Re: Allowing unknown pseudo classes and unknown pseudo element in the selector.

On Thu, 08 Mar 2012 20:05:27 +0100, Daniel Glazman  
<daniel.glazman@disruptive-innovations.com> wrote:

> Le 08/03/12 19:52, Erik Arvidsson a écrit :
>
>> Today we don't drop the whole rule for unknown tag names, attribute
>> names etc. We should not drop rules for unknown pseudos either.
>
>
> Well. There is no such thing as an unknown tag or attribute name,
> that's the big difference. Tag and attribute names belong to the
> document instance's space, the pseudos belong to the CSS parser's space.

Agreed

> I don't think we should change our old and stable error handling rules
> here.

I don't think we should start allowing unknown pseudos, but I have toyed
around with the idea that when an error is encountered in a complex
selector, we should discard that complex selector only, not the entire
selector list (ie: drop things until the next coma, not the entire
selector). That's similar to what we do in media queries.

Here is the tracker issue on that topic:
https://www.w3.org/Style/CSS/Tracker/issues/223

I still think this behavior would be nicer, but there is some truth to
the assertion that many pages probably rely unintentionally on the
entire selector to be dropped, and failing to to so would
suddenly activate rules that were best left off.

If we can do this change without breaking the web, I'd like it, but
I am not sure we can.

  - Florian

Received on Friday, 9 March 2012 08:57:40 UTC