Re: Place of PO files for NLS (was Re: [PATCHES] PG_DUMP NLS (Russian)) - Mailing list pgsql-hackers

From Serguei Mokhov
Subject Re: Place of PO files for NLS (was Re: [PATCHES] PG_DUMP NLS (Russian))
Date
Msg-id 002201c15052$4e3f5bc0$5dd9fea9@gunn
Whole thread Raw
In response to Re: Place of PO files for NLS (was Re: [PATCHES] PG_DUMP NLS (Russian))  (Peter Eisentraut <peter_e@gmx.net>)
List pgsql-hackers
----- Original Message ----- 
From: Peter Eisentraut <peter_e@gmx.net>
Sent: Monday, October 08, 2001 7:25 PM

> > > I just thought I would point out that could get
> > > messy and many projects realizing this have created a po directory off the
> > > source tree to store the translations in.  It might be worthwhile to
> > > create such a directory and implement this now then when more and more
> > > translations start pouring in.  Just a very concrete sequential way of
> > > keeping the "real" code visible.
> >
> > There was discussion at some point going on AFAIR on the topic or
> > close to the topic, but I don't recall how it ended. But this idea is
> > not new, and I don't remember why this approach wasn't followed...
> 
> I don't have a strong opinion about this other than "it's easier the way
> it is now".  Once we get annoyed by it we can always change.

Well, it kind of make sense to keep all the po files under a separate directory,
as a special kind of sub-tree, and it's easier to keep track of because all
translations are in the same place, and there isn't much of "distortion"
and "noise" for an eye of a developer looking for the source files in the source
code tree. For instance for psql we have already Czech, German, French, Russian,
Swedish, and two Chinese translations, that is 6 po files, and one expects their
number ot grow. They don't really belong to the source code... 'me thinks'.

How I invision this is to have a /src/nls dir, and components' dirs under it:

/src/nls/       postgres/*.po       pg_dump/*.po       psql/*.po       libpq/*.po       anyothercomponent/*.po

But it's just me, and if it is easier to keep things the way they are
now, well you are the boss, but it might be harder to change them in the future
then if one will want to.

--
S.






pgsql-hackers by date:

Previous
From: Peter Eisentraut
Date:
Subject: Re: Place of PO files for NLS (was Re: [PATCHES] PG_DUMP NLS (Russian))
Next
From: Brent Verner
Date:
Subject: Re: [patch] ALTER RENAME and indexes