W3C home > Mailing lists > Public > www-style@w3.org > September 2010

Re: [css3-background] border-radius and wrapping inline elements

From: Tab Atkins Jr. <jackalmage@gmail.com>
Date: Tue, 7 Sep 2010 21:03:27 -0700
Message-ID: <AANLkTinpO1OSCm_ge5JLpcd5TgRNx=HTaTrL9NQTai2Q@mail.gmail.com>
To: Sylvain Galineau <sylvaing@microsoft.com>
Cc: "www-style@w3.org" <www-style@w3.org>
On Fri, Sep 3, 2010 at 7:05 PM, Sylvain Galineau <sylvaing@microsoft.com> wrote:
>> From: Tab Atkins Jr. [mailto:jackalmage@gmail.com]
>
>> Webkit and FF render the whole line as one box and then chop the
>> decorations at the break, while Opera and IE9 preview render as if
>> each line was a separate box, and the break edges were just
>> border:none.
>>
>> http://dev.w3.org/csswg/css3-background/#box-decoration-break covers
>> the appropriate behavior here - FF and Webkit are correct.  Inline
>> boxes should be rendered as a single long box, and then just chopped
>> at the linebreaks when laying out the lineboxes.
>
> This feature remains at risk and, afaik, is not implemented by any browser.
> If it does not pass the CR period then the behavior will be undefined. Unless,
> of course, this was agreed to pass CR at the f2f and I missed it in the minutes?
>
> Again, I am fine with leaving it undefined in Level 3. But I'd rather be explicit
> about it, on testing grounds at the least. If we want to formalize the behavior
> described in box-decoration-break, then the prose may have to move up elsewhere
> and we should talk about this at the next telcon.

I believe CSS 2.1 defines the 'box-decoration-break:slice' behavior as
the correct default in section 8.6:
http://www.w3.org/TR/CSS2/box.html#bidi-box-model.  This is probably
not sufficiently precise, though.  It also sounds like it's talking
about something completely different.  ^_^

~TJ
Received on Wednesday, 8 September 2010 04:04:22 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:31 GMT