Microsoft CSS3 selectors testcases

Hi Arron,

I was looking at the testcases in
http://samples.msdn.microsoft.com/ietestcenter/css3/selectors_harness.htm
that Firefox trunk is marked as failing.  Since I suspect Microsoft
either has or is planning to contribute these to CSS test suites,
I'm cc:ing public-css-testsuite as well.  (If there's a better
contact for these tests, feel free to let me know.  I'm guessing
that if it's not you, you probably know the right person, though.)

I think in both of the tests that Firefox trunk is marked as
failing, Firefox's behavior is correct.  (The third that was broken
in Firefox 3.6.3 has been fixed on trunk:
https://bugzilla.mozilla.org/show_bug.cgi?id=553805 .)

In particular, this test:
http://samples.msdn.microsoft.com/ietestcenter/css3/selectors/fieldset.htm
asserts that a fieldset element should match :enabled.  But
http://www.w3.org/TR/html5/interactive-elements.html#selector-enabled
describes which HTML elements should match :enabled, and fieldset is
not listed.  (This makes sense to me, since fieldset isn't really a
form control; it's something that groups form controls.)

Second, this test:
http://samples.msdn.microsoft.com/ietestcenter/css3/selectors/indeterminate.htm
asserts that a radio input matches :indeterminate.  However, the
only HTML elements that :indeterminate applies to are checkboxes,
according to HTML5:
http://www.w3.org/TR/html5/interactive-elements.html#selector-indeterminate

-David

-- 
L. David Baron                                 http://dbaron.org/
Mozilla Corporation                       http://www.mozilla.com/

Received on Saturday, 5 June 2010 00:10:04 UTC