[wp-hackers] [GSoC] Language Packs

Stas Sușcov stas at nerd.ro
Tue Mar 29 11:13:30 UTC 2011

În data de Ma, 29-03-2011 la 12:16 +0200, Marko Novaković a scris:
> I consider to choose language packs module as GSoC project.
> This could be the plugin which download mo file for plugins or themes
> from GlotPress if exists for current language.
> This plugin may also support for multi-language blog.
> There are some plugins like WPMU, see here
> http://www.webdesignbooth.com/wordpress-multi-languages-5-plugins-to-built-a-multilingual-website/
> I cannot understand this sentence in ideas page:
> "Work is also necessary in GlotPress for this (support for plugins and
> themes, ability to build download packages, etc.), so this task can probably
> serve as two separate projects."
> I cannot understand what could be done in the GlotPress side.
> It also could be done automatic support for plugin and theme localization in
> plugin API,
> because it could be only localized plugins with the following segment of
> code:
> $locale_name  = get_locale();
> $mofile_name  = dirname(__FILE__) . '/languages';
> $mofile_name .= "/$domain_name-$locale_name.mo";
> load_textdomain("domain", $mofile_name);
> load_plugin_textdomain('domain', false, dirname( plugin_basename( __FILE__ )
> ) . '/languages/');

Nice idea,
actually I wrote a plugin for the 1st proposed part. I was lucky Nikolay
was around and enabled .mo downloads on GlotPress :)


The problem with current language support is that it can't be applied to
every plugin that exists in plugins directory. And that is for a couple
of reasons, like:
* plugins cannot benefit from GlotPress awesomeness so far (at least not
on translate.wordpress.org)
* plugin translations are mainly "accidental" contributions, and there
are no dedicated teams like <langCode>.wordpress.org that can solve this
problem, nor much people that would do this to cover all existent

To help you develop an idea about all this stuff, here's what I would
* talk to Nikolay about current status of GlotPress and what it needs to
get integrated with plugin repos (readme.txt will have to suffer
modifications and plugins will have to deliver .pot files mandatory in
this case)
* to reduce the "noise" of obsolete/old plugins locale imports, an
approval system should be written and automated that will validate the
plugin code for gettext calls and .pot file existence. From here you are
free to propose all kind of awesomeness, like integration with
profiles.wordpress.org to "number of X people required before plugin
will get accepted for a translation".
* a new pot-bot will have to be written :D

Good luck with this idea, though from my last year experience, you will
have to demonstrate more than just interest to get a project related
to .org infrastructure accepted.

Just my 2 cents...

More information about the wp-hackers mailing list