- From: Tab Atkins Jr. <jackalmage@gmail.com>
- Date: Fri, 7 Mar 2014 11:44:04 -0800
- To: Sergio Villar Senin <svillar@igalia.com>
- Cc: www-style list <www-style@w3.org>
On Fri, Mar 7, 2014 at 11:38 AM, Sergio Villar Senin <svillar@igalia.com> wrote: > En 07/03/14 20:18, Tab Atkins Jr. escribiu: >>> BTW, although the computed value for the positioning properties is >>> clearly specified as "specified value" I wonder what would be the >>> computed value for those properties in those error situations. In the >>> example above for getComputedStyle(grid-column-end) would it be "span C" >>> or "auto" ? >> >> Yes, it used to actually change the computed value in the >> error-handling. We moved away from that because it ended up being >> terrible to spec, so now it just refuses to contribute lines or spans >> to the position. > > Cool, thanks for the clarification. > > Related to that, what about the computed value of <grid-line> in the > grid placement properties? I mean, the specs mention how to resolve for > example a lone <custom-ident>, it might be an implicit grid line created > from a grid area, an explicit grid line or even resolved to "auto". My > question is basically the same, should we use that "resolved" position > as the return value of getComputedStyle() or shall we just use the plain > specified value? > > (similar things happen to <integer> && <custom-ident>? for example) Just follow what the spec says - return the specified value. ~TJ
Received on Friday, 7 March 2014 19:44:52 UTC