[Mailman-Developers] Mailman/Archiver/HyperArch.py has not been marked for i18n

Simone Piunno pioppo@ferrara.linux.it
Tue, 13 Aug 2002 16:39:37 +0200

On Mon, Aug 12, 2002 at 05:14:29PM -0400, Barry A. Warsaw wrote:

>     SP> that no string in Mailman/Archiver/HyperArch.py has been
>     SP> marked for pygettext extraction.

> i18n process.  A brief skim shows a few in-line strings that probably
> need to be wrapped, but not a whole lot.  The nasty bit are the
> templates sitting in the file as triple quoted strings.  These should
> probably be moved out in separate template files.

eh, that's the least...

now I can see why nobody wanted to open this can of worms...
the notion of a MailList object (along with mlist.preferred_language) 
is lost in the process of creating Article objects and 
T.processUnixMailbox() is even worse.

Not to mention the problem with dates localization and mangling
of "2002q3" or "2002-August" volume tags.

There was a template called article.html and it was completely unused
(because Utils.maketext didn't receive any lang or mlist parameter)

Another problem is that I'm using the list language as the archive 
language, so that as soon as you upgrade from 2.1b3 to the patched
release, all your non-english lists will start writing translated
articles.  This is potentially a mess and could force list admins
of non-english lists to regenerate the whole archive (the same will
happen everytime you change the mlist.preferred_language).  
Is this acceptable?

> Care to provide a patch?

I'm working on it.  Expect a large patch.

Adde parvum parvo magnus acervus erit.
Simone Piunno, FerraraLUG - http://members.ferrara.linux.it/pioppo