- From: Mikko Rantalainen <mira@st.jyu.fi>
- Date: Tue, 08 Jul 2003 22:27:48 +0300
- To: www-style@w3.org
Boris Zbarsky wrote:
> Mikko Rantalainen wrote:
>
>>OK, it's the more general problem then. How about:
>>
>>span { position: relative; top: 0; }
>>span:hover { top: 100%; }
>
>
> http://www.w3.org/TR/CSS21/selector.html#x39 explicitly says (and CSS2
> has similar language):
>
> User agents are not required to reflow a currently displayed document
> due to pseudo-class transitions.
>
> In other words, a UA that completely ignores the "top" declaration in
> that span:hover rule would be quite compliant with the spec (and not
> have the circularity problem you raise).
Yes, but it doesn't say "User agents must not reflow..." and I feel that
the spec should define what happens in case UA reflows the document.
Existing user agents try to more or less reflow the document if the
style so requests and AFAIK, page authors usually want this[1]. If the
behaviour should be considered as undefined, the spec should say so.
Currently we have hacks that work in some browsers and logically those
hacks usually make sense so I think the reflow should be allowed.
[1] Eric Meyer has produced a well known example(s) of what we could do
if all UAs supported reflow on hover
<URL:http://www.meyerweb.com/eric/css/edge/menus/demo.html>
--
Mikko
Received on Tuesday, 8 July 2003 15:28:17 UTC