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

List:       kde-i18n-doc
Subject:    Re: Please run ./update_xml
From:       Pablo de Vicente <pvicentea () wanadoo ! es>
Date:       2002-10-26 18:06:32
[Download RAW message or body]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

El Sábado 26 de Octubre de 2002 18:37, Éric Bischoff escribió:
> Le Saturday 26 October 2002 17:34, Pablo de Vicente a écrit:
> > Hello Eric,
>
> Hola Pablo,

 Hello again,


>
> >  I think that the statistics have to take into account the fact that once
> > you modify a PO, scripty will probably create a new version of it several
> > hours later (in the next round). I am not sure this is a general
> > behaviour, but I would think it is.
>
> It is already taken into account. All commits from scripty are just
> ignored.

 Aaah.... That is very good rule indeed.

>
> >   For example I commited a new PO:
> > kde-i18n/es/messages/docs/kdeadmin/kcron.po on 25/10/2002 and several
> > minutes later I commited the docbook. Today the spanish team has nothing
> > to regenerate but tomorrow we will have (wrongly) because scripty has
> > modified that PO, but has left it untouched. Have a look.
>
> Normally, that won't happen tomorrow.

 Yes, I agree.

>
> >   I do not know the reason a modified PO gets checked by scripty several
> > hours later, in the next round, but I think it makes sense so that the
> > teams do not miss possible new strings. This is a bit akward but somebody
> > could confirm it (perhaps Stephan knows this?)
>
> I think I know why. Scripty's script corrects at least indentation and
> length of lines.
>
> >   If one team modifies one PO and commits it to the CVS after the
> > corresponding new POT is commited from where scripty runs (and without
> > having msmerged their PO with that POT locally) and before the PO is
> > commited from where scripty runs, it may happen that the team misses some
> > changes in the PO. Therefore I think it is reasonable that scripty
> > msmerges that file in the next round to check nothing was left behind.
> >
> >  If this is true (perhaps it is a silly speculation), the doc statistics
> > page should take this into account, because all modified POs will have a
> > more recent date than their docbooks.
> >
> > Just a final note: I may be completly wrong :-)
>
> Normally, you are wrong - unless I did a bug in the code that excludes
> scripty's commits, which is always possible.
>
> Thanks for your comments, and I hope my explanations help.

 Yes, now it is clearer. Thanks a lot.

Pablo.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.0 (GNU/Linux)

iD8DBQE9utm0SItUpHl6kJERAtZpAJ4j96ox3em28sV1qLUWNjI9IKqFRACg9mGl
gT/rJHib4ASRCeygTikEv8Q=
=/97F
-----END PGP SIGNATURE-----

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