RE: proposed change for scrollbars in size calculations

You are right, there is redundancy in creating a special case for zero... I hope it is the biggest issue with my proposal ;)

-----Original Message-----
From: Andrew Fedoniouk [mailto:news@terrainformatica.com]
Sent: Tuesday, January 08, 2008 12:03 AM
To: Alex Mogilevsky
Cc: fantasai; www-style@w3.org; Rossen Atanassov; Harel Williams
Subject: Re: proposed change for scrollbars in size calculations

Alex Mogilevsky wrote:
> It appears there is an agreement for the main question of this thread: if width or height is specified on an element with overflow:scroll or overflow:auto, any scrollbars sizes don't affect dimensions of its border box but instead are subtracted from specified width/height.
>
> There is a broad interop on these test cases:
> http://lists.w3.org/Archives/Public/www-style/2008Jan/att-0089/scroll2.htm. 100% browsers use the model where no values of overflow (including 'scroll' and 'auto') affect dimensions of border box.
>
> To make the spec match the implementations, I propose the following updates to section 10:
>
>
>
> 10.3.3 and 10.3.7: Insert (immediately after the formula)
>
>         If 'width' is not 'auto' and scrollbar width is not zero,
>         subtract scrollbar width from 'width'.

Alex, it is just enough to say
          If 'width' is not 'auto' subtract scrollbar width from 'width'.

"and ..." part is redundant there. Or I missed something in principle?

>
> 10.6.4: Insert (immediately after the formula)
>
>         If 'height' is not 'auto' and scrollbar height is not zero,
>         subtract scrollbar height from 'height'.
>

Here too.

--
Andrew Fedoniouk.

http://terrainformatica.com