429
my $outattr = getAttributeString($attrs, 1, $language, \$translate);
431
if ($nodename =~ /^_/) {
437
$spacepreserve = 0 if ((exists $attrs->{"xml:space"}) && ($attrs->{"xml:space"} =~ /^["']?default["']?$/));
438
$spacepreserve = 1 if ((exists $attrs->{"xml:space"}) && ($attrs->{"xml:space"} =~ /^["']?preserve["']?$/));
441
$lookup = getXMLstring($content, $spacepreserve);
442
if (!$spacepreserve) {
443
$lookup =~ s/^\s+//s;
444
$lookup =~ s/\s+$//s;
447
if ($lookup && $translate != 2) {
448
$comments{$lookup} = $XMLCOMMENT if $XMLCOMMENT;
449
$messages{$lookup} = [];
450
} elsif ($translate == 2) {
451
translate_subnodes($fh, \@all, $language, 1, $spacepreserve);
455
my $count = scalar(@all);
458
while ($index < $count) {
459
my $type = $all[$index];
460
my $rest = $all[$index+1];
461
traverse($fh, $type, $rest, $language, $spacepreserve);
471
# Verbatim copy from intltool-merge.in.in, $fh for compatibility
476
my $language = shift || "";
478
my $name = shift @{ $ref };
479
my $cont = shift @{ $ref };
481
while (!$name || "$name" eq "1") {
482
$name = shift @{ $ref };
483
$cont = shift @{ $ref };
486
my $spacepreserve = 0;
487
my $attrs = @{$cont}[0];
488
$spacepreserve = 1 if ((exists $attrs->{"xml:space"}) && ($attrs->{"xml:space"} =~ /^["']?preserve["']?$/));
490
traverse($fh, $name, $cont, $language, $spacepreserve);
493
# Verbatim copy from intltool-merge.in.in
494
sub intltool_tree_comment
497
my $data = $expat->original_string();
498
my $clist = $expat->{Curlist};
503
push @$clist, 1 => $data;
506
# Verbatim copy from intltool-merge.in.in
507
sub intltool_tree_cdatastart
510
my $clist = $expat->{Curlist};
513
push @$clist, 0 => $expat->original_string();
516
# Verbatim copy from intltool-merge.in.in
517
sub intltool_tree_cdataend
520
my $clist = $expat->{Curlist};
523
$clist->[$pos] .= $expat->original_string();
526
# Verbatim copy from intltool-merge.in.in
527
sub intltool_tree_char
531
my $clist = $expat->{Curlist};
534
# Use original_string so that we retain escaped entities
537
if ($pos > 0 and $clist->[$pos - 1] eq '0') {
538
$clist->[$pos] .= $expat->original_string();
540
push @$clist, 0 => $expat->original_string();
544
# Verbatim copy from intltool-merge.in.in
545
sub intltool_tree_start
551
# Use original_string so that we retain escaped entities
552
# in attribute values. We must convert the string to an
553
# @origlist array to conform to the structure of the Tree
556
my @original_array = split /\x/, $expat->original_string();
557
my $source = $expat->original_string();
559
# Remove leading tag.
561
$source =~ s|^\s*<\s*(\S+)||s;
563
# Grab attribute key/value pairs and push onto @origlist array.
567
if ($source =~ /^\s*([\w:-]+)\s*[=]\s*["]/)
569
$source =~ s|^\s*([\w:-]+)\s*[=]\s*["]([^"]*)["]||s;
571
push @origlist, '"' . $2 . '"';
573
elsif ($source =~ /^\s*([\w:-]+)\s*[=]\s*[']/)
575
$source =~ s|^\s*([\w:-]+)\s*[=]\s*[']([^']*)[']||s;
577
push @origlist, "'" . $2 . "'";
585
my $ol = [ { @origlist } ];
587
push @{ $expat->{Lists} }, $expat->{Curlist};
588
push @{ $expat->{Curlist} }, $tag => $ol;
589
$expat->{Curlist} = $ol;
592
# Copied from intltool-merge.in.in and added comment handler.
595
my $xmldoc = shift || return;
596
my $ret = eval 'require XML::Parser';
598
die "You must have XML::Parser installed to run $0\n\n";
600
my $xp = new XML::Parser(Style => 'Tree');
601
$xp->setHandlers(Char => \&intltool_tree_char);
602
$xp->setHandlers(Start => \&intltool_tree_start);
603
$xp->setHandlers(CdataStart => \&intltool_tree_cdatastart);
604
$xp->setHandlers(CdataEnd => \&intltool_tree_cdataend);
606
## differences from intltool-merge.in.in
607
$xp->setHandlers(Comment => \&intltool_tree_comment);
608
## differences end here from intltool-merge.in.in
610
my $tree = $xp->parse($xmldoc);
613
# <foo><!-- comment --><head id="a">Hello <em>there</em></head><bar>Howdy<ref/></bar>do</foo>
615
# [foo, [{}, 1, "comment", head, [{id => "a"}, 0, "Hello ", em, [{}, 0, "there"]], bar,
616
# [{}, 0, "Howdy", ref, [{}]], 0, "do" ] ]
622
### For schemas XML files ###
624
# FIXME: We should handle escaped < (less than)
626
<locale\ name="C">\s*
627
(<default>\s*(?:<!--([^>]*?)-->\s*)?(.*?)\s*<\/default>\s*)?
628
(<short>\s*(?:<!--([^>]*?)-->\s*)?(.*?)\s*<\/short>\s*)?
629
(<long>\s*(?:<!--([^>]*?)-->\s*)?(.*?)\s*<\/long>\s*)?
632
my @totranslate = ($3,$6,$9);
633
my @eachcomment = ($2,$5,$8);
634
foreach (@totranslate) {
635
my $currentcomment = shift @eachcomment;
638
$messages{entity_decode_minimal($_)} = [];
639
$comments{entity_decode_minimal($_)} = $currentcomment if (defined($currentcomment));
645
### For rfc822-style Debian configuration files ###
649
while ($input =~ /\G(.*?)(^|\n)(_+)([^:]+):[ \t]*(.*?)(?=\n\S|$)/sg)
651
my ($pre, $newline, $underscore, $tag, $text) = ($1, $2, $3, $4, $5);
652
while ($pre =~ m/\n/g)
656
$lineno += length($newline);
657
my @str_list = rfc822deb_split(length($underscore), $text);
658
for my $str (@str_list)
661
$messages{$str} = [];
662
$loc{$str} = $lineno;
663
$count{$str} = $strcount;
664
my $usercomment = '';
665
while($pre =~ s/(^|\n)#([^\n]*)$//s)
667
$usercomment = "\n" . $2 . $usercomment;
669
$comments{$str} = $tag . $usercomment;
671
$lineno += ($text =~ s/\n//g);
675
sub rfc822deb_split {
676
# Debian defines a special way to deal with rfc822-style files:
677
# when a value contain newlines, it consists of
678
# 1. a short form (first line)
679
# 2. a long description, all lines begin with a space,
680
# and paragraphs are separated by a single dot on a line
681
# This routine returns an array of all paragraphs, and reformat
683
# When first argument is 2, the string is a comma separated list of
687
$text =~ s/^[ \t]//mg;
688
return (split(/, */, $text, 0)) if $type ne 1;
689
return ($text) if $text !~ /\n/;
691
$text =~ s/([^\n]*)\n//;
694
for my $line (split (/\n/, $text))
697
if ($line =~ /^\.\s*$/)
704
elsif ($line =~ /^\s/)
706
# Line which must not be reformatted
707
$str .= "\n" if length ($str) && $str !~ /\n$/;
713
# Continuation line, remove newline
714
$str .= " " if length ($str) && $str !~ /\n$/;
719
push(@list, $str) if length ($str);
724
while ($input =~ /\"(([^\"]|\\\")*[^\\\"])\"/g) {
727
$message =~ s/\\\"/\"/g;
728
$before =~ s/[^\n]//g;
729
$messages{$message} = [];
730
$loc{$message} = length ($before) + 2;
735
### For translatable Glade XML files ###
737
my $tags = "label|title|text|format|copyright|comments|preview_text|tooltip|message";
739
while ($input =~ /<($tags)>([^<]+)<\/($tags)>/sg) {
740
# Glade sometimes uses tags that normally mark translatable things for
741
# little bits of non-translatable content. We work around this by not
742
# translating strings that only includes something like label4 or window1.
743
$messages{entity_decode($2)} = [] unless $2 =~ /^(window|label|dialog)[0-9]+$/;
746
while ($input =~ /<items>(..[^<]*)<\/items>/sg) {
747
for my $item (split (/\n/, $1)) {
748
$messages{entity_decode($item)} = [];
752
## handle new glade files
753
while ($input =~ /<(property|atkproperty)\s+[^>]*translatable\s*=\s*"yes"(?:\s+[^>]*comments\s*=\s*"([^"]*)")?[^>]*>([^<]+)<\/\1>/sg) {
754
$messages{entity_decode($3)} = [] unless $3 =~ /^(window|label)[0-9]+$/;
755
if (defined($2) and !($3 =~ /^(window|label)[0-9]+$/)) {
756
$comments{entity_decode($3)} = entity_decode($2) ;
759
while ($input =~ /<atkaction\s+action_name="([^>]*)"\s+description="([^>]+)"\/>/sg) {
760
$messages{entity_decode_minimal($2)} = [];
765
my ($line, $i, $state, $str, $trcomment, $char);
766
for $line (split(/\n/, $input)) {
768
$state = 0; # 0 - nothing, 1 - string, 2 - translatable string
769
while ($i < length($line)) {
770
if (substr($line,$i,1) eq "\"") {
772
$comments{$str} = $trcomment if ($trcomment);
773
$messages{$str} = [];
775
$state = 0; $trcomment = "";
776
} elsif ($state == 1) {
778
$state = 0; $trcomment = "";
782
if ($i>0 && substr($line,$i-1,1) eq '_') {
787
if (substr($line,$i,1) eq ";") {
788
$trcomment = substr($line,$i+1);
789
$trcomment =~ s/^;*\s*//;
791
} elsif ($trcomment && substr($line,$i,1) !~ /\s|\(|\)|_/) {
795
if (substr($line,$i,1) eq "\\") {
796
$char = substr($line,$i+1,1);
797
if ($char ne "\"" && $char ne "\\") {
802
$str = $str . substr($line,$i,1);
813
@msgids = sort { $count{$a} <=> $count{$b} } keys %count;
817
@msgids = sort keys %messages;
819
for my $message (@msgids)
822
$offsetlines++ if $message =~ /%/;
823
if (defined ($comments{$message}))
825
while ($comments{$message} =~ m/\n/g)
830
print OUT "# ".($loc{$message} - $offsetlines). " \"$FILE\"\n"
831
if defined $loc{$message};
832
print OUT "/* ".$comments{$message}." */\n"
833
if defined $comments{$message};
834
print OUT "/* xgettext:no-c-format */\n" if $message =~ /%/;
836
my @lines = split (/\n/, $message, -1);
837
for (my $n = 0; $n < @lines; $n++)
841
print OUT "char *s = N_(\"";
848
print OUT escape($lines[$n]);
added
removed
intltool-extract.in
