Re: gettext issue (FreeBSD 5.4)

[Bruno Haible]

Christian Winther wrote:
> Im trying to write a .po file in da_DK (ISO_8859-15) - but every time I run
> the xgettext on the file - and some of the special Danish chars is in the
> .po file - its all converted to UTF-8 - resulting in wrong characters for
> the file and output
>
> xgettext --version =
>
> xgettext (GNU gettext-tools) 0.14.5
> Copyright © 1995-1998, 2000-2005 Free Software Foundation, Inc.
> Dette er frit programmel; se kildekoden for kopieringsbetingelser.  Der er
> INGEN garanti; heller ikke for SALGBARHED eller EGNETHED FOR NOGET SÆRLIGT
> FORMÅL.
> Skrevet af Ulrich Drepper.
>
> uname -a =
>
> FreeBSD chan.guu.dk 5.4-RELEASE FreeBSD 5.4-RELEASE #0: Wed Jun  8 09:25:35
> CEST 2005     address@hidden:/usr/obj/usr/src/sys/system-5.4_firewall
> i386
>
> Is this a known issue ? Have had this problem for 6month+ now..

xgettext is meant to produce a POT file, i.e. a template which can be used
by translators of all languages. Since a Chinese or Russian translator cannot
work with a ISO-8859-15 encoded file, xgettext converts it to UTF-8.

What you probably need is the 'msgcat' program instead.
http://www.gnu.org/software/gettext/manual/html_chapter/gettext_7.html#SEC6

If not, then you can use 'msgconv' to convert the UTF-8 encoded xgettext
output back to ISO-8859-15.

Bruno