D7net
Home
Console
Upload
information
Create File
Create Folder
About
Tools
:
/
usr
/
bin
/
Filename :
po4a
back
Copy
#!/usr/bin/perl eval 'exec perl -S $0 ${1+"$@"}' if $running_under_some_shell; # po4a -- Update both the po files and translated documents in one shoot # # Copyright 2002-2020 by SPI, inc. # # This program is free software; you can redistribute it and/or modify it # under the terms of GPL (see COPYING). =encoding UTF-8 =head1 NAME po4a - update both the PO files and translated documents in one shot =head1 SYNOPSIS B<po4a> [I<options>] I<config_file> =head1 DESCRIPTION po4a (PO for anything) eases the maintenance of documentation translation using the classical gettext tools. The main feature of po4a is that it decouples the translation of content from its document structure. Please refer to the page L<po4a(7)> for a gentle introduction to this project. When you run the B<po4a> program for the first time, with only a configuration file and the documents to translate (called master documents), it produces a POT file (also called translation template) that contains all of the translatable strings in the document in a form that eases the work of translators. Those POT files can either be translated with a specific editor such as the B<GNOME Translation Editor>, KDE's B<Lokalize> or B<poedit>, or they can be integrated in an online localization platform such as B<weblate> or B<pootle>. The translation result is a set of PO files, one per language. When you run the B<po4a> program with both the master documents and the PO files, it produces the translated documents by injecting the content's translation (found in the PO files) into the structure of the original master document. If the master documents changed in the meanwhile, po4a will update the PO and POT files accordingly, so that the translators can easily detect the modifications and update their work. Depending on your settings, po4a will discard the partially translated documents, or produce a document mixing English (for the new or modified paragraphs) and the target language (for paragraphs where translation is already in the PO file). By default, the translated documents are produced when at least 80% of their content is translated (see the I<--keep> option below). Discarding translations as soon as they are not 100% may be discouraging for the translators, while showing "translations" that are too incomplete may be troubling for the end users. =head2 Graphical overview master documents ---+---->-------->---------+ (doc authoring) | | V (po4a executions) >-----+--> translations | | | existing PO files -->--> updated PO files >-+ | ^ | | | V | +----------<---------<-------+ ^ (manual translation process) | | addendum -->--------------------------------------+ The master documents are authored by the documentation writers. Any changes are automatically reflected by po4a in the PO files, that are then updated by the translators. All changes to the PO files (either manual or by po4a) are automatically reflected in translated documents. You can mimic this behavior using the L<po4a-updatepo(1)> and L<po4a-translate(1)> scripts in makefiles, but this quickly becomes bothersome and repetitive (see L<po4a(7)>). It is highly recommended to use the B<po4a> program in your build process. =head1 OPTIONS =over 4 =item B<-k>, B<--keep> Minimal threshold for translation percentage to keep (i.e. write) the resulting file (default: 80). I.e. by default, files have to be translated at least at 80% to be written on disk. =item B<-h>, B<--help> Show a short help message. =item B<-M>, B<--master-charset> Charset of the files containing the documents to translate. Note that all master documents must use the same charset. =item B<-L>, B<--localized-charset> Charset of the files containing the localized documents. Note that all translated documents will use the same charset. =item B<-A>, B<--addendum-charset> Charset of the addenda. Note that all the addenda should be in the same charset. =item B<-V>, B<--version> Display the version of the script and exit. =item B<-v>, B<--verbose> Increase the verbosity of the program. =item B<-q>, B<--quiet> Decrease the verbosity of the program. =item B<-d>, B<--debug> Output some debugging information. =item B<-o>, B<--option> Extra option(s) to pass to the format plugin. See the documentation of each plugin for more information about the valid options and their meanings. For example, you could pass '-o tablecells' to the AsciiDoc parser, while the text parser would accept '-o tabs=split'. =item B<-f>, B<--force> Always generate the POT and PO files, even if B<po4a> considers it is not necessary. The default behavior (when B<--force> is not specified) is the following: =over If the POT file already exists, it is regenerated if a master document or the configuration file is more recent (unless B<--no-update> is provided). The POT file is also written in a temporary document and B<po4a> verifies that the changes are really needed. Also, a translation is regenerated only if its master document, the PO file, one of its addenda or the configuration file is more recent. To avoid trying to regenerate translations which do not pass the threshold test (see B<--keep>), a file with the F<.po4a-stamp> extension can be created (see B<--stamp>). =back If a master document includes files, you should use the B<--force> flag because the modification time of these included files are not taken into account. The PO files are always re-generated based on the POT with B<msgmerge -U>. =item B<--stamp> Tells B<po4a> to create stamp files when a translation is not generated because it does not reach the threshold. These stamp files are named according to the expected translated document, with the F<.po4a-stamp> extension. Note: This only activates the creation of the F<.po4a-stamp> files. The stamp files are always used if they exist, and they are removed with B<--rm-translations> or when the file is finally translated. =item B<--no-translations> Do not generate the translated documents, only update the POT and PO files. =item B<--no-update> Do not change the POT and PO files, only the translation may be updated. =item B<--keep-translations> Keeps the existing translation files even if the translation doesn't meet the threshold specified by B<--keep>. This option does not create new translation files with few content, but it will save existing translations which decay because of changes to the master files. WARNING: This flag changes the po4a behavior in a rather drastic way: your translated files will not get updated at all until the translation improves. Only use this flag if you prefer shipping an outdated translated documentation rather than only shipping an accurate untranslated documentation. =item B<--rm-translations> Remove the translated files (implies B<--no-translations>). =item B<--no-backups> This flag does nothing since 0.41, and may be removed in later releases. =item B<--rm-backups> This flag does nothing since 0.41, and may be removed in later releases. =item B<--translate-only> I<translated-file> Translate only the specified file. It may be useful to speed up processing if a configuration file contains a lot of files. Note that this option does not update PO and POT files. This option can be used multiple times. =item B<--variable> I<var>B<=>I<value> Define a variable that will be expanded in the B<po4a> configuration file. Every occurrence of I<$(var)> will be replaced by I<value>. This option can be used multiple times. =item B<--srcdir> I<SRCDIR> Set the base directory for all input documents specified in the B<po4a> configuration file. If both I<destdir> and I<srcdir> are specified, input files are searched in the following directories, in order: I<destdir>, the current directory and I<srcdir>. Output files are written to I<destdir> if specified, or to the current directory. =item B<--destdir> I<DESTDIR> Set the base directory for all the output documents specified in the B<po4a> configuration file (see B<--srcdir> above). =back =head2 Options modifying the POT header =over 4 =item B<--porefs> I<type> Specify the reference format. Argument I<type> can be one of B<never> to not produce any reference, B<file> to only specify the file without the line number, B<counter> to replace line number by an increasing counter, and B<full> to include complete references (default: full). =item B<--wrap-po> B<no>|B<newlines>|I<number> (default: 76) Specify how the po file should be wrapped. This gives the choice between either files that are nicely wrapped but could lead to git conflicts, or files that are easier to handle automatically, but harder to read for humans. Historically, the gettext suite has reformatted the po files at the 77th column for cosmetics. This option specifies the behavior of po4a. If set to a numerical value, po4a will wrap the po file after this column and after newlines in the content. If set to B<newlines>, po4a will only split the msgid and msgstr after newlines in the content. If set to B<no>, po4a will not wrap the po file at all. The reference comments are always wrapped by the gettext tools that we use internally. Note that this option has no impact on how the msgid and msgstr are wrapped, ie on how newlines are added to the content of these strings. =item B<--master-language> Language of the source files containing the documents to translate. Note that all master documents must use the same language. =item B<--msgid-bugs-address> I<email@address> Set the report address for msgid bugs. By default, the created POT files have no Report-Msgid-Bugs-To fields. =item B<--copyright-holder> I<string> Set the copyright holder in the POT header. The default value is "Free Software Foundation, Inc." =item B<--package-name> I<string> Set the package name for the POT header. The default is "PACKAGE". =item B<--package-version> I<string> Set the package version for the POT header. The default is "VERSION". =back =head2 Options to modify the PO files =over 4 =item B<--msgmerge-opt> I<options> Extra options for B<msgmerge>(1). Note: B<$lang> will be extended to the current language. =item B<--no-previous> This option removes B<--previous> from the options passed to B<msgmerge>. This permits to support versions of B<gettext> earlier than 0.16. =item B<--previous> This option adds B<--previous> to the options passed to B<msgmerge>. It requires B<gettext> 0.16 or later, and is activated by default. =back =head1 CONFIGURATION FILE po4a expects a configuration file as argument. This file must contain the following elements: =over =item The path to the PO files and the list of languages existing in the project; =item Optionally, some global options and so-called configuration aliases that are used as templates to configure individual master files; =item The list of each master file to translate, along with specific parameters. =back All lines contain a command between square braces, followed by its parameters. Comments begin with the char '#' and run until the end of the line. You can escape the end of line to spread a command over several lines. Some full examples are presented on this page, while other examples can be found in the C<t/cfg> directory of the source distribution. =head2 Finding the PO and POT files The simplest solution is to explicitly give the path to POT and PO files, as follows: [po4a_paths] man/po/project.pot de:man/po/de.po fr:man/po/fr.po This specifies the path to the POT file first, and then the paths to the German and French PO files. The same information can be written as follows to reduce the risk of copy/paste errors: [po4a_langs] fr de [po4a_paths] man/po/project.pot $lang:man/po/$lang.po The C<$lang> component is automatically expanded using the provided languages list, reducing the risk of copy/paste error when a new language is added. You can further compact the same information by only providing the path to the directory containing your translation project, as follows. [po_directory] man/po/ The provided directory must contain a set of PO files, each named F<XX.po> with C<XX> the ISO 639-1 of the language used in this file. The directory must also contain a single POT file, with the C<.pot> file extension. For the first run, this file can be empty but it must exist (po4a cannot guess the name to use before the extension). Note that you must choose only one between C<po_directory> and C<po4a_paths>. The first one (C<po_directory>) is more compact, further reduces the risk of copy/paste error, but forces you to use the expected project structure and file names. The second one (C<po4a_paths>), is more explicit, probably more readable, and advised when you setup your first project with po4a. =head3 Centralized or split PO files? By default, po4a produces one single PO file per target language, containing the whole content of your translation project. As your project grows, the size of these files may become problematic. When using weblate, it is possible to specify priorities for each translation segment (i.e., msgid) so that the important ones get translated first. Still, some translation teams prefer to split the content in several files. To have one PO file per master file, you simply have to use the string C<$master> in the name of your PO files on the C<[po4a_paths]> line, as follows. [po4a_paths] doc/$master/$master.pot $lang:doc/$master/$lang.po With this line, po4a will produce separate POT and PO files for each document to translate. For example, if you have 3 documents and 5 languages, this will result in 3 POT files and 15 PO files. These files are named as specified on the C<po4a_paths> template, with C<$master> substituted to the basename of each document to translate. In case of name conflict, you can specify the POT file to use as follows, with the C<pot=> parameter. This feature can also be used to group several translated files into the same POT file. [po4a_langs] de fr ja [po4a_paths] l10n/po/$master.pot $lang:l10n/po/$master.$lang.po [type: xml] foo/gui.xml $lang:foo/gui.$lang.xml pot=foo-gui [type: xml] bar/gui.xml $lang:bar/gui.$lang.xml pot=bar [type: xml] bar/cli.xml $lang:bar/cli.$lang.xml pot=bar In split mode, B<po4a> builds a temporary compendium during the PO update, to share the translations between all the PO files. If two PO files have different translations for the same string, B<po4a> will mark this string as fuzzy and will submit both translations in all the PO files containing this string. When unfuzzied by the translator, the translation is automatically used in every PO files. =head2 Specifying the documents to translate You must also list the documents that should be translated. For each master file, you must specify the format parser to use, the location of the translated document to produce, and optionally some configuration. Here is an example: [type: sgml] doc/my_stuff.sgml fr:doc/fr/mon_truc.sgml \ de:doc/de/mein_kram.sgml [type: man] script fr:doc/fr/script.1 de:doc/de/script.1 [type: docbook] doc/script.xml fr:doc/fr/script.xml \ de:doc/de/script.xml But again, these complex lines are difficult to read and modify, e.g. when adding a new language. It is much simpler to reorganize things using the C<$lang> template as follows: [type: sgml] doc/my_stuff.sgml $lang:doc/$lang/my_stuff.sgml [type: man] script.1 $lang:po/$lang/script.1 [type: docbook] doc/script.xml $lang:doc/$lang/script.xml =head2 Specifying options There is two types of options: I<po4a options> are default values to the po4a command line options while I<format options> are used to change the behavior of the format parsers. As a I<po4a options>, you could for example specify in your configuration file that the default value of the B<--keep> command line parameter is 50% instead of 80%. I<Format options> are documented on the specific page of each parsing module, e.g. L<Locale::Po4a::Xml(3pm)>. You could for example pass B<nostrip> to the XML parser to not strip the spaces around the extracted strings. You can pass these options for a specific master file, or even for a specific translation of that file, using C<opt:> and C<opt_XX:> for the C<XX> language. In the following example, the B<nostrip> option is passed to the XML parser (for all languages), while the threshold will be reduced to 0% for the French translation (that is thus always kept). [type:xml] toto.xml $lang:toto.$lang.xml opt:"-o nostrip" opt_fr:"--keep 0" In any case, these configuration chunks must be located at the end of the line. The declaration of files must come first, then the addendum if any (see below), and then only the options. The grouping of configuration chunks is not very important, since elements are internally concatenated as strings. The following examples are all equivalent: [type:xml] toto.xml $lang:toto.$lang.xml opt:"--keep 20" opt:"-o nostrip" opt_fr:"--keep 0" [type:xml] toto.xml $lang:toto.$lang.xml opt:"--keep 20 -o nostrip" opt_fr:"--keep 0" [type:xml] toto.xml $lang:toto.$lang.xml opt:--keep opt:20 opt:-o opt:nostrip opt_fr:--keep opt_fr:0 Note that language specific options are not used when building the POT file. It is for example impossible to pass B<nostrip> to the parser only when building the French translation, because the same POT file is used to update every languages. So the only options that can be language-specific are the ones that are used when producing the translation, as the C<--keep> option. =head3 Configuration aliases To pass the same options to several files, the best is to define a type alias as follows. In the next example, C<--keep 0> is passed to every Italian translation using this C<test> type, that is an extension of the C<man> type. [po4a_alias:test] man opt_it:"--keep 0" [type: test] man/page.1 $lang:man/$lang/page.1 You can also extend an existing type reusing the same name for the alias as follows. This is not interpreted as as an erroneous recursive definition. [po4a_alias:man] man opt_it:"--keep 0" [type: man] man/page.1 $lang:man/$lang/page.1 =head3 Global default options You can also use C<[options]> lines to define options that must be used for all files, regardless of their type. [options] --keep 20 --option nostrip As with the command line options, you can abbreviate the parameters passed in the configuration file: [options] -k 20 -o nostrip =head3 Option priorities The options of every sources are concatenated, ensuring that the default values can easily be overridden by more specific options. The order is as follows: =over =item C<[options]> lines provide default values that can be overridden by any other source. =item Type aliases are then used. Language specific settings override the ones applicable to all languages. =item Settings that are specific to a given master file override both the default ones and the ones coming from the type alias. In this case also, language specific settings override the global ones. =item Finally, parameters provided on the B<po4a> command line override any settings from the configuration file. =back =head3 Example Here is an example showing how to quote the spaces and quotes: [po_directory] man/po/ [options] --master-charset UTF-8 [po4a_alias:man] man opt:"-o \"mdoc=NAME,SEE ALSO\"" [type:man] t-05-config/test02_man.1 $lang:tmp/test02_man.$lang.1 \ opt:"-k 75" opt_it:"-L UTF-8" opt_fr:--verbose =head2 Addendum: Adding extra content in the translation If you want to add an extra section to the translation, for example to give credit to the translator, then you need to define an addendum to the line defining your master file. Please refer to the page L<po4a(7)> for more details on the syntax of addendum files. [type: pod] script fr:doc/fr/script.1 \ add_fr:doc/l10n/script.fr.add You can also use language templates as follow: [type: pod] script $lang:doc/$lang/script.1 \ add_$lang:doc/l10n/script.$lang.add If an addendum fails to apply, the translation is discarded. =head3 Modifiers for the addendum declaration Addendum modifiers can simplify the configuration file in the case where not all languages provide an addendum, or when the list of addenda changes from one language to the other. The modifier is a single char located before the file name. =over 2 =item B<?> Include I<addendum_path> if this file does exist, otherwise do nothing. =item B<@> I<addendum_path> is not a regular addendum but a file containing a list of addenda, one by line. Each addendum may be preceded by modifiers. =item B<!> I<addendum_path> is discarded, it is not loaded and will not be loaded by any further addendum specification. =back The following includes an addendum in any language, but if only it exists. No error is reported if the addendum does not exist. [type: pod] script $lang:doc/$lang/script.1 add_$lang:?doc/l10n/script.$lang.add The following includes a list of addendum for every language: [type: pod] script $lang:doc/$lang/script.1 add_$lang:@doc/l10n/script.$lang.add =head2 Filtering the translated strings Sometimes, you want to hide some strings from the translation process. To that extend, you can give a C<pot_in> parameter to your master file to specify the name of the file to use instead of the real master when building the POT file. Here is an example: [type:docbook] book.xml \ pot_in:book-filtered.xml \ $lang:book.$lang.xml With this setting, the strings to translate will be extracted from the F<book-filtered.xml> (that must be produced before calling B<po4a>) while the translated files will be built from F<book.xml>. As a result, any string that is part of F<book.xml> but not in F<book-filtered.xml> will not be included in the PO files, preventing the translators from providing a translation for them. So these strings will be left unmodified when producing the translated documents. This naturally decreases the level of translation, so you may need the C<--keep> option to ensure that the document is produced anyway. =head2 CONFIGURATION EXAMPLE TODO: Is this section really useful? Let's assume you maintain a program named B<foo> which has a man page F<man/foo.1> which naturally is maintained in English only. Now you as the upstream or downstream maintainer want to create and maintain the translation. First you need to create the POT file necessary to send to translators using L<po4a-gettextize(1)>. So for our case we would call cd man && po4a-gettextize -f man -m foo.1 -p foo.pot You would then send this file to the appropriate language lists or offer it for download somewhere on your website. Now let's assume you received three translations before your next release: F<de.po> (including an addendum F<de.add>), F<sv.po> and F<pt.po>. Since you don't want to change your F<Makefile>(s) whenever a new translation arrives you can use B<po4a> with an appropriate configuration file in your F<Makefile>. Let's call it F<po4a.cfg>. In our example it would look like the following: [po_directory] man/po4a/po/ [type: man] man/foo.1 $lang:man/translated/$lang/foo.1 \ add_$lang:?man/po4a/add_$lang/$lang.add opt:"-k 80" In this example we assume that your generated man pages (and all PO and addenda files) should be stored in F<man/translated/$lang/> (respectively in F<man/po4a/po/> and F<man/po4a/add_$lang/>) below the current directory. In our example the F<man/po4a/po/> directory would include F<de.po>, F<pt.po> and F<sv.po>, and the F<man/po4a/add_de/> directory would include F<de.add>. Note the use of the modifier B<?> as only the German translation (F<de.po>) is accompanied by an addendum. To actually build the translated man pages you would then (once!) add the following line in the B<build> target of the appropriate F<Makefile>: po4a po4a.cfg Once this is set up you don't need to touch the F<Makefile> when a new translation arrives, i.e. if the French team sends you F<fr.po> and F<fr.add> then you simply drop them respectively in F<man/po4a/po/> and F<man/po4a/add_fr/> and the next time the program is built the French translation is automatically build as well in F<man/translated/fr/>. Note that you still need an appropriate target to install localized manual pages with English ones. Finally if you do not store generated files into your version control system, you will need a line in your B<clean> target as well: -rm -rf man/translated =head1 SEE ALSO L<po4a-gettextize(1)>, L<po4a-normalize(1)>, L<po4a-translate(1)>, L<po4a-updatepo(1)>, L<po4a(7)>. =head1 AUTHORS Denis Barbier <barbier@linuxfr.org> Nicolas Francois <nicolas.francois@centraliens.net> Martin Quinson (mquinson#debian.org) =head1 COPYRIGHT AND LICENSE Copyright 2002-2020 by SPI, inc. This program is free software; you may redistribute it and/or modify it under the terms of GPL (see the COPYING file). =cut use 5.006; use strict; use warnings; use Getopt::Long qw(GetOptions); use Locale::Po4a::Chooser; use Locale::Po4a::TransTractor; use Locale::Po4a::Common; use Locale::Po4a::Po qw(move_po_if_needed); use Pod::Usage qw(pod2usage); use File::Temp; use File::Basename; use File::Copy; use File::Spec; use Fcntl; # sysopen flags use Cwd; # cwd use Config; Locale::Po4a::Common::textdomain('po4a'); sub show_version { Locale::Po4a::Common::show_version("po4a"); exit 0; } # keep the command line arguments my @ORIGINAL_ARGV = @ARGV; # Use /NUL instead of /dev/null on Windows my $devnull = ( $^O =~ /Win/ ) ? '/NUL' : '/dev/null'; # Parse the options provided on the command line, or in argument sub get_options { if ( defined $_[0] ) { @ARGV = @_; # map {print "o: $_\n"} @ARGV; } else { @ARGV = @ORIGINAL_ARGV; } # temporary array for GetOptions my @verbose = (); my @options = (); my @variables = (); my $previous; my $noprevious; my $msgmerge_opt = ''; my %opts = ( "help" => 0, "type" => "", "debug" => 0, "verbose" => 0, "quiet" => 0, "force" => 0, "no-translations" => 0, "no-update" => 0, "keep-translations" => 0, "rm-translations" => 0, "no-backups" => 0, "rm-backups" => 0, "threshold" => 80, "mastlang" => "", "mastchar" => "", "locchar" => "", "addchar" => "", "options" => { "verbose" => 0, "debug" => 0 }, "variables" => {}, "partial" => [], "porefs" => "full", "wrap-po" => undef, "copyright-holder" => undef, "msgid-bugs-address" => undef, "package-name" => undef, "package-version" => undef, "srcdir" => undef, "destdir" => undef, "calldir" => cwd() ); Getopt::Long::config( 'bundling', 'no_getopt_compat', 'no_auto_abbrev' ); GetOptions( 'help|h' => \$opts{"help"}, 'master-language=s' => \$opts{"mastlang"}, 'master-charset|M=s' => \$opts{"mastchar"}, 'localized-charset|L=s' => \$opts{"locchar"}, 'addendum-charset|A=s' => \$opts{"addchar"}, 'verbose|v' => \@verbose, 'debug|d' => \$opts{"debug"}, 'force|f' => \$opts{"force"}, 'stamp' => \$opts{"stamp"}, 'quiet|q' => \$opts{"quiet"}, 'keep|k=s' => \$opts{"threshold"}, 'no-translations' => \$opts{"no-translations"}, 'no-update' => \$opts{"no-update"}, 'keep-translations' => \$opts{"keep-translations"}, 'rm-translations' => \$opts{"rm-translations"}, 'translate-only=s' => \@{ $opts{"partial"} }, 'no-backups' => \$opts{"no-backups"}, 'rm-backups' => \$opts{"rm-backups"}, 'version|V' => \&show_version, 'option|o=s' => \@options, 'variable=s' => \@variables, 'porefs=s' => \$opts{"porefs"}, 'wrap-po=s' => \$opts{"wrap-po"}, 'copyright-holder=s' => \$opts{"copyright-holder"}, 'msgid-bugs-address=s' => \$opts{"msgid-bugs-address"}, 'package-name=s' => \$opts{"package-name"}, 'package-version=s' => \$opts{"package-version"}, 'no-previous' => \$noprevious, 'previous' => \$previous, 'msgmerge-opt=s' => \$msgmerge_opt, 'srcdir=s' => \$opts{"srcdir"}, 'destdir=s' => \$opts{"destdir"} ) or pod2usage(); $opts{"verbose"} = scalar @verbose; $opts{"verbose"} = 0 if $opts{"quiet"}; $opts{"verbose"} ||= 1 if $opts{"debug"}; $opts{"msgmerge-opt"} = ''; $opts{"msgmerge-opt"} .= "--previous " unless $noprevious; $opts{"msgmerge-opt"} .= "--add-location=file " if ( $opts{'porefs'} =~ m/^file/ ); $opts{"msgmerge-opt"} .= $msgmerge_opt; # options to transmit to the modules $opts{"options"} = { "verbose" => $opts{"verbose"}, "debug" => $opts{"debug"} }; foreach (@options) { if (m/^([^=]*)=(.*)$/) { $opts{"options"}{$1} = "$2"; } else { $opts{"options"}{$_} = 1; } } foreach (@variables) { if (m/^([^=]*)=(.*)$/) { $opts{"variables"}{$1} = "$2"; } } # The rm- options imply the no- $opts{"no-translations"} = 1 if $opts{"rm-translations"}; if ( defined $opts{"srcdir"} and not -d $opts{"srcdir"} ) { die wrap_msg( gettext("Option %s invalid. Directory %s does not exist (current directory: %s)."), "srcdir", $opts{"srcdir"}, cwd() ); } if ( defined $opts{"destdir"} and not -d $opts{"destdir"} ) { die wrap_msg( gettext("Option %s invalid. Directory %s does not exist (current directory: %s)."), "destdir", $opts{"destdir"}, cwd() ); } # print STDERR "msgmerge: ".$opts{"msgmerge-opt"}."\n"; return %opts; } # Parse a config line and extract the parameters that correspond to options. # These options are appended to the options provided in argument (as a # reference to an hash). The options are sorted by category in this hash. # The categories are: global and the various languages. sub parse_config_options { my $ref = shift; # a line reference for the die messages my $line = shift; # the line to parse my $orig_line = $line; # keep the original line for die messages my $options = shift; # reference to an hash of options while ( defined $line and $line !~ m/^\s*$/ ) { if ( $line =~ m/^\s*(?:opt(?:_(.+?))?:)?\s*(.+?)\s*$/ ) { my $lang = $1 // 'global'; $line = $2; my $opt = ""; if ( $line =~ m/^\s*"(.+?(?<!\\)(?:\\\\)*)"(?:\s+(.+))?$/ ) { # take up to the next " not preceded by an odd number of \ $opt = $1; $line = $2; } elsif ( $line =~ m/^\s*([^\s]+?)(?:\s+(.+))?$/ ) { # Use the first space separated arg $opt = $1; $line = $2; } else { # TRANSLATOR: There is two chunks in the following: opt:"--keep 20" opt:"-o nowrap" die wrap_ref_mod( "$ref", "", gettext("Unparsable option chunk '%s' (%s)."), $line, $orig_line ); } if ( !defined $options->{$lang} ) { $options->{$lang} = $opt; } else { $options->{$lang} .= " $opt"; } } else { print STDERR wrap_ref_mod( "$ref", "", gettext("Skipping option chunk '%s'."), $line ); last; } } $line = "" unless defined $line; return $line; } my %po4a_opts = get_options(@ARGV); our ( $destdir, $srcdir, $calldir ) = ( $po4a_opts{'destdir'}, $po4a_opts{'srcdir'}, $po4a_opts{'calldir'} ); sub find_input_file { my $filename = $_[0]; my $debug = $_[1]; print STDERR "Search for input file $filename\n" if $debug; return $filename if ( File::Spec->file_name_is_absolute($filename) ); foreach ( ( $destdir, $calldir, $srcdir ) ) { next unless defined $_; my $p = File::Spec->catfile( $_, $filename ); print STDERR "Test $p: " . ( -e $p ? "OK" : "nope" ) . "\n" if defined $debug; return $p if -e $p; } return $filename; } sub find_output_file { my $filename = $_[0]; return $filename if ( File::Spec->file_name_is_absolute($filename) ); foreach ( ( $destdir, $calldir ) ) { next unless defined $_; return File::Spec->catfile( $_, $filename ) if -d $_ and -w $_; } return $filename; } # Argument check $po4a_opts{"help"} && pod2usage( -verbose => 1, -exitval => 0 ); sub run_cmd { my $cmd = shift; print $cmd. "\n" if $po4a_opts{"debug"}; my $out = qx/$cmd 2>&1/; unless ( $? == 0 ) { my $err = ""; if ( $? == -1 ) { $err = sprintf( gettext("failed to execute '%s': %s."), $cmd, $! ); } elsif ( $? & 127 ) { if ( $? & 128 ) { $err = sprintf( gettext("'%s' died with signal %d, with coredump."), $cmd, $? & 127 ); } else { $err = sprintf( gettext("'%s' died with signal %d, without coredump."), $cmd, $? & 127 ); } } else { $err = sprintf( gettext("'%s' exited with value %d."), $cmd, $? >> 8 ); } die wrap_msg( gettext("Error: %s"), $err ); } return $out; } my $config_file = shift(@ARGV) || pod2usage(); # Check file existence -e $config_file || die wrap_msg( gettext("File %s does not exist."), $config_file ); # Parse the config file my (@langs); my (%aliases); # module aliases ([po4a_alias:...] my $pot_filename = ""; # In split mode, this is the temp file name for the big POT my $pot_filename_cfg = ""; # This is never modified, useful in split mode to find the cfg content my (%po_filename); # po_files: '$lang'=>'$path' my (%add_filename); # Global addendum_files: '$lang'=>array of '$path' # The document hash table contains all the information about the files to translate. # Each key of the hash map a master file, ie a file to translate. For each such master file, we have a sub-hashmap containing: # 'format' => '$format': string designating the po4a module to use # '$lang' => '$path': one string entry per language, the path of the translated file # Optional parts: # 'add_$lang'=>('$path','$path'): one vector per language, the list of all addendum for that language # 'pot_in' => '$master': The master file to use in input when building the POT (allows to filter out some things) # 'pot' => the filename to use when building the POT file in split mode (when not provided, basename of master is used) my (%document); my $doc_count = 0; my %partial = ( 'master' => {}, 'files' => {}, 'lang' => {} ); open CONFIG, "<", "$config_file" or die wrap_msg( gettext("Cannot open %s: %s"), $config_file, $! ); my ( $line, $nb ) = ( "", 0 ); while (<CONFIG>) { $nb++; s/#.*//; $line .= $_; $line =~ s/\t/ /g; $line =~ s/ +/ /g; while ( $line =~ m/\$\((\w+)\)/ ) { if ( defined $po4a_opts{"variables"}{$1} ) { $line =~ s/\$\((\Q$1\E)\)/$po4a_opts{"variables"}{$1}/g; } else { die wrap_ref_mod( "$config_file:$nb", "", gettext("Unknown variable: %s"), $1 ); } } $line =~ s/^ //; $line =~ s/ $//; chomp($line); next if ( $line =~ s/\\$// ); next unless ( $line =~ /\S/ ); my $args = $line; die wrap_ref_mod( "$config_file:$nb", "", gettext("Syntax error: %s"), $line ) unless ( $args =~ s/^\[([^\]]*)\] *// ); my $cmd = $1; my $main = ( $args =~ s/^(\S+) *// ? $1 : "" ); my $pot_in = $main; # Default for POT file input is $main if (@langs) { # Expand the $lang templates my ($args2) = ""; foreach my $arg ( split( / /, $args ) ) { if ( $arg =~ /\$lang\b/ ) { # Expand for all the langs foreach my $lang (@langs) { my ($arg2) = $arg; $arg2 =~ s/\$lang\b/$lang/g; $args2 .= $arg2 . " "; } } else { # Leave the argument as is $args2 .= $arg . " "; } } $args = $args2; } print "cmd=[$cmd]; $main / $args\n" if $po4a_opts{"debug"}; if ( $cmd eq "po4a_paths" ) { die wrap_ref_mod( "$config_file:$nb", "", gettext("'%s' redeclared"), "po4a_path" ) if ( length $pot_filename ); $pot_filename = $main; $pot_filename_cfg = $main; warn wrap_ref_mod( "$config_file:$nb", "", gettext( "Your POT file '%s' file name does not end with the 'pot' extension. This is unusual and will trouble the translators." ), $pot_filename ) unless $pot_filename =~ m/\.pot$/; my @split_args = split( / /, $args ); warn wrap_ref_mod( "$config_file:$nb", "", gettext("No PO file provided as subsequent parameters to '%s'. Your documentation will not be translated."), "po4a_path" ) unless scalar @split_args > 0; foreach my $arg (@split_args) { die wrap_ref_mod( "$config_file:$nb", "", gettext("Unparsable argument '%s'."), $arg ) unless ( $arg =~ /^([^:]*):(.*)/ ); my ( $component, $path ) = ( $1, $2 ); if ( $component =~ m/^add_(.*)$/ ) { # This is an addendum declaration my $lang = $1; my @addlist = ($path); while ( scalar @addlist ) { my $elem = shift @addlist; my $modifiers; $elem =~ s/^([@!?]+)// and $modifiers = $1; if ( defined $modifiers ) { next if ( $modifiers =~ m/!/ ); next if ( $modifiers =~ m/\?/ and not -e find_input_file($elem) ); if ( $modifiers =~ m/@/ ) { open LIST, "<", find_input_file($elem) or die wrap_msg( gettext("Cannot open %s: %s"), $elem, $! ); while (<LIST>) { chomp; s/\s+$//; next if length($_) == 0 or $_ =~ m/^\s*#/; while (m/\$\((\w+)\)/) { if ( defined $po4a_opts{"variables"}{$1} ) { s/\$\((\Q$1\E)\)/$po4a_opts{"variables"}{$1}/g; } else { die wrap_ref_mod( "$config_file:$nb", "", gettext("Unknown variable: %s"), $1 ); } } push @addlist, $_; } close LIST; next; } } push @{ $add_filename{$lang} }, $elem; } } else { # Not an addendum. It must be a po filename $po_filename{$1} = $path; } } } elsif ( $cmd eq "po4a_langs" ) { die wrap_ref_mod( "$config_file:$nb", "", gettext("'%s' redeclared"), "po4a_langs" ) if (@langs); @langs = split( / /, $main . " " . $args ); } elsif ( $cmd eq "po_directory" ) { die wrap_ref_mod( "$config_file:$nb", "", gettext("The list of languages cannot be set twice.") ) if scalar @langs; die wrap_ref_mod( "$config_file:$nb", "", gettext("The POT file cannot be set twice.") ) if length $pot_filename; # Search for the candidate PO and POT files # The difficulty is that the directory itself must be searched in destdir, calldir, srcdir and then the PO files must be searched in there. # This mechanism is easy to get wrong (see eg #960892), so let's get cautionous and verbose on error. # Step 1. collect the files in every candidate po_directory we find. my $po_directory = $main; my @po_files; print STDERR "Searching for po_directory '$po_directory'." if $po4a_opts{"debug"}; if ( File::Spec->file_name_is_absolute($po_directory) ) { print STDERR wrap_msg( gettext("Using '%s' as a %s."), $po_directory, "po_directory" ) if $po4a_opts{"verbose"}; die wrap_ref_mod( "$config_file:$nb", "", gettext("'%s' is not a directory (current dir: %s)"), $po_directory, cwd() ) unless ( -d $po_directory ); opendir PO_DIR, find_input_file($po_directory) or die wrap_ref_mod( "$config_file:$nb", "", gettext("Cannot list the '%s' directory"), $po_directory ); map { push @po_files, $_ if -e $_; } ( sort readdir PO_DIR ); } else { my $found = 0; foreach my $basedir ( $destdir, $calldir // ".", $srcdir ) { next unless defined $basedir; my $candidate = "$basedir/$po_directory"; if ( -e $candidate && -d $candidate ) { $found = 1; print STDERR wrap_msg( gettext("Using '%s' as a %s."), $candidate, "po_directory" ) if $po4a_opts{"verbose"}; opendir PO_DIR, $candidate or die wrap_ref_mod( "$config_file:$nb", "", gettext("Cannot list directory '%s' in '%s' "), $po_directory, $basedir ); map { push @po_files, "$po_directory/$_" if -e "$candidate/$_"; } ( sort readdir PO_DIR ); } } unless ($found) { die wrap_ref_mod( "$config_file:$nb", "", gettext("Directory '%s' not found in '%s'."), $po_directory, $calldir ) if ( not defined $destdir && not defined $srcdir ); die wrap_ref_mod( "$config_file:$nb", "", gettext("Directory '%s' not found in '%s' nor in '%s'."), $po_directory, $destdir, $calldir ) if ( defined $destdir && not defined $srcdir ); die wrap_ref_mod( "$config_file:$nb", "", gettext("Directory '%s' not found in '%s' nor in '%s'."), $po_directory, $calldir, $srcdir ) if ( not defined $destdir && defined $srcdir ); die wrap_ref_mod( "$config_file:$nb", "", gettext("Directory '%s' not found in '%s' nor in '%s' nor in '%s'."), $po_directory, $destdir, $calldir, $srcdir ); } } # Step 2: separate the PO and POT files we found in all traversed directories my %seen = (); # the elements of po_files that we saw already (we need to ignore the dupplicates) foreach my $f (@po_files) { next if $seen{$f}++; # we saw it already if ( $f =~ m|^(.*?)/([^/]*)\.po$| ) { push @langs, $2; die wrap_ref_mod( "$config_file:$nb", gettext("Language '%s' found twice in '%s' and '%s'."), $po_filename{$2}, $f ) if defined $po_filename{$2} && $po_filename{$2} ne $f; $po_filename{$2} = $f; print wrap_msg( gettext("Found language '%s' in the provided po_directory: %s"), $2, $f ) if $po4a_opts{"verbose"}; } if ( $f =~ m/\.pot$/ ) { die wrap_msg( gettext("%s:%d: too many POT files: '%s' and '%s'."), $config_file, $nb, $pot_filename, $f ) if ( length $pot_filename ); $pot_filename = $f; print wrap_msg( gettext("Found POT file '%s' in the provided po_directory."), $pot_filename ) if $po4a_opts{"verbose"}; } } undef %seen; # Step 3: if no PO file found, make a sensible error message. # We could factorize the error message, but it would break gettext() if ( not @langs ) { if ( not defined $destdir && not defined $srcdir ) { warn wrap_ref_mod( "$config_file:$nb", "", gettext("No PO files found in '%s'/'%s'."), $calldir, $main ); } elsif ( defined $destdir && not defined $srcdir ) { warn wrap_ref_mod( "$config_file:$nb", "", gettext("No PO files found in '%s'/'%s' nor in '%s'/'%s'."), $destdir, $main, $calldir, $main ); } elsif ( not defined $destdir && defined $srcdir ) { warn wrap_ref_mod( "$config_file:$nb", "", gettext("No PO files found in '%s'/'%s' nor in '%s'/'%s'."), $calldir, $main, $srcdir, $main ); } else { warn wrap_ref_mod( "$config_file:$nb", "", gettext("No PO files found in '%s'/'%s' nor in '%s'/'%s' nor in '%s'/'%s'."), $destdir, $main, $calldir, $main, $srcdir, $main ); } warn wrap_ref_mod( "$config_file:$nb", "", gettext("Your project will not be translated to any language.") ); } # Step 4: if no POT file found, make a sensible error message if ( length $pot_filename == 0 ) { if ( not defined $destdir && not defined $srcdir ) { warn wrap_ref_mod( "$config_file:$nb", "", gettext("No POT files found in '%s'/'%s'."), $calldir, $main ); } elsif ( defined $destdir && not defined $srcdir ) { warn wrap_ref_mod( "$config_file:$nb", "", gettext("No POT files found in '%s'/'%s' nor in '%s'/'%s'."), $destdir, $main, $calldir, $main ); } elsif ( not defined $destdir && defined $srcdir ) { warn wrap_ref_mod( "$config_file:$nb", "", gettext("No POT files found in '%s'/'%s' nor in '%s'/'%s'."), $calldir, $main, $srcdir, $main ); } else { warn wrap_ref_mod( "$config_file:$nb", "", gettext("No POT files found in '%s'/'%s' nor in '%s'/'%s' nor in '%s'/'%s'."), $destdir, $main, $calldir, $main, $srcdir, $main ); } warn wrap_ref_mod( "$config_file:$nb", "", gettext("Please create an empty file with the 'pot' extension (e.g. 'myproject.pot').") ); } } elsif ( $cmd =~ m/type: *(.*)/ ) { if ( defined $document{$main}{'format'} ) { warn wrap_ref_mod( "$config_file:$nb", "", gettext( "The master file '%s' was specified earlier in the configuration file. This may cause problems with options." ), $main ) unless ( $po4a_opts{"quiet"} ); } elsif ( not -e find_input_file($main) ) { die wrap_ref_mod( "$config_file:$nb", "", gettext("The master file '%s' does not exist."), find_input_file($main) ); } if ( scalar @{ $po4a_opts{"partial"} } ) { foreach my $file ( @{ $po4a_opts{"partial"} } ) { if ( $args =~ m/(\S+):\Q$file\E\b/ ) { $partial{'lang'}{$1} = 1; $partial{'master'}{$main} = 1; $partial{'files'}{$file} = 1; last; } } } $document{$main}{'format'} = $1; $document{$main}{'pot_in'} = $pot_in; $document{$main}{'pos'} = $doc_count; $doc_count++; # Options my %options; # 1. Use the global options ([opt] ...) # %options = %{ $document{''}{'options'} } # if defined $document{''}{'options'}; # 2. Use the alias options if any if ( defined $aliases{$1} ) { $document{$main}{'format'} = $aliases{$1}{"module"}; if ( defined $aliases{$1}{"options"} ) { # print STDERR "Override the global options with the alias ones\n"; %options = %{ $aliases{$1}{"options"} }; # XXX not a merge, but overwrite } } # 3. If this file was already specified, reuse the previous # options (no merge) %options = %{ $document{$main}{'options'} } if defined $document{$main}{'options'}; # 4. Save the name of the master doc, to handle "$master" substitution in strings warn( wrap_ref_mod( "$config_file:$nb", "", gettext("Option 'master:file' is deprecated. Please use 'pot' instead for the same effect.") ) ) if ( $args =~ m/ +master:file=\S/ ); if ( $args =~ s/ +master:file=(\S+)// || $args =~ s/ +pot=(\S+)// ) { $document{$main}{'pot'} = $1; } # 5. Merge the document specific options # separate the end of the line, which contains options. # Something more clever could be done to allow options in the # middle of a line. if ( $args =~ m/^(.*?) +(opt(_.+)?:(.*))$/ ) { $args = $1 // ""; $args .= " " . parse_config_options( "$config_file:$nb", $2, \%options ); } %{ $document{$main}{'options'} } = %options; my %discarded = (); my @remaining_args = split( / /, $args ); while (@remaining_args) { my $arg = shift(@remaining_args); die wrap_ref_mod( "$config_file:$nb", "", gettext("Unparsable argument '%s' (%s)."), $arg, $line ) unless ( $arg =~ /^([^:]*):(.*)/ ); my ( $lang, $trans ) = ( $1, $2 ); die wrap_ref_mod( "$config_file:$nb", "", gettext("The translated and master file are the same.") ) if ( $main eq $trans ); if ( $lang =~ /^add_/ ) { my $modifiers; $trans =~ s/^([@!?]+)// and $modifiers = $1; next if defined( $discarded{$trans} ); if ( defined $modifiers ) { if ( $modifiers =~ m/!/ ) { $discarded{$trans} = 1; next; } next if ( $modifiers =~ m/\?/ and not -e find_input_file($trans) ); if ( $modifiers =~ m/@/ ) { open LIST, "<", find_input_file($trans) or die wrap_msg( gettext("Cannot open %s: %s"), $trans, $! ); my @new_list = (); while (<LIST>) { chomp; s/\s+$//; next if length($_) == 0 or $_ =~ m/^\s*#/; while (m/\$\((\w+)\)/) { if ( defined $po4a_opts{"variables"}{$1} ) { s/\$\((\Q$1\E)\)/$po4a_opts{"variables"}{$1}/g; } else { die wrap_ref_mod( "$config_file:$nb", "", gettext("Unknown variable: %s"), $1 ); } } push( @new_list, "$lang:$_" ); } close LIST; unshift( @remaining_args, @new_list ); } else { push @{ $document{$main}{$lang} }, $trans; } } else { push @{ $document{$main}{$lang} }, $trans; } } elsif ( $lang eq "pot_in" ) { $document{$main}{$lang} = $trans; # override POT file input print " .......... divert pot_in=$trans\n" if $po4a_opts{"debug"}; } else { die wrap_ref_mod( "$config_file:$nb", "", gettext("Translation of %s in %s redefined"), $main, $lang ) if ( defined $document{$main}{$lang} ); $document{$main}{$lang} = $trans; } } } elsif ( $cmd =~ m/po4a_alias: *(.*)/ ) { my $name = $1; my %alias = (); $alias{"module"} = $main; my %options; $args = parse_config_options( "$config_file:$nb", $args, \%options ); %{ $alias{"options"} } = %options; %{ $aliases{$name} } = %alias; } elsif ( $cmd eq "options" ) { my %options; %options = %{ $document{''}{"options"} } if ( defined $document{''}{"options"} ); # Add to the existing options parse_config_options( "$config_file:$nb", "$main $args", \%options ); %{ $document{''}{"options"} } = %options; } else { die wrap_ref_mod( "$config_file:$nb", "", gettext("Unparsable command '%s'."), $cmd ); } $line = ""; } close CONFIG; # don't care about error here die wrap_msg( gettext("'po4a_paths' is not defined in the configuration file. Where are the POT and PO files?") ) unless ( length $pot_filename ); # Parse a vector of parameters, and split them on spaces. # If passed ('aa', 'bb', 'cc dd'), it splits the last parameter and returns ('aa', 'bb', 'cc', 'dd'), sub split_opts { my $options = shift; my $options_orig = $options; my @opts = (); $options =~ s/\\"/"/g; while ( length $options ) { my $o = ""; while ( length $options and $options !~ /^ /s ) { if ( ( $options =~ m/^(["'])/ ) and ( $options !~ m/^(["'])(?:\\.|(?!\1)[^\\])*\1/ ) ) { die wrap_msg( gettext("Cannot parse option line (missing >%s<?): %s"), $1, $options_orig ); } # Extract non quoted parts $options =~ s/^([^\\"' ]*)//s; $o .= $1 if defined $1; # And extract quoted parts $options =~ s/^(["'])((?:\\.|(?!\1)[^\\])*)\1//s; $o .= $2 if defined $2; } $options =~ s/^ *//s; push @opts, $o; } return @opts; } # Adds the parameters passed in the global section of the config file to the ones passed on the command line if ( defined $document{''}{"options"} and defined $document{''}{"options"}{"global"} ) { %po4a_opts = get_options( split_opts( $document{''}{"options"}{"global"} ), @ORIGINAL_ARGV ); } # make a big pot my $update_pot_file = 0; if ( $pot_filename =~ m/\$master/ ) { printf( gettext("Split mode, creating a temporary POT:") ) if $po4a_opts{"verbose"}; if ( scalar @{ $po4a_opts{"partial"} } ) { print wrap_msg( gettext("Disabling --translate-only option, it is not supported in split mode") . "\n" ); $po4a_opts{"partial"} = []; } # The POT needs to be generated anyway. $update_pot_file = 1; $po4a_opts{"split"} = 1; } else { if ( scalar @{ $po4a_opts{"partial"} } ) { # Skip documents not specified, strings are read directly from POT file foreach my $master ( keys %document ) { next unless length $master; delete $document{$master} unless exists $partial{'master'}{$master}; } # Do not read PO files if no file is processed for this language foreach my $lang ( keys %po_filename ) { delete $po_filename{$lang} unless exists $partial{'lang'}{$lang}; } } if ( not scalar @{ $po4a_opts{"partial"} } ) { if ( -e find_input_file($pot_filename) ) { my $modtime = ( stat find_input_file($pot_filename) )[9]; # The POT needs to be re-generated if a master document is more recent # than the POT. foreach my $master ( keys %document ) { next if ( $master eq '' ); my $pot_in = find_input_file( $document{$master}{'pot_in'} ); if ( ( stat $pot_in )[9] >= $modtime ) { $update_pot_file = 1; last; } } if ( ( stat $config_file )[9] > $modtime ) { # The configuration file was modified after the POT. # Maybe a new document, or new options $update_pot_file = 1; } if ( $po4a_opts{"force"} ) { $update_pot_file = 1; } if ( $po4a_opts{"no-update"} ) { print wrap_msg( gettext("NOT updating %s as requested (--no-update)."), $pot_filename ) if ( $update_pot_file and $po4a_opts{"verbose"} ); $update_pot_file = 0; } else { printf( gettext("Updating %s:"), $pot_filename ) if ( $update_pot_file and $po4a_opts{"verbose"} ); } } elsif ( $po4a_opts{"no-update"} ) { print wrap_msg( gettext("NOT creating %s as requested (--no-update)."), $pot_filename ) if $po4a_opts{"verbose"}; } else { printf( gettext("Creating %s:"), $pot_filename ) if $po4a_opts{"verbose"}; $update_pot_file = 1; } } } if ($update_pot_file) { my %pot_options; foreach (qw(porefs wrap-po msgid-bugs-address copyright-holder package-name package-version)) { if ( defined $po4a_opts{$_} ) { $pot_options{$_} = $po4a_opts{$_}; } } $pot_options{'pot-language'} = $po4a_opts{"mastlang"}; $pot_options{'pot-charset'} = $po4a_opts{"mastchar"}; my $potfile = Locale::Po4a::Po->new( \%pot_options ); foreach my $master ( sort { return -1 if ( $a eq "" ); return 1 if ( $b eq "" ); $document{$a}{'pos'} <=> $document{$b}{'pos'} } keys %document ) { next if ( $master eq '' ); my $pot_in = $document{$master}{'pot_in'}; my %file_opts = %po4a_opts; my $options = ( $document{''}{"options"}{"global"} // '' ) . ' ' . # global options (if any) as default values ( $document{$master}{"options"}{"global"} // '' ) . ' '; # file's global options (if any) to override them $options =~ s/ *$//; print STDERR "Options for $master while generating pot: $options\n" if $po4a_opts{'debug'}; %file_opts = get_options( split_opts($options), @ORIGINAL_ARGV ) if ( length $options ); my $doc = Locale::Po4a::Chooser::new( $document{$master}{'format'}, %{ $file_opts{"options"} } ); # If we know that the input isn't UTF-8, say so to the TransTractor if ( length $po4a_opts{"mastchar"} && ( not $po4a_opts{"mastchar"} =~ /UTF-8/i ) && $po4a_opts{"mastchar"} ne "CHARSET" ) { print STDERR "mastchar: " . $po4a_opts{"mastchar"} . "\n"; $doc->{TT}{ascii_input} = 0; # this actually mean "don't mess up with the encoding", not litteraly that it's in ascii } # We ensure that the generated po will be in utf-8 if the input document isn't entirely in ascii $doc->{TT}{utf_mode} = 1; $doc->setpoout($potfile); my @file_in_name; push @file_in_name, $pot_in; $doc->process( 'file_in_name' => \@file_in_name, 'file_in_charset' => $file_opts{"mastchar"}, 'srcdir' => $po4a_opts{"srcdir"}, 'destdir' => $po4a_opts{"destdir"}, 'calldir' => $po4a_opts{"calldir"} ); $potfile = $doc->getpoout(); } if ( $po4a_opts{"split"} ) { ( undef, $pot_filename ) = File::Temp::tempfile( "po4a-temp-pot-XXXX", DIR => $ENV{TMPDIR} || "/tmp", SUFFIX => ".pot", OPEN => 0, UNLINK => 0 ) or die wrap_msg( gettext("Cannot create a temporary POT file: %s"), $! ); $potfile->write($pot_filename); } else { if ( $po4a_opts{"force"} ) { $potfile->write( find_output_file($pot_filename) ); } else { $potfile->write_if_needed( find_output_file($pot_filename) ); } } print wrap_msg( gettext(" (%d entries)"), $potfile->count_entries() ) unless ( $po4a_opts{"quiet"} ); } else { print wrap_msg( gettext("POT file %s already up to date."), $pot_filename ) if ( $po4a_opts{"verbose"} and not $po4a_opts{"no-update"} ); } my %split_po; # po_files: '$lang','$master' => '$path' my %split_pot; # pot_files: '$master' => '$path' my %split_pot_files; # List of files in a given POT, as a parameter to msggrep: '$potfile' => "-N file1 -N file2" if ( $po4a_opts{"split"} ) { foreach my $master ( keys %document ) { next if ( $master eq '' ); my $m = $document{$master}{'pot'} || basename $master; my $master_pot = $pot_filename_cfg; $master_pot =~ s/\$master/$m/g; $split_pot{$master} = $master_pot; my $searched_file = $document{$master}{'pot_in'} // $master; $split_pot_files{$master_pot} //= ''; $split_pot_files{$master_pot} .= " -N \"$searched_file\""; } # Generate a .pot for each document by msggrep on the big POT file. # These files are used to update the PO files later on (and also by humans to start new translations) if ( not $po4a_opts{"no-update"} ) { foreach my $master_pot ( keys %split_pot_files ) { my $tmp_file; # Create a temporary POT, and check if the old one needs to be # updated (unless --force was specified). unless ( $po4a_opts{"force"} ) { ( undef, $tmp_file ) = File::Temp::tempfile( "po4aXXXX", DIR => $ENV{TMPDIR} || "/tmp", SUFFIX => ".pot", OPEN => 0, UNLINK => 0 ) or die wrap_msg( gettext("Cannot create a temporary POT file: %s"), $! ); } my $dir = dirname( find_output_file($master_pot) ); if ( not -d $dir ) { mkdir $dir or die wrap_msg( gettext("Cannot create directory '%s': %s"), $dir, $! ); } my $outfile = $po4a_opts{"force"} ? find_output_file($master_pot) : $tmp_file; my $cmd = "msggrep$Config{_exe} $split_pot_files{$master_pot} -o $outfile $pot_filename"; run_cmd($cmd); die wrap_msg( gettext( "msggrep did not create '%s'. Used command:\n %s\nPlease report that bug, along the information allowing to reproduce it.\n" ), $outfile, $cmd ) unless ( -e $outfile ); move_po_if_needed( $tmp_file, find_output_file($master_pot), 0 ) unless ( $po4a_opts{"force"} ); } } # Generate a complete .po for each language, with the translations of all master files # That's done with msgcat on all existing PO files of a given language foreach my $lang ( sort keys %po_filename ) { my $tmp_bigpo; ( undef, $tmp_bigpo ) = File::Temp::tempfile( "po4aXXXX", DIR => $ENV{TMPDIR} || "/tmp", SUFFIX => "-$lang.po", OPEN => 0, UNLINK => 0 ) or die wrap_msg( gettext("Cannot create a temporary PO file: %s"), $! ); my $cmd_cat = ""; foreach my $master ( sort keys %document ) { next if ( $master eq '' ); my $m = $document{$master}{'pot'} || basename $master; my $master_po = $po_filename{$lang}; $master_po =~ s/\$master/$m/g; if ( -e find_input_file($master_po) ) { $cmd_cat .= " " . find_input_file($master_po); } else { print wrap_msg( gettext("The translation of master file '%s' in language '%s' is missing (file: %s) -- skipping."), $master, $lang, $master_po ) if ( $po4a_opts{"verbose"} ); } $split_po{$lang}{$master} = $master_po; } if ( length $cmd_cat ) { $cmd_cat = "msgcat" . $Config{_exe} . " -o $tmp_bigpo $cmd_cat"; run_cmd($cmd_cat); } # We do not need to keep the original name with $master $po_filename{$lang} = $tmp_bigpo; } } if ( not $po4a_opts{"no-update"} ) { # update all po files my $lang; if ( not scalar @{ $po4a_opts{"partial"} } ) { foreach $lang ( sort keys %po_filename ) { my ( $infile, $outfile ) = ( find_input_file( $po_filename{$lang} ), find_output_file( $po_filename{$lang} ) ); my $updated_potfile = find_input_file($pot_filename); if ( -e $infile ) { my $dir = dirname($outfile); if ( not -d $dir ) { mkdir $dir or die wrap_msg( gettext("Cannot create directory '%s': %s"), $dir, $! ); } my $msgmerge_opt = $po4a_opts{"msgmerge-opt"}; $msgmerge_opt =~ s/\$lang\b/$lang/g if scalar @langs; my $cmd = "msgmerge" . $Config{_exe} . " $infile $updated_potfile " . $msgmerge_opt; if ( $infile eq $outfile ) { # in place $cmd .= " --backup=none --update"; } else { $cmd .= " --output $outfile"; } run_cmd($cmd); if ( $po4a_opts{"verbose"} ) { if ( $po4a_opts{'split'} ) { printf( gettext("Updating the translation in language %s:"), $lang ); } else { printf( gettext("Updating %s:"), $po_filename{$lang} ); } my $stat = qx(msgfmt$Config{_exe} --statistics -v -o $devnull $outfile 2>&1); $stat =~ s/^[^:]*://; print $stat; } } else { my $read_pot_filename = find_input_file($pot_filename); my $cmd = "msginit$Config{_exe} -i $read_pot_filename --locale $lang -o $outfile --no-translator >$devnull"; run_cmd($cmd); } } if ( $po4a_opts{"split"} ) { # Split the complete PO in multiple POs foreach $lang ( sort keys %po_filename ) { foreach my $master ( keys %document ) { next if ( $master eq '' ); my $tmp_file; # Create a temporary PO, and check if the old one needs to be # updated (unless --force was specified). ( undef, $tmp_file ) = File::Temp::tempfile( "po4aXXXX", DIR => $ENV{TMPDIR} || "/tmp", SUFFIX => ".po", OPEN => 0, UNLINK => 0 ) or die wrap_msg( gettext("Cannot create a temporary POT file: %s"), $! ); my $cmd; # Create an empty PO or copy the original PO header (to keep the header) if ( -f find_input_file( $split_po{$lang}{$master} ) ) { # Pass msggrep to C (english), but not under windows as this syntax is not understood by the shell my $env = ( $^O =~ /Win/ ) ? "" : "LC_ALL=C.UTF-8"; $cmd = $env . " msggrep$Config{_exe}" . " --force-po --invert-match --msgid --regexp '.'" . " --output $tmp_file " . find_input_file( $split_po{$lang}{$master} ); } else { $cmd = "msginit$Config{_exe} " . "--no-translator -l $lang --input " . find_output_file( $split_pot{$master} ) . " --output $tmp_file >$devnull"; } run_cmd($cmd); # Update the PO according to the new POT and to the big PO (compendium). $cmd = "msgmerge$Config{_exe} " . " --compendium " . $po_filename{$lang} . " --update --backup=none " . $po4a_opts{"msgmerge-opt"} . " $tmp_file " . find_output_file( $split_pot{$master} ); run_cmd($cmd); my $dir = dirname( $split_po{$lang}{$master} ); if ( not -d $dir ) { mkdir $dir or die wrap_msg( gettext("Cannot create directory '%s': %s"), $dir, $! ); } unless ( $po4a_opts{"force"} ) { move_po_if_needed( $tmp_file, $split_po{$lang}{$master}, 0 ); } else { move $tmp_file, find_output_file( $split_po{$lang}{$master} ) or die wrap_msg( dgettext( "po4a", "Cannot move %s to %s: %s." ), $tmp_file, find_output_file( $split_po{$lang}{$master} ), $! ); } } } } # if ( $po4a_opts{"split"} ) } # if ( not scalar @{ $po4a_opts{"partial"} } ) } # not $po4a_opts{"no-update"} if ( $po4a_opts{"split"} ) { # We don't need the tmp big POT anymore unlink($pot_filename); print STDERR "Unlink the big POT file $pot_filename\n" if $po4a_opts{'debug'}; } if ( not $po4a_opts{"no-translations"} ) { # update all translations LANG: foreach my $lang ( sort keys %po_filename ) { # Read the $lang PO once, no options for the creation my $po = Locale::Po4a::Po->new(); my $pofile = find_input_file( $po_filename{$lang} ); if ( not -e $pofile ) { if ( $po4a_opts{"verbose"} ) { if ( File::Spec->file_name_is_absolute( $po_filename{$lang} ) ) { print wrap_msg( gettext("PO file for language %s is missing -- skipping."), $lang ); } else { print wrap_msg( gettext("PO file %s for language %s is missing -- skipping."), $po_filename{$lang}, $lang ); } } next LANG; } print "Reading PO file $pofile for language $lang: " if ( $po4a_opts{"debug"} ); $po->read($pofile); system( "msgfmt" . $Config{_exe} . " --statistics -v -o $devnull $pofile" ) if ( $po4a_opts{"debug"} ); DOC: foreach my $master ( sort { return -1 if ( $a eq "" ); return 1 if ( $b eq "" ); $document{$a}{'pos'} <=> $document{$b}{'pos'} } keys %document ) { next if ( $master eq '' ); next unless defined $document{$master}{$lang}; if ( scalar @{ $po4a_opts{"partial"} } ) { next unless defined $partial{'files'}{ $document{$master}{$lang} }; } unless ( $po4a_opts{"force"} ) { my $outfile = find_input_file( $document{$master}{$lang} ); my $stampfile = ( -e $outfile ? $outfile : "$outfile.po4a-stamp" ); $stampfile = File::Spec->rel2abs($stampfile); my @files = ( $master, $po_filename{$lang}, File::Spec->rel2abs($config_file) ); if ( defined $document{$master}{"add_$lang"} ) { push @files, @{ $document{$master}{"add_$lang"} }; } push @files, @{ $add_filename{$lang} } if ( defined $add_filename{$lang} ); unless ( is_older( $stampfile, @files ) ) { print wrap_msg( gettext("%s doesn't need to be updated."), $document{$master}{$lang} ) if ( $po4a_opts{"verbose"} ); next DOC; } } my %file_opts = %po4a_opts; my $options = ( $document{''}{"options"}{"global"} // '' ) . ' ' . # global options (if any) as default values ( $document{$master}{"options"}{"global"} // '' ) . ' ' . # file's global options (if any) to override them ( $document{$master}{"options"}{$lang} // '' ); # file's language options (if any) overriding everything $options =~ s/ *$//; print STDERR "Options for $master in $lang: $options\n" if $po4a_opts{'debug'}; %file_opts = get_options( split_opts($options), @ORIGINAL_ARGV ) if ( length $options ); my $doc = Locale::Po4a::Chooser::new( $document{$master}{'format'}, %{ $file_opts{"options"} } ); my @file_in_name; push @file_in_name, $master; # Reuse the already parsed PO, do not use the po_in_name option of process. $doc->{TT}{po_in} = $po; $doc->{TT}{po_in}->stats_clear(); my $translation_out; if ( $po4a_opts{"keep-translations"} && -e find_output_file( $document{$master}{$lang} ) ) { ( undef, $translation_out ) = File::Temp::tempfile( "$document{$master}{$lang}.new-XXXX", OPEN => 0, UNLINK => 0 ); } else { $translation_out = $document{$master}{$lang}; } $doc->process( 'file_in_name' => \@file_in_name, 'file_out_name' => $translation_out, 'file_in_charset' => $file_opts{"mastchar"}, 'file_out_charset' => $file_opts{"locchar"}, 'addendum_charset' => $file_opts{"addchar"}, 'srcdir' => $po4a_opts{"srcdir"}, 'destdir' => $po4a_opts{"destdir"}, 'calldir' => $po4a_opts{"calldir"} ); my ( $percent, $hit, $queries ) = $doc->stats(); if ( $percent < $file_opts{"threshold"} ) { print wrap_msg( gettext("Discard %s (%s of %s strings; only %s%% translated; need %s%%)."), $document{$master}{$lang}, $hit, $queries, $percent, $file_opts{"threshold"} ); unlink( find_output_file( $document{$master}{$lang} ) ) if ( !$po4a_opts{"keep-translations"} ); unlink( find_output_file($translation_out) ) if ( $po4a_opts{"keep-translations"} ); if ( $po4a_opts{"stamp"} && ( not $po4a_opts{"force"} ) ) { touch( find_output_file( $document{$master}{$lang} ) . ".po4a-stamp" ); print wrap_msg( gettext("Timestamp %s created."), $document{$master}{$lang} . ".po4a-stamp" ) if ( $po4a_opts{"verbose"} ); } next DOC; } else { # The translated document reached the threshold and was kept rename( $translation_out, find_output_file( $document{$master}{$lang} ) ); unless ( $po4a_opts{"force"} ) { if ( -e $document{$master}{$lang} . ".po4a-stamp" ) { unlink find_output_file( $document{$master}{$lang} ) . ".po4a-stamp"; print wrap_msg( gettext("Timestamp %s removed."), $document{$master}{$lang} . ".po4a-stamp" ) if ( $po4a_opts{"verbose"} ); } } } my @addlist; push @addlist, @{ $document{$master}{"add_$lang"} } if ( defined( $document{$master}{"add_$lang"} ) ); push @addlist, @{ $add_filename{$lang} } if ( defined( $add_filename{$lang} ) ); foreach my $add (@addlist) { if ( !$doc->addendum( find_input_file($add) ) ) { die wrap_msg( gettext("Addendum %s does NOT apply to %s (translation discarded)."), $add, $document{$master}{$lang} ); my $outfile = find_output_file( $document{$master}{$lang} ); unlink($outfile) if ( -e $outfile ); next DOC; } } if ( $file_opts{"verbose"} ) { if ( $percent == 100 ) { print wrap_msg( gettext("%s is 100%% translated (%s strings)."), $document{$master}{$lang}, $queries ); } else { print wrap_msg( gettext("%s is %s%% translated (%s of %s strings)."), $document{$master}{$lang}, $percent, $hit, $queries ); } } $doc->write( find_output_file( $document{$master}{$lang} ) ); } } } if ( $po4a_opts{"split"} ) { # We don't need the tmp big POs anymore foreach my $lang ( keys %po_filename ) { unlink $po_filename{$lang}; } } if ( $po4a_opts{"rm-translations"} ) { # Delete the translated documents foreach my $lang ( keys %po_filename ) { foreach my $master ( keys %document ) { next if ( $master eq '' ); my $outfile = find_output_file( $document{$master}{$lang} ); unlink $outfile; unlink "$outfile.po4a-stamp"; } } } sub touch { my $file = shift; if ( -e $file ) { utime undef, undef, $file; } else { sysopen( FH, $file, O_WRONLY | O_CREAT | O_NONBLOCK | O_NOCTTY ) or croak("Cannot create $file : $!"); close FH or croak("Cannot close $file : $!"); } } sub is_older { my $file = shift; my @files = @_; return 1 unless ( -e $file ); my $older = 0; my $modtime = ( stat $file )[9]; for my $f (@files) { if ( -e $f and ( stat $f )[9] > $modtime ) { $older = 1; last; } } return $older; } __END__