Developers of Sympa

[Hatuka*nezumi - IKEDA Soji <address@concealed>]

On Fri, 03 Nov 2006 16:14:11 +0100
Olivier Salaün - CRU <address@concealed> wrote:

> Hatuka*nezumi - IKEDA Soji wrote:
> > I got it about strftime().
> >
> > strftime() accepts/returns a string encoded by charset that is
> > determined by current system locale.  So (also, preferred formats
> > of date/time depend on locale) ---
> >
> >     POSIX::strftime(FORMAT, TIME)
> >
> > may be replaced with:
> >
> >     Encode::decode(gettext("_charset_"),
> >                POSIX::strftime(Encode::encode(gettext("_charset_"),
> >                                               gettext(FORMAT)),
> >                                TIME))
> >
> > (I assume gettext("_charset_") designates the equivalent one of
> > $locale2charset{...}.  Is it right?)
> >   
> We've applied the sympa-MAIN-20061103-strftime.patch patch you proposed, 
> except for the part that concerns xgettext.pl.
> Could you please present what is the goal of the patch for xgettext.pl ?

Preferred formats of date/time depend on locale: For example on 
Japanese, year-month-day(-dayofweek) order is almost exclusively 
used, months aren't written by their names but their numbers, of 
course "at" is English word;) etc.

Modification on xgettext.pl aims to pick up format arguments  of 
gettext_strftime, a new function introduced by remainder of the 
patch, and to take it into PO catalogs.  It may recognize following 
patterns in Perl sources ---

    gettext_strftime "FORMAT", ARRAY...
    gettext_strftime("FORMAT", ARRAY...)
    ...


  --- nezumi