CH_DISPLAY and gettext

tridge at tridge at
Tue Jun 21 00:03:01 MDT 2011

Hi Andrew,

The tdb based internationalisation was originally added as it offered
a way for translators to immediately update the translations in Samba
without rebuilding. At the time that wasn't easy to do with .po files.

These days the best approach is to use one of the web based
translation tools. We would register Samba with one of those tools,
and point translators at a web page where they will be presented with
a list of untranslated words/phrases for their language. They can
translate them directly in the web interface. I think it would be good
to adopt that for Samba if someone has the time. We would remove the
old tdb based system.

Regarding the separation of CH_DISPLAY and CH_UNIX, Jeremy may be able
to give more background, but I believe the separation came about
because there were some sites in Japan which used different character
sets for their terminals and their filesystems. This was done because
of some legacy compatibility issues which I think originally came from
the CAP appleshare for unix file server, which had problems with 8 bit
unix character sets. So some weird filesystem encodings were used to
avoid this problem, but sysadmins used terminals that had different
encodings that were better for display.

I suspect this sort of setup is either gone completely now, or so rare
as to hardly matter. Basically UTF-8 has won, and if you're not using
UTF-8 on unix these days then you're pretty crazy.

So I'd support dropping CH_DISPLAY, unless someone can point out a
good reason to keep it. It's even arguable that we could drop CH_UNIX
and just do UTF-8 at some stage, although I suspect there are still
sites around who use something other than UTF-8, just not sites that
use something different for terminals and filesystems.

Cheers, Tridge

More information about the samba-technical mailing list