- From: Tab Atkins Jr. <jackalmage@gmail.com>
- Date: Thu, 14 Mar 2013 16:29:02 -0700
- To: Rossen Atanassov <Rossen.Atanassov@microsoft.com>
- Cc: "www-style@w3.org" <www-style@w3.org>
On Tue, Dec 18, 2012 at 9:30 AM, Rossen Atanassov <Rossen.Atanassov@microsoft.com> wrote: > Reading through the current flexbox spec I see that we're treating the flex flow as orthogonal only based on the writing mode. This is fine for writing mode differences but I don't see why the same wouldn't apply to multiline flex that has a main axis orthogonal to its container? > > The test case below [1] show that exact problem and it seems like Webkit currently does something close to treating it as orthogonal but not quite. Opera and IE are currently in agreement with the spec but the behavior is not good. > > Can we add a clarification about this if we agree on that behavior? Treating it as orthogonal is definitely the wrong thing to do. You want it to take advantage of the available space, not the viewport size. We think we've fixed the issue now, by defining that the min/max cross size of flexboxes (which was previously undefined) to be the cross size after performing layout into the available space. This is similar to how the extent of blocks and the measure of multicols are defined. Take a look and let us know if you agree this solves the issue. (To be specific, we stripped out some of the sizing requirements in step 4, and defined the intrinsic sizing stuff all together in a new section 9.8 <http://dev.w3.org/csswg/css3-flexbox/#intrinsic-sizes>.) ~TJ and fantasai
Received on Thursday, 14 March 2013 23:29:50 UTC