[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kfm-devel
Subject:    Re: Netscape's page and KHTML [Was: openssl license]
From:       Dirk Mueller <mueller () kde ! org>
Date:       2001-09-10 23:46:34
[Download RAW message or body]

On Die, 11 Sep 2001, Ilya Konstantinov wrote:

> > table height is not specified..
> <Toastie> but there's no "specified height" - we got "auto"
> <Hixie> that's specified
> <Hixie> :-)
> <Toastie> but that's rediculous - every piece of content has height,
> non-overflowing one too
> <Hixie> yep
> <Hixie> it's quite simple... the spec is underspecified for this

but who is Hixie and why does he have the authority to declare this case as 
unspecified?

> > > There is a small hint that Mozilla does it right:
> > > "the formatted content may span any number of lines but may not overflow
> > > the cell box." (from the Tables chapter in the CSS2 spec).
> In the 'Automatic table layout' section. That's why I said it's only a
> hint :)

I browsed that section. where should it be ?

> According to the Mozilla guys, the spec doesn't really define it, so we
> should stop looking for meaning that wasn't put there in the first
> place and optimize for the real-world behavior :)

Its just legacy. IE doesn't fully handle the means of overflowing contents 
as defined by CSS as far as I tried it. So as long as there is nowhere defined 
that table cells just cannot have overflowing contents I'm just a bit 
careful. Of course this allows quite some parts in khtml to optimize now, 
but we're loosing a "feature" this way. 
 
The table layout behaviour is not defined at all by CSS actually ("user 
agents are not required to follow this algorithm, they might choose any 
other implemenation that is suitable") so that comment from the Mozilla guys 
doesn't say anything relevant at all. 


Dirk

[prev in list] [next in list] [prev in thread] [next in thread]