Re: Translation coordination robot, who?
[ Sorry for being late at the party. Been busy with other stuff in Debian :]
On Tue, Dec 27, 2005 at 12:03:41AM +0100, Javier Fernández-Sanguino Peña wrote:
> I'm setting up the translation coordination robot for the spanish team,
> since the server were it was set up previously (asgard) no longer works.
> I was wondering which translation teams are using that coordination robot,
> I'm currently aware of, at least:
> - Catalan team, available at http://ca.debian.net/
> (in CC:)
That would be me. =) The bot we (the Catalan team) are using is a "fork"
from the Dutch team one. The bot itself (the part that listens to the
mailing list) is in Perl, and the web interface is in PHP, it uses MySQL
to keep track of the translations. I patched it to fix some of the bugs
that we found at the beginning when we were setting it up, and
replaced the BTS html spider code with the LDAP gateway stuff.
> - Spanish team, available at http://www.debian.org.es/cgi-bin/l10n.cgi?team=es
This one is a rewrite in Python by Ricardo Javier Cardenes (in CC:),
this one has multiteam support and selectable database backend (among
> - French team, uses the pseudo-url format but is there a web interface?
> It would be interesting to have all the code of the translation coordination
> robot somewhere. I would like i18n teams to forward me their code (if they
> are using it) so I can put together a common release that we could publish at
You'll have a hard time trying to merge all those forks and rewrites. =)
I'll clean our bot sources a bit and will send you a tarball. What's the
status on this anyway, any progress?
> On the other hand, which translation teams would be willing to use the
> translation robot themselves and currently don't have the means to set it up
> themselves? Would it be useful to have a common infraestructure for that?
I've wanted from the beginning to get rid of our single-team bot and to
setup one to be shared with all teams in Debian (sadly I've not found
the time to do so), we even talked about that at the latest Debconf. That
would reduce the duped work alot, and we should have more or less the same
needs for all teams.