RE: [RC4] 39 'font-size: 0' testcases

On Tuesday, January 11, 2011 2:44 PM Gérard Talbot wrote:
> Le Mar 11 janvier 2011 14:14, "Gérard Talbot" a écrit :
> >
> > There is now at least 2 browsers (Chrome 8.0.552.224 and Konqueror
> > 4.5.5; I have not checked with Safari 5.0.3 yet) which will fail
> > testcases where the pass condition relies on rendering 'font-size: 0'.
> >
> > One example (but there are a bunch of others) failed by Chrome 8 and
> > Konqueror 4.5.5:
> > http://test.csswg.org/suites/css2.1/20101210/html4/font-size-001.htm
> >
> > Konqueror 4.5.5 (minimal font-size) can not and will not go below a
> > font-size: 2pt (rounded up to 3px).
> >
> > It is kind of weird to have browsers fail testcases because they can
> > not go below a 3px font-size and they can not render appropriately a
> > 'font-size: 0' declaration.

That is a bug in their implementation per the specification. The specification states that a positive length is supported thus 0, 1px, 2px, 3px should all work for font sizes. This is the case for all units that resolve to 1, 2 or 3 pixels.

> 
> Chrome 8.0.552.224, Safari 5.0.3 and Konqueror 4.5.5 will fail this list of 39
> testcases just because they can not render appropriately
> 'font-size: 0' or 'font-size: 1px', 'font-size: 1pt':
> 
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-001.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-002.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-003.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-004.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-005.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-012.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-013.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-014.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-015.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-023.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-024.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-026.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-027.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-034.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-035.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-037.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-038.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-045.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-046.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-047.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-048.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-049.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-056.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-057.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-059.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-067.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-068.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-070.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-078.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-079.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-080.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-081.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-089.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-090.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-092.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-093.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-100.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-101.htm
> http://test.csswg.org/suites/css2.1/20101210/html4/font-size-102.htm
> 
> The negative font-size testcases (eg: font-size: -1px) are coded in a way that
> rely on the tested browsers to render appropriately 'font-size:
> 0'. Those browsers would pass the negative font-size testcases if it was not
> of 'font-size: 0'.
> 
> I think there should be a (positive) minimal font-size assumed for the test
> suite.
> 

There is nothing in the specification that states that the font can't be set at 1px or 0. There is nothing that makes these cases invalid. The user agent if it wants to be compatible needs to support a way to get to 0px. It can still have a default minimum font size it just needs to be adjustable to 0.

--
Thanks,
Arron Eicholz

Received on Wednesday, 12 January 2011 00:15:10 UTC