Jukka K. Korpela wrote:[color=blue]
> Michael Rozdoba <mroz@nowhere.invalid> wrote:
>
>[color=green][color=darkred]
>>>It's largely a matter of taste whether such appearance is good or
>>>bad, but it seems to be difficult to style it in CSS - actually I
>>>don't think there's a way to _describe_ all of its features in CSS
>>>even in principle.[/color]
>>
>>As a matter of interest, can I ask which features you have in mind
>>that present a problem?[/color]
>
>
> I don't there's a way to describe the rounded corners that e.g. IE
> on WinXP uses.[/color]
Ah, I've not seen that; yes, it would be tricky.
[color=blue]
> Setting border properties for fieldset affects its
> special border in some cases (property/browser combinations) only.[/color]
Ok.
[color=blue]
> The placement of the legend might be describeable using positioning or
> negative margins, but I'm not sure how well that would work in
> practice, especially since we don't really know what browsers are
> doing.[/color]
Yes, this is the bit I had a fiddle with. Thought I'd cracked it with
http://osiris.sunderland.ac.uk/~cd2m...gySeeOpera.htm
but as the file name implies, it doesn't work with Opera (& in any case,
the use of span is hacky in that context).
Putting the legend text into the group div with suitable 0 margin &
padding & setting line height to 0 gets the right result in IE 6 &
Firebird 0.7, but not in Opera 7.23.
A read of the spec suggests Opera is getting it right by assuming
vertical-align of baseline within this zero height box aligned with the
div top border. Setting other aligns on the Span fixes this in Opera but
breaks it in IE. So even if the spec allows it, it doesn't seem to be
easily doable in practice.
A previous attempt with negative margins also looked promising but broke
in IE & I think I've read use of negative margins is less reliable in
practice than tinkering with line-height.
--
Michael
m r o z a t u k g a t e w a y d o t n e t
