openerp-community-leaders team mailing list archive
-
openerp-community-leaders team
-
Mailing list archive
-
Message #00115
Re: Efficient translation impossible
On Monday 26 July 2010, you wrote:
> Am Montag 26 Juli 2010, 08:21:08 schrieben Sie:
> Thanks
> Actually not
> 1) I probably do not have commit rights in trunk
If you send a patch with the updated translations, that would be welcome, of
course.
And I think this is a much better procedure, than blindly accepting anonymous
translation strings through LP. There /should/ be a human at the end of the
procedure, that will do a quick check on the strings being submitted, and
perhaps resolve merges of multiple translations (supported by the gettext file
standard).
> 2) IMHO for efficient and coherent translation it would be helpful to get
> ONE big po file with all modules.
Not really. One .po file would be too large, and would also multiply the danger
of having different context strings merged (example the "State" string that has
two meanings).
Since many translation editors are familiar with "project structure", I have a
little bash script that will symbolic link all the translation files into one
directory like ./de/account.po ./de/crm.po etc. That directory will be
recognized by the local editor and help you translate all modules at once.
> of course this could be achieved by loading all modules (whenever there are
> no conflicts), exporting all translations into ONE po file - translation -
> importing - exporting per module - committing - but this takes extra
> hours.
>
> @xrg - IIRC you mentioned another way to translate efficiently ?
>
All of the scripts I'm referring to are at my "openerp" base repo, under the
mandriva dir:
http://git.hellug.gr/?p=xrg/openerp;a=tree;f=mandriva;hb=HEAD
Follow ups