- From: Daniel Holbert <dholbert@mozilla.com>
- Date: Wed, 18 Feb 2015 23:23:03 -0800
- To: Greg Whitworth <gwhit@microsoft.com>, www-style list <www-style@w3.org>
On 02/18/2015 04:43 PM, Greg Whitworth wrote: > In our current implementation we render the same as firefox Good to know, thanks. > and I would don't think it would be intuitive if we actually want > to allow width to influence flex-basis but not min and max width. > I think the wording should stay as is. ...but actually, we *don't* allow width to influence the flex-basis in this case. The section of the algorithm I'm talking about (9.2, step 3, part E) says: # [...] lay out the item into the available space using its # used flex basis *in place of its main size*, treating # a value of content as max-content. http://dev.w3.org/csswg/css-flexbox-1/#algo-main-item The "in place of its main size" quote here means that the "width" property is *ignored* here, if you had e.g. "flex-basis:content; width: 100px". That's the goal. But, the language does not specify whether *min/max-width* should be applied when doing this content measurement. I think they should not be applied, since they have an explicit effect later in the layout algorithm; but Blink's implementation disagrees. Given this, do you agree that the wording needs clarification, Greg? ~Daniel
Received on Thursday, 19 February 2015 07:23:34 UTC