[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-i18n-it
Subject:    Fwd: Fwd: SVN Account Request: Zarko Pintar
From:       Nicola Ruggero <nicola () nxnt ! org>
Date:       2009-06-19 10:12:41
Message-ID: 51980fe0906190312j3c8dd869m6cd68a1bb299a58e () mail ! gmail ! com
[Download RAW message or body]

Ciao,
leggendo questo ho scoperto che altri gruppi usano interessantissimi
automatismi... noi come siamo messi? mi pare traduciamo ancora
old-style (passaggi stable/trunk e viceversa). Qualcuno ha provato le
altre soluzioni? sono effettivamente valide?

http://techbase.kde.org/Localization/Workflows/PO_Summit
http://techbase.kde.org/Localization/Tools/Pology

Nicola


---------- Forwarded message ----------
From: Karl Ove Hufthammer <karl@huftis.org>
Date: 2009/6/19
Subject: Re: Fwd: SVN Account Request: Zarko Pintar
To: kde-i18n-doc@kde.org


Chusslove Illich:

> On the side of KDE repository, it is only necessary that every commit
> comes _from a particular, real person. I.e. you must never allow anyone
> else to use your KDE repository account (give someone your password or
> key), and whatever else you do to organize work within your team is
> perfectly fine.
>
> In particular, I think Norwegian translators are using exactly the
> workflow you describe, so their coordinator (Karl Ove) may provide further
> advice -- or even some ready-made synchronization scripts! :)

Yes, we do use this workflow, and it seems to work very well. Some points:

We use the excellent summit framework that Chusslove has created, and I very
much recommend you do the same. See:
http://techbase.kde.org/Localization/Workflows/PO_Summit
Basically, it amounts to translating files that are the unions of ‘stable’
and ‘trunk’, instead of translating ‘stable’ and ‘trunk’ separately, which
would result in much wasted and duplicated work, and all sorts of problems.

We use svn:external in our SVN to automatically pull in the latest summit
templates and scripts. Then we run our own scripts that do the merging. This
has the advantage that we have complete control over the merging, giving us,
among other things:

– Automatic generation of a compendium to use when merging, for automatic
 translation of new strings that are identical to old strings in different
 files.

– Custom wrapping (using porewrap) of the strings, to make them more
 comfortable to work with.

– Automatic correction of some common spelling and typography mistakes
 (e.g. using ... instead of a real …, inserting non-breaking space in
 front of it, and changing ‘(C)’ to a real copyright symbol), and setting
 the correct PO file headers (PO editors are often wrongly configured).

Also, when transferring the PO files to the KDE repository, we run an
additional sed script to make a few more ortographic changes that we
don’t want to have in our source repository (because they may very well
change in the future).

I can supply the script we use, but be advised that all the comments are in
Norwegian …

We also use a simple script to check all the translation files for syntax
errors every three days, and send an e-mail to the translators nagging them
to fix them. Very useful! :)

--
Karl Ove Hufthammer
_______________________________________________
Traduzioni italiane di KDE: http://kde.gulp.linux.it
https://mail.kde.org/mailman/listinfo/kde-i18n-it

[prev in list] [next in list] [prev in thread] [next in thread]