Re: [CSS21] computed value for percentage 'height'

On Fri, 02 Sep 2011 16:09:45 +0200, Peter Moulder  
<peter.moulder@monash.edu> wrote:

> On Fri, Sep 02, 2011 at 11:10:12AM +0200, Øyvind Stenhaug wrote:
>> On Fri, 02 Sep 2011 08:19:31 +0200, Peter Moulder
>> <peter.moulder@monash.edu> wrote:
>>
>> >Furthermore, when I try to test the behaviour (based on the behaviour  
>> of
>> >height:inherit in a child), all UAs I've tested inherit the specified
>> >percentage rather than 'auto' in this condition.
>>
>> Hm, I guess you didn't test Opera.
>
> (I've replied privately re testing Opera.)
>
> Am I right that it only affects the behaviour of height:inherit when
> inheriting from a height specified as a percentage even though that
> percentage acts like height:auto in other respects?  That sounds like a
> very rare case, to the extent that I'd be inclined not to bother
> implementing it: I wouldn't expect anyone to notice.

Well, the spec as it is currently written requires the height to be 'auto'  
in order to trigger the part of 10.6.3 that calculates a height based on  
the contents. I don't know if there are other places that refer to the  
computed value.

> As someone who has implemented it, would you comment on the engineering
> cost of the feature?  Am I right that it's a bit tricky when combined
> with run-in?  (I haven't been following run-in's development since it was
> removed from CSS2.1.)

I haven't been involved with implementing it personally and I'm not  
familiar with the code. I can forward the question.

-- 
Øyvind Stenhaug
Core Norway, Opera Software ASA

Received on Monday, 5 September 2011 16:10:53 UTC