W3C home > Mailing lists > Public > www-style@w3.org > December 2012

RE: [css3-flexbox] stretch apply to percentage heights WAS: Children of flex items ignoring cross-axis percentage length

From: Rossen Atanassov <Rossen.Atanassov@microsoft.com>
Date: Mon, 10 Dec 2012 01:09:17 +0000
To: Tab Atkins Jr. <jackalmage@gmail.com>, Ojan Vafai <ojan@chromium.org>
CC: Rudolph Gottesheim <r.gottesheim@loot.at>, Morten Stenshorne <mstensho@opera.com>, "www-style@w3.org" <www-style@w3.org>
Message-ID: <9A57384B9CE9AE4F9C01F3BB5C3D88440173F368@TK5EX14MBXW604.wingroup.windeploy.ntdev.microsoft.com>
> From: Tab Atkins Jr. [mailto:jackalmage@gmail.com]
> Sent: Monday, December 3, 2012 2:23 PM
> To: Ojan Vafai
> Cc: Rudolph Gottesheim; Morten Stenshorne; www-style@w3.org
> Subject: Re: [css3-flexbox] stretch apply to percentage heights WAS:
> Children of flex items ignoring cross-axis percentage length
> 
> On Tue, Oct 9, 2012 at 11:19 AM, Ojan Vafai <ojan@chromium.org> wrote:
> > http://jsfiddle.net/GgzGf/6/

> >
> > If a percentage height doesn't apply (e.g. because it's standards mode
> > and the containing block is auto-sized), should stretch apply to that flex
> item?
> >
> > Currently in WebKit and, per the flexbox spec, it does not. I don't
> > feel strongly either way. Just want to confirm that this is the desired
> behavior.
> 
> Okay, based on this thread and the existing browser behavior, we've clarified
> step 11 in the layout algo to say that it looks at the
> *computed* value of the cross-size property to tell whether it can stretch or
> not.  Based on the errata mentioned in this thread, that means that
> percentages which later turn into auto behavior won't get stretched.
> 
> If someone feels strongly that we should instead be using used-value-time
> auto (which is theoretically possible, since this is happening in the middle of
> layout), speak up!

What was the reason to refer to the computed instead of used value here, perf? I am more interested to see behavior where flex items don't overlap, so let's try to work on that.

Thanks,
Rossen
Received on Monday, 10 December 2012 01:09:59 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:21:03 GMT