- From: L. David Baron <dbaron@dbaron.org>
- Date: Tue, 6 Dec 2011 18:31:05 -0800
- To: "Tab Atkins Jr." <jackalmage@gmail.com>
- Cc: Brian Manthos <brianman@microsoft.com>, "www-style@w3.org" <www-style@w3.org>
On Monday 2011-11-07 14:55 -0800, Tab Atkins Jr. wrote: > On Mon, Nov 7, 2011 at 2:44 PM, Brian Manthos <brianman@microsoft.com> wrote: > > If three or four values are specified, two pairs of a keyword plus a length > > or percentage. > > > > Does that mean thatβ¦ > > center top 5px > > computes to > > left 50% top 5px > > rather than > > 50% 5px (minimized) > > or > > center top 5px (conceptually aligned with the specified > > value) > > ? > > Ah, I see what you're referring to. The B&B text is in conflict with > the suggestion for serialization in CSSOM. You should have mentioned > that in the OP. > > In that case, I'll defer to Elika and Brad. The "Computed Value:" lines in the spec are about semantics rather than syntax. They affect what values are inherited, but serialization specs don't refer to them (or if any do, they shouldn't). Some of them may be phrased in terms of syntax, but that's probably a mistake that should be corrected. (It's often sensible to phrase them in terms of conceptual data structures, such as "list of", "pair of", etc.) So I think what the "Computed Value:" line says isn't relevant to Brian's question (quoted above), or the question relevant to what the line should say. (The getComputedValue() API is about used values, due to some unfortunate turns of history. I think what it does should be specified in cssom-view.) -David -- π L. David Baron http://dbaron.org/ π π’ Mozilla http://www.mozilla.org/ π
Received on Wednesday, 7 December 2011 02:31:30 UTC