W3C home > Mailing lists > Public > www-validator-css@w3.org > July 2003

Re: Something weird goin? on with the CSS-validator.

From: Jukka K. Korpela <jkorpela@cs.tut.fi>
Date: 25 Jul 2003 18:18:57 -0400
To: Ville Koivuranta <ville@virastojazz.net>
Cc: www-validator-css@w3.org
Message-ID: <Pine.GSO.4.50.0307252251140.15526-100000@korppi.cs.tut.fi>




On Fri, 25 Jul 2003, Ville Koivuranta wrote:

> 	An invalid XML character (Unicode: 0x1ace70) was found in the element
> content of the document. "

This would be easier to analyze if you had posted the URLs of the
documents tested.

But I'll make a guess - you use XHTML, and the document is in Finnish and
contains ISO 8859-1 characters like Scandinavian letters as actual 8-bit
data. I got a similar message when I was virtually forced to use an XHTML
doctype for a document in Finnish, and the problem disappeared when
I replaced the Scandinavian letters by entity references like &auml;.

"CSS Validator" apparently runs some XHTML parsing and postulates UTF-8
encoding, then goes wild when it sees your Scandinavian letters, since
those octets have a quite different meaning as part of UTF-8 datastream.

I suppose I don't want to know what the problem really is - it occurred
when my document has a <meta> tag specifying ISO 8859-1 _and_ it was sent
by the server with a Content-Type header saying the same. The media type
and character encoding issues in XHTML are just a big mess to me, and
I avoid them when I can. In your case, using good old HTML 4.01 would most
probably be a better solution, unless someone is pointing a gun at you and
commanding you to use XHTML.

-- 
Jukka "Yucca" Korpela, http://www.cs.tut.fi/~jkorpela/
Received on Friday, 25 July 2003 18:18:58 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 27 June 2012 00:14:12 GMT