- From: Benjamin Hawkes-Lewis <bhawkeslewis@googlemail.com>
- Date: Sat, 05 Apr 2008 15:46:50 +0100
- To: Bjoern Hoehrmann <derhoermi@gmx.net>
- CC: www-style@w3.org
Bjoern Hoehrmann wrote: > * Benjamin Hawkes-Lewis wrote: >> http://www.w3.org/TR/CSS21/syndata.html: >>> The meaning of input that cannot be tokenized or parsed is undefined >>> in CSS 2.1. >> And then http://www.w3.org/TR/CSS21/syndata.html#parsing-errors says: >>> In some cases, user agents must ignore part of an illegal style >>> sheet. This specification defines ignore to mean that the user agent >>> parses the illegal part (in order to find its beginning and end), but >>> otherwise acts as if it had not been there. >> It seems to me ignoring something is equivalent to defining it as having >> no meaning (at least to the CSS 2.1-conforming implementation). > > Yes, this is a well-known contradiction in the specification (that it > says the meaning of some cases is undefined, but then goes on to de- > fine error recovery rules for a subset of those cases, in other parts > than the one you quote). The class of errors under discussion is handled > by the "Malformed declarations" recovery rule despite the first quote > above. Hmm. I don't suppose just dropping the line "The meaning of input that cannot be tokenized or parsed is undefined in CSS 2.1." would help? -- Benjamin Hawkes-Lewis
Received on Saturday, 5 April 2008 14:47:27 UTC