- From: Tab Atkins Jr. <jackalmage@gmail.com>
- Date: Mon, 24 Nov 2014 14:51:32 -0800
- To: Manuel Rego Casasnovas <rego@igalia.com>
- Cc: www-style list <www-style@w3.org>
On Wed, Nov 19, 2014 at 6:03 AM, Manuel Rego Casasnovas <rego@igalia.com> wrote: > On 12/11/14 23:07, Tab Atkins Jr. wrote: >> No, we're just being sloppy with terminology in 8.4, and with the >> "Applies to" lines in the propdef tables. ("Applies to" is just an >> informative note, anyway.) >> >> The grid-placement properties definitely apply to abspos elements >> whose containing block is a grid container, even if they're not >> children. They alter the containing block to be the grid area >> described by the properties (with the alteration, described in 8.4, >> that an "auto" value *does* contribute a line, namely the >> corresponding padding edge of the grid container). >> >> I'll fix to make this clearer. > > Just an extra thing about this, probably the spec should include also > fixed positioned elements (not only absolutely positioned). > > As in some cases (for example using "transform" property [1]) the grid > could be the containing block of the fixed elements. I guess that the > same behavior than absolutely positioned elements should be applied. The term "absolutely positioned" applies to fixpos as well. The only difference between abspos and fixpos is the containing block. If a fixpos is switched to having an element as a containing block, it'll already get caught by wording about "absolutely positioned" elements. ~TJ
Received on Monday, 24 November 2014 22:52:20 UTC