Re: [CSS 2.1] cases that do not pass in any browser

On Mon, Aug 2, 2010 at 6:44 PM, Boris Zbarsky <bzbarsky@mit.edu> wrote:
> On 8/2/10 6:34 PM, Arron Eicholz wrote:
>>
>>
>> http://test.csswg.org/source/contributors/mozilla/submitted/css2.1/selectors/first-line-floats-002.htm
>
> This is a very interesting testcase that I wrote because the spec is pretty
> clear here but the behavior of some UA's didn't make sense.  The testcase is
> basically:
>
>  <style>
>    div { color: red; }
>    div:first-line { color: green }
>  </style>
>  <div>
>    <span>
>      <span style="float: left">This should be green</span>
>    </span>
>  </div>
>
> Ignoring the first-line business for the moment, the float inherits its
> color from the outer <span>.  But section 5.12.1 explicitly says that this
> markup should behave just like this markup:
>
>  <div style="color: red">
>    <div:first-line style="color: green">
>      <span>
>        <span style="float: left">This should be green</span>
>      </span>
>    </div:first-line>
>  </div>

Agreed; the spec is very clear that this is the correct element-tree.
The color of the float then follows naturally.

~TJ