W3C home > Mailing lists > Public > www-style@w3.org > September 2013

RE: [css-shapes-2] Auto-height and shapes sizing constraints

From: François REMY <francois.remy.dev@outlook.com>
Date: Wed, 4 Sep 2013 20:49:59 +0200
Message-ID: <DUB120-W64ACBB82917E39AAECE6CA5320@phx.gbl>
To: Tab Atkins Jr. <jackalmage@gmail.com>
CC: "www-style@w3.org" <www-style@w3.org>, "stearns@adobe.com" <stearns@adobe.com>
>> I propose to change that to
>>
>> An auto-sized element should use the shape as a constraint in determining its maximum size. In the case of auto-sized elements, values expressed as percentages in the auto-sized direction resolve to 0px to avoid cyclic dependency.
> 
> Resolving percentages to 0px is hostile and unusual. Typically they
> resolve to 'auto' or something equivalent for the property.

In the case of a shape, there's no "auto" thing. The shape defines the zone where the content goes, and the amount of room the content needs to be displayed defines the height of the element. If the shape depends on the height of the element, there's no efficient algorithm that will predict you which size (if any) would allow you to respect both the shape and the content-fit constraints.

In the current spec, the behavior of percentages is not described but in webkit/blink, the behavior is that ALL percentages are resolved to zero when the object is auto-height, even width-related percentages. I would like to lift this restriction for the direction which is not auto-sized. 		 	   		  
Received on Wednesday, 4 September 2013 18:50:26 UTC

This archive was generated by hypermail 2.4.0 : Friday, 25 March 2022 10:08:34 UTC