[Localization] Re: [OLPC-translation] Translations are afoot! (fwd)

[SJ Klein]

forwarding to the list.

---------- Forwarded message ----------
Date: Tue, 27 Mar 2007 14:02:17 +0200
From: Marco Pesenti Gritti <mpg at redhat.com>
To: Christopher Blizzard <blizzard at redhat.com>
Cc: Sabine Cretella <s.cretella at openprogress.org>,
     Samuel Klein <sjklein at hcs.harvard.edu>,
     Chris Blizzard <blizzard at laptop.org>, Ivan Krstic <ivan at laptop.org>,
     Todd Kelsey <todd at cftwinc.com>, Gerard Meijssen <gerard.meijssen at gmail.com>,
     Adam Brandi <abrandi at laptop.org>, Walter Bender <walter at laptop.org>
Subject: Re: [OLPC-translation] Translations are afoot!

On Mon, 2007-03-26 at 15:26 -0400, Christopher Blizzard wrote:
> Marco == mpg at redhat.com.
>
> --Chris
>
> On Sun, 2007-03-25 at 20:12 +0200, Sabine Cretella wrote:
>> Hi, I just came back home from Pisa. Couldn't we go away from .po? It is
>> just a pain if you need to work on them with OmegaT since you first have
>> to convert them to monolingual .po files and then in the end re-convert
>> them into bilingual ones. We need to create TMX files here to be able to
>> have easy and fast updates.
>>
>> I already asked for it time ago: I would like to see the files that need
>> translation BEFORE they are converted to .po files. It could make sense
>> to create an OmegaT filter for them once and that's it. Or we need a
>> filter that lets us work directly on .po files.
>>

I can see the point for HTML but for python I suppose you don't want to
go hunting down the strings in the code... We can make sure the .po
files have comments to describe the context when it's unclear from the
strings itself.

Also from a quick look OmegaT does not seem to support any format which
can be used for code translation... Hence writing a filter for it would
be necessary in any case. Am I missing something?

Marco