Re: Any conclusion about (C) and ©, and non-ASCII msigid?

[Pablo Saratxaga <pablo mandrakesoft com>]

Kaixo!

On Wed, Dec 18, 2002 at 01:31:24PM -0300, German Poo Caaman~o wrote:
> What will be the "official" position about use 
> non-ASCII msgid?
> 
> I ask, because gettext-0.11 complains about rhythmbox
> messages, it has a multibyte secuence (©) in msgid.

What needs to be done is that msgid and msgstr be both in the same
charset encoding.

In practice, if you want to allow translations in any language, that means
either to restrict msgid to ascii-only (as ascii is a subset of all charset
encodings commonly used), or either to force the exclusive use of utf-8
and nothing else, for both msgid and msgstr.

> what will be the policy about that?

As UTF-8 is being made mandatory for po files anyway, and as in the future
there will exist only utf-8, I would say that if you need to use non-ascii
in msgid entries, just go ahead and do it; you only need to ensure that
you are using UTF-8 in the source files.
And you need to tell translators sending non-UTF-8 po files that they must
convert to UTF-8.

(There is a compatibility mode in gettext tools that allows to ignore
msgid/msgstr charset inconsistencies; but you are better pretending to
ignore it than allowing translators to take very bad habitudes with po
files management)

-- 
Ki ça vos våye bén,
Pablo Saratxaga

http://chanae.walon.org/pablo/		PGP Key available, key ID: 0xD9B85466
[you can write me in Walloon, Spanish, French, English, Italian or Portuguese]

PGP signature