NAME¶
Locale::Maketext::Extract - Extract translatable strings from source
SYNOPSIS¶
my $Ext = Locale::Maketext::Extract->new;
$Ext->read_po('messages.po');
$Ext->extract_file($_) for <*.pl>;
# Set $entries_are_in_gettext_format if the .pl files above use
# loc('%1') instead of loc('[_1]')
$Ext->compile($entries_are_in_gettext_format);
$Ext->write_po('messages.po');
-----------------------------------
### Specifying parser plugins ###
my $Ext = Locale::Maketext::Extract->new(
# Specify which parser plugins to use
plugins => {
# Use Perl parser, process files with extension .pl .pm .cgi
perl => [],
# Use YAML parser, process all files
yaml => ['*'],
# Use TT2 parser, process files with extension .tt2 .tt .html
# or which match the regex
tt2 => [
'tt2',
'tt',
'html',
qr/\.tt2?\./
],
# Use My::Module as a parser for all files
'My::Module' => ['*'],
},
# Warn if a parser can't process a file or problems loading a plugin
warnings => 1,
# List processed files
verbose => 1,
);
DESCRIPTION¶
This module can extract translatable strings from files, and write them back to
PO files. It can also parse existing PO files and merge their contents with
newly extracted strings.
A command-line utility, xgettext.pl, is installed with this module as well.
The format parsers are loaded as plugins, so it is possible to define your own
parsers.
Following formats of input files are supported:
- Perl source files (plugin: perl)
- Valid localization function names are:
"translate", "maketext", "gettext",
"loc", "x", "_" and "__".
For a slightly more accurate, but much slower Perl parser, you can use the
PPI plugin. This does not have a short name (like "perl"), but
must be specified in full.
- HTML::Mason (Mason 1) and Mason (Mason 2) (plugin:
mason)
- HTML::Mason (aka Mason 1)
Strings inside <&|/l>...</&> and
<&|/loc>...</&> are extracted.
Mason (aka Mason 2) Strings inside <% $.floc { %>...</%> or
<% $.fl { %>...</%> or <% $self->floc {
%>...</%> or <% $self->fl { %>...</%> are
extracted.
- Template Toolkit (plugin: tt2)
- Valid forms are:
[% | l(arg1,argn) %]string[% END %]
[% 'string' | l(arg1,argn) %]
[% l('string',arg1,argn) %]
FILTER and | are interchangeable
l and loc are interchangeable
args are optional
- Text::Template (plugin: text)
- Sentences between "STARTxxx" and
"ENDxxx" are extracted individually.
- YAML (plugin: yaml)
- Valid forms are _"string" or _'string', eg:
title: _"My title"
desc: _'My "quoted" string'
Quotes do not have to be escaped, so you could also do:
desc: _"My "quoted" string"
- HTML::FormFu (plugin: formfu)
- HTML::FormFu uses a config-file to generate forms, with
built in support for localizing errors, labels etc.
We extract the text after "_loc: ":
content_loc: this is the string
message_loc: ['Max string length: [_1]', 10]
- Generic Template (plugin: generic)
- Strings inside {{...}} are extracted.
METHODS¶
Constructor¶
new()
new(
plugins => {...},
warnings => 1 | 0,
verbose => 0 | 1 | 2 | 3,
)
See "Plugins", "Warnings" and "Verbose" for
details
Plugins¶
$ext->plugins({...});
Locale::Maketext::Extract uses plugins (see below for the list) to parse
different formats.
Each plugin can also specify which file types it can parse.
# use only the YAML plugin
# only parse files with the default extension list defined in the plugin
# ie .yaml .yml .conf
$ext->plugins({
yaml => [],
})
# use only the Perl plugin
# parse all file types
$ext->plugins({
perl => '*'
})
$ext->plugins({
tt2 => [
'tt', # matches base filename against /\.tt$/
qr/\.tt2?\./, # matches base filename against regex
\&my_filter, # codref called
]
})
sub my_filter {
my ($base_filename,$path_to_file) = @_;
return 1 | 0;
}
# Specify your own parser
# only parse files with the default extension list defined in the plugin
$ext->plugins({
'My::Extract::Parser' => []
})
By default, if no plugins are specified, then it uses all of the builtin
plugins, and overrides the file types specified in each plugin
- instead, each plugin is tried for every file.
Available plugins
- "perl" :
Locale::Maketext::Extract::Plugin::Perl
- For a slightly more accurate but much slower Perl parser,
you can use the PPI plugin. This does not have a short name, but must be
specified in full, ie: Locale::Maketext::Extract::Plugin::PPI
- "tt2" :
Locale::Maketext::Extract::Plugin::TT2
- "yaml" :
Locale::Maketext::Extract::Plugin::YAML
- "formfu" :
Locale::Maketext::Extract::Plugin::FormFu
- "mason" :
Locale::Maketext::Extract::Plugin::Mason
- "text" :
Locale::Maketext::Extract::Plugin::TextTemplate
- "generic" :
Locale::Maketext::Extract::Plugin::Generic
Also, see Locale::Maketext::Extract::Plugin::Base for details of how to write
your own plugin.
Warnings¶
Because the YAML and TT2 plugins use proper parsers, rather than just regexes,
if a source file is not valid and it is unable to parse the file, then the
parser will throw an error and abort parsing.
The next enabled plugin will be tried.
By default, you will not see these errors. If you would like to see them, then
enable warnings via
new(). All parse errors will be printed to STDERR.
Also, if developing your own plugin, turn on warnings to see any errors that
result from loading your plugin.
Verbose¶
If you would like to see which files have been processed, which plugins were
used, and which strings were extracted, then enable "verbose". If no
acceptable plugin was found, or no strings were extracted, then the file is
not listed:
$ext = Locale::Extract->new( verbose => 1 | 2 | 3);
OR
xgettext.pl ... -v # files reported
xgettext.pl ... -v -v # files and plugins reported
xgettext.pl ... -v -v -v # files, plugins and strings reported
Accessors¶
header, set_header
lexicon, set_lexicon, msgstr, set_msgstr
entries, set_entries, entry, add_entry, del_entry
compiled_entries, set_compiled_entries, compiled_entry,
add_compiled_entry, del_compiled_entry
clear
PO File manipulation¶
method read_po ($file)
method write_po ($file, $add_format_marker?)
extract
extract_file
Compilation¶
compile($entries_are_in_gettext_style?)
Merges the "entries" into "compiled_entries".
If $entries_are_in_gettext_style is true, the previously extracted entries are
assumed to be in the
Gettext style (e.g. %1).
Otherwise they are assumed to be in
Maketext style (e.g.
"[_1]") and are converted into
Gettext style before merging
into "compiled_entries".
The "entries" are
not cleared after each compilation; use
"-"
set_entries()> to clear them if you need to extract
from sources with varying styles.
normalize_space
Lexicon accessors¶
msgids, has_msgid,
msgstr, set_msgstr
msg_positions, msg_variables, msg_format, msg_out
Internal utilities¶
_default_header
_maketext_to_gettext
_escape
_format
ACKNOWLEDGMENTS¶
Thanks to Jesse Vincent for contributing to an early version of this module.
Also to Alain Barbet, who effectively re-wrote the source parser with a
flex-like algorithm.
SEE ALSO¶
xgettext.pl, Locale::Maketext, Locale::Maketext::Lexicon
AUTHORS¶
Audrey Tang <cpan@audreyt.org>
COPYRIGHT¶
Copyright 2003-2008 by Audrey Tang <cpan@audreyt.org>.
This software is released under the MIT license cited below.
The "MIT" License¶
Permission is hereby granted, free of charge, to any person obtaining a copy of
this software and associated documentation files (the "Software"),
to deal in the Software without restriction, including without limitation the
rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
sell copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO
EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES
OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
DEALINGS IN THE SOFTWARE.