W3C home > Mailing lists > Public > www-style@w3.org > February 2011

Re: list-style-image was(Re: [CSS WG] Minutes and Resolutions 2011-02-23)

From: Tab Atkins Jr. <jackalmage@gmail.com>
Date: Thu, 24 Feb 2011 11:08:29 -0800
Message-ID: <AANLkTik1P83ZqEOHaikH-8N9jrQs1qRdWU7svB9xXbds@mail.gmail.com>
To: Alan Gresley <alan@css-class.com>
Cc: Bert Bos <bert@w3.org>, W3C style mailing list <www-style@w3.org>
On Thu, Feb 24, 2011 at 10:57 AM, Alan Gresley <alan@css-class.com> wrote:
> On 25/02/2011 4:49 AM, Bert Bos wrote:
>> 4. List-style-image
>
>>    SG: Do you have the test case?
>>    TA: looking…
>>    <TabAtkins_>
>>        [16]http://lists.w3.org/Archives/Public/www-style/2011Feb/0364.html
>>    TA: IE9 does what the spec says now. So I'm asking IE to change.
>>    SG: So apparently the spec can be implemented.
>>    TA: Yes, but it is inconsistent with images in other places.
>>    SG: Do you know of a use case?
>>    TA: Such images seem weird to me.
>>    EE: Yes, I think we just spec'ed that for completeness.
>>    SG: Are there any testcases for this now?
>>    TA: Now, there aren't.
>>    SG: Do we need to add some?
>>    TA: We have one implementation, Opera, that does it. So we don't have
>>        two implementations yet.
>>    SG: We have one implementation for the current spec, too. Should we add
>>        a test case at all? Should the edit be in CSS3 instead?
>>    EE: CSS2 and 3 cannot contradict each other, 3 can only be more
>>        precise.
>>    DB: I think we can make the change and add the tests to the pool of
>>        tests to add after the PR test snapshot.
>
>>    RESOLUTION: conditionally accept Tab's edit on list-style-image size,
>>        pending review of tests by Elika.
>
>
> I believe that only Opera 11 renders this test case correctly. Each SVG
> list-style-image has a ratio of 1:1 in the list marker box region and each
> SVG marker is relative to the percentage width and height of the original
> SVG.
>
>
> <http://css-class.com/test/svg/list-marker2.htm>
>
>
> Safari 5 becomes unstable upon loading the test case. Firefox 4b shows the
> SVG marker relative to a default bullet size. IE9 resizes the SVG marker
> according to the width of the viewport. This is not what we want.

This exercises the "percentage intrinsic widths" thing, which we
explicitly don't support in CSS any longer, if I understand correctly.

If you want to create an SVG image with an intrinsic ratio but no
intrinsic dimensions, omit @width and @height but add @viewBox, like:

...doctype stuff...
<svg viewBox="0 0 100 200">
  ...stuff...
</svg>

This creates an SVG image with no intrinsic dimensions but an
intrinsic ratio of 1:2.

~TJ
Received on Thursday, 24 February 2011 19:11:53 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:37 GMT