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";
42 ## Scalars used by the option stuff
46 my $VERSION_ARG = "0";
54 my $gettext_type = "";
62 ## Use this instead of \w for XML files to handle more possible characters.
63 my $w = "[-A-Za-z0-9._:]";
70 "type=s" => \$TYPE_ARG,
71 "local|l" => \$LOCAL_ARG,
72 "help|h" => \$HELP_ARG,
73 "version|v" => \$VERSION_ARG,
74 "update" => \$UPDATE_ARG,
75 "quiet|q" => \$QUIET_ARG,
76 "srcdir=s" => \$SRCDIR_ARG,
83 ## This section will check for the different options.
85 sub split_on_argument {
93 } elsif ($LOCAL_ARG) {
97 } elsif ($UPDATE_ARG) {
101 } elsif (@ARGV > 0) {
114 $OUTFILE = "$FILE.h";
119 $OUTFILE = fileparse($FILE, ());
121 system("mkdir tmp/");
123 $OUTFILE = "./tmp/$OUTFILE.h"
127 if ($TYPE_ARG =~ /^gettext\/(.*)/) {
132 ## Sub for printing release information
135 ${PROGRAM} (${PACKAGE}) $VERSION
136 Copyright (C) 2000, 2003 Free Software Foundation, Inc.
137 Written by Kenneth Christiansen, 2000.
139 This is free software; see the source for copying conditions. There is NO
140 warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
145 ## Sub for printing usage information
148 Usage: ${PROGRAM} [OPTION]... [FILENAME]
149 Generates a header file from an XML source file.
151 It grabs all strings between <_translatable_node> and its end tag in
152 XML files. Read manpage (man ${PROGRAM}) for more info.
154 --type=TYPE Specify the file type of FILENAME. Currently supports:
155 "gettext/glade", "gettext/ini", "gettext/keys"
156 "gettext/rfc822deb", "gettext/schemas",
157 "gettext/scheme", "gettext/xml"
158 -l, --local Writes output into current working directory
159 (conflicts with --update)
160 --update Writes output into the same directory the source file
161 reside (conflicts with --local)
162 --srcdir Root of the source tree
163 -v, --version Output version information and exit
164 -h, --help Display this help and exit
165 -q, --quiet Quiet mode
167 Report bugs to http://bugzilla.gnome.org/ (product name "$PACKAGE")
168 or send email to <xml-i18n-tools\@gnome.org>.
173 ## Sub for printing error messages
175 print STDERR "Try `${PROGRAM} --help' for more information.\n";
180 print "Generating C format header file for translation.\n" unless $QUIET_ARG;
188 open OUT, ">$OUTFILE";
192 print "Wrote $OUTFILE\n" unless $QUIET_ARG;
200 local $/; #slurp mode
201 open (IN, "<$SRCDIR_ARG/$FILE") || die "can't open $SRCDIR_ARG/$FILE: $!";
205 &type_ini if $gettext_type eq "ini";
206 &type_keys if $gettext_type eq "keys";
207 &type_xml if $gettext_type eq "xml";
208 &type_glade if $gettext_type eq "glade";
209 &type_scheme if $gettext_type eq "scheme";
210 &type_schemas if $gettext_type eq "schemas";
211 &type_rfc822deb if $gettext_type eq "rfc822deb";
214 sub entity_decode_minimal
240 return '\"' if $_ eq '"';
241 return '\n' if $_ eq "\n";
242 return '\\' if $_ eq '\\';
250 return join "", map &escape_char, split //, $string;
254 ### For generic translatable desktop files ###
255 while ($input =~ /^_.*=(.*)$/mg) {
261 ### For generic translatable mime/keys files ###
262 while ($input =~ /^\s*_\w+=(.*)$/mg) {
268 ### For generic translatable XML files ###
270 while ($input =~ /(?:<!--([^>]*?)-->[^\n]*\n?[^\n]*)?\s_$w+\s*=\s*\"([^"]*)\"/sg) { # "
271 $messages{entity_decode_minimal($2)} = [];
272 $comments{entity_decode_minimal($2)} = $1 if (defined($1));
275 while ($input =~ /(?:<!--([^>]*?)-->\s*)?<_($w+)(?: xml:space="($w+)")?[^>]*>(.*?)<\/_\2>/sg) {
277 if (!defined($3) || $3 ne "preserve") {
283 $comments{$_} = $1 if (defined($1));
288 ### For schemas XML files ###
290 # FIXME: We should handle escaped < (less than)
292 <locale\ name="C">\s*
293 (<default>\s*(?:<!--([^>]*?)-->\s*)?(.*?)\s*<\/default>\s*)?
294 (<short>\s*(?:<!--([^>]*?)-->\s*)?(.*?)\s*<\/short>\s*)?
295 (<long>\s*(?:<!--([^>]*?)-->\s*)?(.*?)\s*<\/long>\s*)?
298 my @totranslate = ($3,$6,$9);
299 my @eachcomment = ($2,$5,$8);
300 foreach (@totranslate) {
301 my $currentcomment = shift @eachcomment;
304 $messages{entity_decode_minimal($_)} = [];
305 $comments{entity_decode_minimal($_)} = $currentcomment if (defined($currentcomment));
311 ### For rfc822-style Debian configuration files ###
315 while ($input =~ /\G(.*?)(^|\n)(_+)([^:]+):[ \t]*(.*?)(?=\n\S|$)/sg)
317 my ($pre, $newline, $underscore, $tag, $text) = ($1, $2, $3, $4, $5);
318 while ($pre =~ m/\n/g)
322 $lineno += length($newline);
323 my @str_list = rfc822deb_split(length($underscore), $text);
324 for my $str (@str_list)
327 $messages{$str} = [];
328 $loc{$str} = $lineno;
329 $count{$str} = $strcount;
330 my $usercomment = '';
331 while($pre =~ s/(^|\n)#([^\n]*)$//s)
333 $usercomment = "\n" . $2 . $usercomment;
335 $comments{$str} = $tag . $usercomment;
337 $lineno += ($text =~ s/\n//g);
341 sub rfc822deb_split {
342 # Debian defines a special way to deal with rfc822-style files:
343 # when a value contain newlines, it consists of
344 # 1. a short form (first line)
345 # 2. a long description, all lines begin with a space,
346 # and paragraphs are separated by a single dot on a line
347 # This routine returns an array of all paragraphs, and reformat
349 # When first argument is 2, the string is a comma separated list of
353 $text =~ s/^[ \t]//mg;
354 return (split(/, */, $text, 0)) if $type ne 1;
355 return ($text) if $text !~ /\n/;
357 $text =~ s/([^\n]*)\n//;
360 for my $line (split (/\n/, $text))
363 if ($line =~ /^\.\s*$/)
370 elsif ($line =~ /^\s/)
372 # Line which must not be reformatted
373 $str .= "\n" if length ($str) && $str !~ /\n$/;
379 # Continuation line, remove newline
380 $str .= " " if length ($str) && $str !~ /\n$/;
385 push(@list, $str) if length ($str);
390 ### For translatable Glade XML files ###
392 my $tags = "label|title|text|format|copyright|comments|preview_text|tooltip|message";
394 while ($input =~ /<($tags)>([^<]+)<\/($tags)>/sg) {
395 # Glade sometimes uses tags that normally mark translatable things for
396 # little bits of non-translatable content. We work around this by not
397 # translating strings that only includes something like label4 or window1.
398 $messages{entity_decode($2)} = [] unless $2 =~ /^(window|label|dialog)[0-9]+$/;
401 while ($input =~ /<items>(..[^<]*)<\/items>/sg) {
402 for my $item (split (/\n/, $1)) {
403 $messages{entity_decode($item)} = [];
407 ## handle new glade files
408 while ($input =~ /<(property|atkproperty)\s+[^>]*translatable\s*=\s*"yes"(?:\s+[^>]*comments\s*=\s*"([^"]*)")?[^>]*>([^<]+)<\/\1>/sg) {
409 $messages{entity_decode($3)} = [] unless $3 =~ /^(window|label)[0-9]+$/;
410 if (defined($2) and !($3 =~ /^(window|label)[0-9]+$/)) {
411 $comments{entity_decode($3)} = entity_decode($2) ;
414 while ($input =~ /<atkaction\s+action_name="([^>]*)"\s+description="([^>]+)"\/>/sg) {
415 $messages{entity_decode_minimal($2)} = [];
420 my ($line, $i, $state, $str, $trcomment, $char);
421 for $line (split(/\n/, $input)) {
423 $state = 0; # 0 - nothing, 1 - string, 2 - translatable string
424 while ($i < length($line)) {
425 if (substr($line,$i,1) eq "\"") {
427 $comments{$str} = $trcomment if ($trcomment);
428 $messages{$str} = [];
430 $state = 0; $trcomment = "";
431 } elsif ($state == 1) {
433 $state = 0; $trcomment = "";
437 if ($i>0 && substr($line,$i-1,1) eq '_') {
442 if (substr($line,$i,1) eq ";") {
443 $trcomment = substr($line,$i+1);
444 $trcomment =~ s/^;*\s*//;
446 } elsif ($trcomment && substr($line,$i,1) !~ /\s|\(|\)|_/) {
450 if (substr($line,$i,1) eq "\\") {
451 $char = substr($line,$i+1,1);
452 if ($char ne "\"" && $char ne "\\") {
457 $str = $str . substr($line,$i,1);
468 @msgids = sort { $count{$a} <=> $count{$b} } keys %count;
472 @msgids = sort keys %messages;
474 for my $message (@msgids)
477 $offsetlines++ if $message =~ /%/;
478 if (defined ($comments{$message}))
480 while ($comments{$message} =~ m/\n/g)
485 print OUT "# ".($loc{$message} - $offsetlines). " \"$FILE\"\n"
486 if defined $loc{$message};
487 print OUT "/* ".$comments{$message}." */\n"
488 if defined $comments{$message};
489 print OUT "/* xgettext:no-c-format */\n" if $message =~ /%/;
491 my @lines = split (/\n/, $message, -1);
492 for (my $n = 0; $n < @lines; $n++)
496 print OUT "char *s = N_(\"";
503 print OUT escape($lines[$n]);