Re: CSS selectors and xml:id

On Monday, May 9, 2005, 7:42:28 PM, Bjoern wrote:

BH> * Chris Lilley wrote:
>>BH> Could you propose text that should be added?
>>
>>If an implementation supports CSS 2.1 and also supports xml:id, then the
>>xml:id attribute must be treated as an ID for the purposes of CSS
>>selector processing.

BH> Okay, so, CSS 2.1 states

BH>   The ID attribute

That is imprecise in the first two words. It sounds like an attribute
whose name is ID. I can imagine implementors thinking that it means an
attribute called ID (or an attribute called, cases-insensitively, ID). I
can also imagine tham thinking that an attribute called foo, or xml:id,
or whatever, is thereby excluded.

BH> of a document language allows authors to assign
BH>   an identifier to one element instance in the document tree. CSS
BH>   ID selectors match an element instance based on its identifier.

BH> It seems that if your text is meant to change requirements for CSS 2.1
BH> implementations, it must be possible to "support xml:id" in a way that
BH> does not cause xml:id IDs (if any) to be assigned to an element.

Uh, no. Exactly the reverse. My suggested text means that if an
implementation supports xml:id and also supports CSS selectors then (one
consequence is that) xml:id is treated as what CSS calls an 'ID
attribute' in the first sentence and an 'identifier' in the second
sentence.


-- 
 Chris Lilley                    mailto:chris@w3.org
 Chair, W3C SVG Working Group
 W3C Graphics Activity Lead

Received on Monday, 9 May 2005 18:01:26 UTC