I'll buy that. My main point was that a fall-back to iso-8859-1 was
even more broken. If, somewhere in the application, there is a fall
back, it should be to the codeset of the current locale. You are
right that it shouldn't be at the utf-8 API level. If it exists it
should be "further up".
On 07 May 2001 14:16:36 +0200, Lauris Kaplinski wrote:
> Hello!
>
> On 05 May 2001 08:16:45 -0400, Richard Schoeller wrote:
> > Better still would be to treat invalid utf-8 as though it were
> > in the codeset of the current global locale. It is very likely
> > that end users outside of western europe will exchange most of
> > their email with users of the same codeset as themselves.
>
> No. This would be completely broken.
> If method name is g_utf8_whatever, then it should accept utf8 and FAIL
> (OK, no crash still), if sting is not well-formed.
> For guessing one can add separate set of methods, like
> g_guess_encoding_xyz()
> and add these to public interface instead of strict utf8 ones.
>
> Lauris
>
>
>
--
Dick Schoeller
mailto:[EMAIL PROTECTED]
http://people.ne.mediaone.net/schoeller/
781.449.5476
"Er ist ein Narr, der meint, es sei nicht schad, das Kind
auszuschütten mit dem Bad" - Thomas Murner 1512
_______________________________________________
evolution-hackers maillist - [EMAIL PROTECTED]
http://lists.helixcode.com/mailman/listinfo/evolution-hackers
