Re: Possible problem with rounded corners

Sent from my iPhone

On Apr 30, 2009, at 10:06 AM, David Hyatt <hyatt@apple.com> wrote:

> https://bugs.webkit.org/show_bug.cgi?id=9543 ?
>
> I assume overflow != visible implies you clip, but that otherwise  
> you don't clip.
>
> dave
> (hyatt@apple.com)

Yes, that is sensible and intuitive.


> On Apr 30, 2009, at 9:47 AM, Brad Kemper wrote:
>
>> The Working Draft does say say nested items are to be clipped, but  
>> so far I don't think any implementations are doing that yet. It was  
>> a more recent change to the Working Draft (a few months ago, as I  
>> recall), and not without some small disagreement from an  
>> implementor. Eventually, implementors will adjust to the change and  
>> implement that part as written, or if no one does then it will be  
>> marked as "at risk" and eventually taken out of the final spec.
>>
>> On Apr 30, 2009, at 5:37 AM, Niels Matthijs wrote:
>>
>>> (repost of my question/remark, apparently the original message was  
>>> posted in a different thread. Excuse my lack of knowledge of  
>>> mailing lists and threaded mails, I’ll try to pay more attention 
>>>  next time)
>>>
>>> I was experimenting with the border-radius implementation of  
>>> Firefox and WebKit and ran into some problems. I checked the ccs3  
>>> specs but couldn't find anything about it, so I'll guess I'll just  
>>> post this here. Again, my apologies if this problem was already  
>>> raised before and is being worked on right now.
>>>
>>> The problem occurs when elements are nested inside a container  
>>> with rounded corners. The background of the container behaves  
>>> well, but the nested elements are still rectangles. So when they  
>>> are given a background-color, they overlap the rounded corners.
>>>
>>> I made a very simple test page, with a horizontal navigation (4  
>>> rounded corners) and hover color on the links in the list items.  
>>> Just hover over the first list item to see the undesired effect.
>>>
>>> http://www.onderhond.com/testpages/rounded-corners.html
>>>
>>> It would be a real shame if this behaviour would still exist when  
>>> we could finally use the property, as workarounds to fix this are  
>>> pretty crappy and really mess up the css.
>>>
>>> Greets,
>>> Niels Matthijs
>>> Internet Architects
>>>
>>
>