Re: Specifying intrinsic width and table layout behavior

----- Original Message ----- 
From: "L. David Baron" <dbaron@dbaron.org>
To: <www-style@w3.org>
Sent: Saturday, September 02, 2006 2:11 PM
Subject: Re: Specifying intrinsic width and table layout behavior

>On Saturday 2006-09-02 13:10 -0700, Andrew Fedoniouk wrote:
>> David, I cannot see why fourth box should have the same width as others 
>> in
>> these tests:
>>
>> http://dbaron.org/css/test/intrinsic/block-overflowing-block-min
>> http://dbaron.org/css/test/intrinsic/block-overflowing-text-max
>> http://dbaron.org/css/test/intrinsic/block-overflowing-text-min
>>
>> I've attached screenshot of what I have on them.
>>
>> Last block is defined as:
>>
>> <div id="ref" style="width: 1em"> </div>
>
>The fourth block is the reference rendering.  The issue is that the
>other blocks should all be narrower, as they are in Mozilla, Opera, and
>Konqueror (rather than the way you show, as they are in IE/Windows).  In
>other words, an explicit width on a block should prevent that block's
>descendants from affecting the intrinsic width of that block's
>ancestors.

I see, so it should be rendered as
http://www.terrainformatica.com/w3/test2.jpg

And it will if I implicitly define
  div#testo { width: 1em; overflow:visible; }

in your block-overflowing-block-min

I do have such thing as overflow:none; as a default value
of overflow attribute. To prevent things like
http://www.terrainformatica.com/w3/w3c-fp.jpg
Such renderings are very undesirable in cases I am dealing with.
(embedded HTML islands and small screen formats + IE legacy)

It appears that overflow:visible in IE is broken
it never overflows on overflow:visible.

Andrew Fedoniouk.
Terra Informatica Software, Inc.

Received on Saturday, 2 September 2006 21:55:12 UTC