- From: Christoph Schneegans <Christoph@Schneegans.de>
- Date: Mon, 21 Aug 2006 18:27:40 GMT
- To: <www-validator-css@w3.org>
olivier Thereaux wrote: > The CSS validator is using Xerces for XML parsing. You're completely missing my point. The CSS Validator only uses Xerces when it detects an XML namespace declaration. When no namespace declaration is present or the namespace declaration is preceded by too many characters, the CSS Validator does not use an XML parser. In the latter case, it does not accept attribute specifications such as "xml:space='preserve'", and it does not enforce XML well-formedness constraints. That's why <http://jigsaw.w3.org/css-validator/validator?uri=http://schneegans.de/temp/not-wellformed-without-namespace.html> incorrectly "validates as CSS", while <http://jigsaw.w3.org/css-validator/validator?uri=http://schneegans.de/temp/not-wellformed-with-namespace.html> correctly tells me to "validate my XML document first". Please note that both documents contain an XML declaration and a proper XHTML 1.0 Strict document type declaration, so the decision whether to use an XML parser or not should by quite simple. However, the CSS Validator seems to rely on namespace declarations only. This behavior should be changed. -- <http://schneegans.de/> |
Received on Monday, 21 August 2006 18:28:58 UTC