Re: [css3-background] Curved borders intersecting backgrounds of inner boxes

On Apr 12, 2010, at 12:46 PM, Brian Manthos <brianman@microsoft.com>  
wrote:

> On Apr 12, 2010, at 11:22 AM, fantasai <fantasai.lists@inkedblade.net>
> wrote:
>> What if there are multiple 'background-clip' values?
>
> On 04/12/2010 10:38 AM, Brad Kemper wrote:
>> What I'm getting at, is that this is not just an issue for
>> 'border-radius', it is also for 'background-clip'. We should add
>> something to 'background-clip' similar to what we are saying about
>> border-radius:
>>
>>   Other effects that clip to the border or padding edge (such as
>>   'overflow' other than 'visible') also must clip to the box  
>> indicated
>>   by 'background-clip'.
>
> Brad, how do you see fantasai's question reconciling with your  
> proposed text?

I took it as rhetorical, given that she was saying we therefore  
shouldn't let background-clip affect overflow clipping. I responded to  
the list that this stance does not seem to jibe with the text about  
clipping to the "appropriate" curve for border-radius (ie inner  
[padding box] or outer [border box]) that we (the wg) previously  
agreed to, since determining which curve was appropriate means looking  
at 'background-clip'. If we do look at  'background-clip', then  
perhaps we use whichever one in the background that is biggest.

I don't know the answer, but it looks like something needs to change  
or be clarified. I WOULD like overflow to clip to the inner curve when  
appropriate, and to me that means also clipping the straight portions  
of the box too, to a box that aligns with those curves.