NAME Locale::TextDomain::OO::Extract - Extracts internationalization data
$Id: Extract.pm 709 2017-11-03 10:40:26Z steffenw $
$HeadURL: svn+ssh://steffenw@svn.code.sf.net/p/perl-gettext-oo/code/extract/trunk/lib/Locale/TextDomain/OO/Extract.pm $
VERSION
2.014
DESCRIPTION
This module extracts internationalization data.
The extractor runs the following steps:
- preprocess the file
- find all possible positions in content
- match exactly and create a stack
- map the stack to a lexicon entry
If you want to write the lexicon structure to pot files see folder example of this distribution how it works.
SYNOPSIS
There are different extract module for different files. All this extracted data are stored into one or selective lexicons. At the end of extraction this lexicons can be stored into pot files or anywhere else.
Use an existing extractor
use strict;
use warnings;
use Locale::TextDomain::OO::Extract::...;
use Path::Tiny qw(path);
my $extractor = Locale::TextDomain::OO::Extract::...->new(
# all parameters are optional
lexicon_ref => \my %lexicon, # default is {}
domain => 'MyDomain', # default is q{}
category => 'LC_MESSAGES', # default is q{}
debug_code => sub { # default is undef
my ($group, $message) = @_;
print $group, ', ', $message, "\n";
return;
},
);
my @files = (
'relative_dir/filename1.suffix',
'relative_dir/filename2.suffix',
);
for ( @files ) {
$extractor->clear;
$extractor->filename($_);
$extractor->content_ref( \( path($_)->slurp_utf8 ) );
$extractor->extract;
}
# do something with that
# maybe write a pot file using Locale::PO
... = extractor->lexicon_ref;Write your own extractor
package MyExtractor;
use Moo;
extends qw(
Locale::TextDomain::OO::Extract::Base::RegexBasedExtractor
);
with qw(
Locale::TextDomain::OO::Extract::Role::File
);Optional method to uncomment or interpolate the file content or anything else.
sub preprocess {
my $self = shift;
my $content_ref = $self->content_ref;
# modify anyhow
${$content_ref}=~ s{\\n}{\n}xmsg;
return;
}Map the reference, all the matches or defaults. See Perl extractor how it works. Maybe ignore some stack entries.
sub stack_item_mapping {
my $self = shift;
my $match = $_->{match};
$self->add_message({
reference => ...,
domain => ...,
category => ...,
msgctxt => ...,
msgid => ...,
msgid_plural => ...,
});
return;
}Match all positions, the searched string is starting with. You can match to the end of the searched string but there is no need for.
my $start_rule = qr{ ... }xms;Match exactly the different things. All the values from () are stored in stack. Prepare the stack in a way you are able to merge the data. Maybe an empty match helps to have all on the right position.
'or' means: if not then try the following. 'or' is only allowed between 2 array references.
my $rules = [
[
'begin',
qr{ ... ( ... ) ...}xms, # match this
'and',
qr{ ... ( ... ) ...}xms, # then that
'end',
]
'or',
[
'begin',
[
qr{ ... ( ... ) ... }xms,
],
'or',
[
qr{ ... ( ... ) ... }xms,
'or',
qr{ ... ( ... ) ... }xms,
],
'end',
],
];Tell your extractor what steps he should run.
sub extract {
my $self = shift;
$self->start_rule($start_rule);
$self->rules($rules);
$self->preprocess;
$self->SUPER::extract;
for ( @{ $self->stack } ) {
$self->stack_item_mapping;
}
return;
}The whole process (extract, translate, clean, format)
See also module Locale::TextDomain::OO::Extract::Process
1.1. If the PO file not exists for a language create it with the right header
or change the defaults after initial write (spew).
|
| .------------------------------------.
| | |
v v |
.-------------. |
| de.po |-. |
'-------------' | |
'-------------' |
| | |
.------' 2.1. Read the existing PO files |
| of all languages |
| into the lexicon sturcture. |
| | |
| v |
| .-------------------. |
| | lexicon structure |<--------------------. |
| | de:: fr:: ... |<----------------. | |
| '-------------------' | | |
| | | | ^ ^ ^
| | | '-------------------->|->-|->-|->-.
| | | ^ ^ ^ |
| | 3.1. Remove all gettext | | | |
| | references | | | |
| | | | | | |
| | '-------------------------' | | |
| | | | |
| | .------------------. | | |
| | | MyProjectFile.pm |-. | | |
| | '------------------' | | | |
| | '------------------' | | |
| | | | | |
| | 4. Extract the files | | |
| | of your project. | | |
| | 4.1. Add new message id's | | |
| | if not already exists. | | |
| | 4.2. Add the new references. | | |
| | | | | |
| | '----------------------' | |
| | | |
| 5.1. Merge new and changed messages. | |
| 5.2. Write back the PO files. | |
| | ^ |
| '---------------------------------------->O |
v ^ |
O->-----. | |
v | | |
| 6.1. Translate the PO files. | |
| | ^ |
| '---------------------------------------->O |
v ^ |
O->-----. | |
v | | |
| 7.1. Clean all messages | |
| without any reference. | |
| | | |
| '-----------------------------------------' |
| |
'-------. |
| |
8.1. Write MO files. |
| |
v |
.-------------. |
| de.mo |-. |
'-------------' | |
'-------------' |
|
Update the references in region files. |
|
1.2. If the PO file not exists for a region |
create it with the right header. |
| |
v |
.----------. |
| de-at.po |-. v
'----------' |<-----------------------------------O
'----------' ^
| | |
.--------' | |
| | |
| 2.1. Read the existing PO files |
| of all regions |
| into the lexicon sturcture. |
| 3.2. Remove all references. |
| 4.3. Add the new references (from language) |
| for existing region messages only. |
| 5.1. Merge changed messages. |
| 5.2. Write back to PO files. |
| 6.2. Translate the PO files. |
| 7.2. Clean all messages |
| without any reference. |
| | |
| '---------------------------------------'
'--------.
|
8.2. Write MO files.
|
v
.----------.
| de-at.mo |-
'----------' |
'----------'SUBROUTINES/METHODS
none
EXAMPLE
Inside of this distribution is a directory named example. Run this *.pl files.
DIAGNOSTICS
none
CONFIGURATION AND ENVIRONMENT
none
DEPENDENCIES
none
INCOMPATIBILITIES
not known
BUGS AND LIMITATIONS
none
SEE ALSO
AUTHOR
Steffen Winkler
LICENSE AND COPYRIGHT
Copyright (c) 2014 - 2017, Steffen Winkler <steffenw at cpan.org>. All rights reserved.
This module is free software; you can redistribute it and/or modify it under the same terms as Perl itself.