- From: Frank Ellermann <nobody@xyzzy.claranet.de>
- Date: Wed, 03 Nov 2004 01:49:50 +0100
- To: www-validator@w3.org
- Cc: abr@abrunner.com
Adrian Brunner wrote:
> 1) Why don't you accept this code (€) ?
u+0080 is not allowed for SGML (or whatever the error message
said, it won't be better with XML). Latin-1 0x80 is u+0080.
> 2) What should be the correct code to validate the page
> correctly ?
The code is correct for windows-1252 instead of Latin-1, just
replace the charset, that's also the solution working with
many legacy browsers. You could also use € working with
any document charset, but not with all browsers.
Don't try € That's incorrect with all charsets even if
it appears to work with some browsers. It took me two years to
get this simple idea, starting with HTML 4 stuff like €
is always interpreted as Unicode point (no matter what the
document charset is), and u+0080 (Unicode point 128) is always
illegal.
Instead of € you could also use € but my good old
Mozilla 3 is consequent and doesn't recognize both forms. It
can handle the byte 0x80 correctly, and the validator believes
it if you use windows-1252.
Bye, Frank
Received on Wednesday, 3 November 2004 00:52:49 UTC