* Nick Kew wrote:
> On Sat, 18 Jun 2011 22:29:02 +0200
>
> André Malo <[email protected]> wrote:
> > > Doesn't look like it from a glance at the svn. No change to the XSLT
> > > since 2009. More likely the server suddenly started incorrectly
> > > sending it as utf-8.
> > >
> > > However, that does beg the question: shouldn't we move to utf-8
> >
> > This begs another question: Why would we want that?
>
> Well, Vincent has given us one reason. More below ...
Nope, he's talking about the xml input files...
Now your reason...
>
> > > How has infra handled multiple charsets in the past?
> >
> > mod_mime, I believe (modulo some exceptional handling for special
> > extensions). Most certainly, it was delivered as iso-8859-1 before, if
> > not specified otherwise via file extension.
> >
> > AddDefaultCharset iso-8859-1 or even better an AddCharset iso-8859-1
> > .html for the manual section should solve the problem.
>
> For English, yes. For French - up to a point, as noted by Vincent.
> For Japanese, Chinese, etc ... no chance.
The non-latin-1 charsets are overridden by extensions already. Only the
default was changed...
Please do at least *some* checking about what's going on and don't spread
FUD. I won't fight religious wars about UTF-8 being universal or not as
long as XML is universal already regardless of the document charset. As
long there's no reason to change the output, why would we waste time with
that?
(Note that we *did* change the output of the japanese documents to UTF-8 a
long time ago, because there was a reason. The iso-2022-jp charset is a
mess).
nd
--
"Solides und umfangreiches Buch"
-- aus einer Rezension
<http://pub.perlig.de/books.html#apache2>
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
