Bug 56994 – :not should accept "a comma-separated list of compound selectors"

• Status: RESOLVED
• Resolution: FIXED
• Component: CSS

[Ojan Vafai]

http://dev.w3.org/csswg/selectors4/#negation

This does mean we'll fail http://samples.msdn.microsoft.com/ietestcenter/css3/selectors/notSelectorSequence.htm, but I think it's better that we match the better behavior of the Selectors 4 spec. I'll try and find someone at Microsoft to see if they're willing to change the test.

[Ojan Vafai]

Created attachment 86903 [details]
Patch

[Ojan Vafai]

I've emailed Resig and MS to see if they're willing to change their tests to match the new spec.

[Eric Seidel (no email)]

Any update?  It's been about a month.

[Adam Barth]

Comment on attachment 86903 [details]
Patch

Please re-nominate when you hear back from John.

[Ojan Vafai]

The Resig test suite appears to also live at http://dev.w3.org/2006/webapi/selectors-api-testsuite/, which covers selectors levels 2.1 and 3. And it's not clear who maintains it.

This is changed in http://dev.w3.org/csswg/selectors4/, which does not have a test suite yet.

Is anyone opposed to me committing this without getting this test suite updated since it's testing selectors 3 and the API change this implements is part of selectors 4 given that I can't get responses from Resig or MS?

[Benjamin Poulain]

WebKit supports :not() of Level 4.