On Thu, 2008-02-07 at 17:11 +0100, Terje Røsten wrote:
> Jeremy Katz
> >> Some languages don't work well with UTF-8 po files (fx. danish and 
> >> norvegian they have some special 'åæø' chars there looks wrong if the po 
> >> file is not in iso-8859-1 )
> >>     
> >
> > Ermmm, no.  ISO-8859-1 is a subset of UTF-8.
> 
> Some ways yes, in others no. ISO-8859-1 is not a subset of UTF-8, the 
> encoding for e.g. øæåØÆÅ
> is different in ISO-8859-1 and UTF-8. However all chars in ISO-8859-1 
> are available in UTF-8 and much
> more of course. ascii however is true subset of UTF-8, the encoding is 
> identical for the all ascii chars.

The encodings are different, yes.  That's why you have to have
consistency and not mix encodings.  In a perfect world, UTF-8 would have
existed from the beginning and these sorts of legacy problems wouldn't
have to exist.

> That's reason people  in the US etc are using UTF-8, while folks in 
> europe (which needs really unicode) still are using
> ISO-8859-1. Not funny at all...

Some people just can't let go of the past.  Please don't frame this as
US vs Europe as it's anything *but* -- the transition in Red Hat Linux
back in the day was actually driven by a *Norwegian*.  And this was over
5 years ago now.

The key is consistency.  We mandate that strings (and po files, etc) in
yum are UTF-8 and then don't worry about it.  If someone happens to be
running in a non-UTF8 locale, then gconv bits in glibc will take care of
things for them.  

Jeremy

_______________________________________________
Yum-devel mailing list
[email protected]
https://lists.dulug.duke.edu/mailman/listinfo/yum-devel

Reply via email to