On Fri, Feb 13, 2009 at 3:12 PM, Enrique Nell <[email protected]> wrote: > On Thu, Feb 12, 2009 at 8:23 AM, Gabor Szabo <[email protected]> wrote:
> > The perldiag file contains around 30.000 words. Why don't we extract > the error messages --not the descriptions-- > as a first step? That would reduce the wordcount to ~5.000 words. > After all, that's what shows up in Padre. I am not sure I follow you. The way you need to translate perldiag in order to make it work is to leave the entries =item something in English and translate the rest. Look at http://search.cpan.org/dist/POD2-FR/FR/perldiag.pod I know it is a bit strange and I have not thought about this earlier but your ide of translating the error messages themselves is also good. On a second thought, probably it is better for now to keep them in English. After all in normal usage - when not using Padre - people will see the English error message so better to get used to them. In any case right now Padre does not have a way to show the translated error messages and it uses the fact that the English version is there. A better way to reduce the initial load is to only translate the 10-20 most important explanations. See the http://padre.perlide.org/wiki/TranslationIntro and the PerlMonks page for a list. > I will review/update the Spanish GUI translations in the next few > days. What's the expected release date of v0.28? Don't worry about release dates. We normally release every week or so so whatever gets in is good. The rest will get in the next version. > I have created a multilingual glossary (which you can find attached) > containing Padre's GUI strings for all the languages currently > available. This reference may be useful to get inspiration from other > languages or to decide the correct translation > of ambiguous strings. > The code I have used to create it (named L10N::PO::Glossary) will be > available in CPAN soon. I may be reinventing > the wheel (there may be several tools available to create glossaries > from PO files, like gettext lint tools), but there's nothing > wrong with that, and I plan to add some nice features. That's awesome. If you can create a nice HTML report based on this then we can hook it up to our Subversion repository and on every commit to the .po files we can recreate the page(s). Again you should start with a simple one, I can hook it up and then you can improve it. regards Gabor _______________________________________________ Madrid-pm mailing list [email protected] http://mail.pm.org/mailman/listinfo/madrid-pm
