W3C home > Mailing lists > Public > www-style@w3.org > June 2005

Re: [CSS21] properties for table-column (In HTML: COL) & table-column-group (In HTML: COLGROUP) items.

From: Emrah BASKAYA <emrahbaskaya@hesido.com>
Date: Thu, 30 Jun 2005 01:23:46 +0300
To: "Laurens Holst" <lholst@students.cs.uu.nl>
Cc: www-style@w3.org
Message-ID: <op.ss5khwlk8nstxa@lomarnona>

On Thu, 30 Jun 2005 01:02:36 +0300, Laurens Holst  
<lholst@students.cs.uu.nl> wrote:

> Emrah BASKAYA wrote:
>> To the skeptical: let the UA's think about it. This is not even hard.  
>> The  second pass would be done right after each element is laid out.  
>> The whole  document tree is kept in memory anyway.
> It is not. Perhaps in your application it is, but in browsers the whole  
> document tree is only available after the document has transferred over  
> the network in its entirity. This may take up to tens of seconds,  
> depending on the document size, client connection and server connection.  
> Not rendering anything until that is done is not acceptable...

I am not saying that the second pass should be done after the document  
loads, that would not useful at all, I agree with you on that indeed.

The key sentence is "The  second pass would be done right after each  
element is laid out." Those elements being table-cells, in this example.

The pass concerning whether that table cell belongs to a certain coloumn  
would be done immediately after that specific cell is laid out, not after  
the document is loaded, and table cells should use the size defines  
padding + border + content-size method so any change in border and padding  
parameters would not require a reflow. Of course, changes in color does  
not require a reflow either. Even with a reflow that may be caused by a  
font-size change, the change is made per cell, so only the last row is  
affected for the reflow, which is very easy on the UA's.

> ~Grauw

Received on Wednesday, 29 June 2005 22:24:00 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:27:18 UTC