W3C home > Mailing lists > Public > www-style@w3.org > June 2008

Re: CSS 2.1 'overflow' applicability to table row groups

From: Andrew Fedoniouk <news@terrainformatica.com>
Date: Tue, 24 Jun 2008 16:06:18 -0700
Message-ID: <48617DEA.2080503@terrainformatica.com>
To: robert@ocallahan.org
CC: "www-style@w3.org" <www-style@w3.org>

Robert O'Callahan wrote:
> On Wed, Jun 25, 2008 at 10:26 AM, Andrew Fedoniouk 
> <news@terrainformatica.com <mailto:news@terrainformatica.com>> wrote:
>     Robert O'Callahan wrote:
>         On Wed, Jun 25, 2008 at 9:40 AM, Robert O'Callahan
>         <robert@ocallahan.org <mailto:robert@ocallahan.org>
>         <mailto:robert@ocallahan.org <mailto:robert@ocallahan.org>>>
>         wrote:
>            The latest CSS 2.1 draft says that 'overflow' does not apply to
>            table row groups. However, a lot of Web sites depend on it
>            applying to table row groups. The spec should be changed to
>            reflect reality.
>         Sorry. It seems only Gecko supports 'overflow' on table row
>         groups. I assumed IE did too since quite a few people are
>         using it and filing bugs about it. (Seems like every time I
>         post to this list I make a fool of myself!)
>         Still, since people are using it, should we add it to the
>         spec? Or should we remove it from Gecko?
>     If you are looking for solution of scrollable tables like here:
>     http://www.terrainformatica.com/htmlayout/images/grid1.jpg
>     then tbody { overflow: auto} will not help you for the following
>     reasons:
>     1) scrollbar has to be a part of the table not tbody. Otherwise
>     widths of cells in thead and tbody (with scrollbar)
>     will not match. And scrolling in horizontal direction has to
>     scroll thead too.
> That's definitely a problem. The testcases I've seen tend to use 
> overflow-x:hidden and overflow-y:auto/scroll so horizontal scrolling 
> doesn't happen.
>     2)  Scrollable tbody would require flex length units:  so you can
>     say tbody height equal the rest that is left from
>     header and footer in table height.
> Not really a problem, you can do that with tables already.
Sorry but to do what? Try this:

    table { border:1px solid black; }
    tbody { border:1px solid blue; }
    thead { border:1px solid green; }

  <table height="100%">
    <tbody height="100%">


And anyway that is about HTML tables where % are close to flex units.
There is no way to define this in CSS.

>     "since people are using it" - haven't seen such uses to be honest.
>     Any links?
> I don't have links to live sites, only people's testcases. However in 
> https://bugzilla.mozilla.org/show_bug.cgi?id=423823 you can see a bug 
> report about a Firefox 3 regression related to scrollable table row 
> groups; there are already five duplicates filed by different people 
> with different testcases, which means they must all be using it. (And 
> of course there must be other people using it who don't encounter the 
> background bug or who haven't tested with Firefox 3 yet or who haven't 
> noticed it or who haven't reported it or who found an existing bug and 
> refrained from posting a duplicate...)
That simply means that people are looking for solution but does not mean 
that they have found it.

I would rather to start thinking about something brand new like <grid 
style="overflow:auto"> with proper behavior
attached (e.g. handling of keys, row selection etc).

Andrew Fedoniouk.

Received on Tuesday, 24 June 2008 23:07:15 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:27:37 UTC