- From: Tantek Çelik <tantek@cs.stanford.edu>
- Date: Tue, 21 Apr 2015 20:33:26 -0700
- To: Florian Rivoal <florian@rivoal.net>
- Cc: www-style list <www-style@w3.org>, Tantek Çelik <tantek@cs.stanford.edu>
On Fri, Apr 10, 2015 at 8:47 AM, Florian Rivoal <florian@rivoal.net> wrote: > I hadn't noticed earlier, but the definition of caret-color as > currently in the spec has 2 issues. > > 1) The computed value of auto is currentColor > > That makes the auto value identical to currentColor, and not useful. Agreed. > I believe we should follow the example of auto in outline-style, and to leave UAs > free to be innovative about what what it does, and offering a suggestion of a basic > behavior (same as currentColor being reasonable here) if they have no particularly > useful idea. > > Proposed phrasing: > > The computed value for auto is auto. > > The auto value permits the user agent to automatically adjust the color of caret > to ensure good visibility and contrast with the surrounding content, possibly > based on the text color, background, shadows, etc. User agents may treat > auto as currentColor. I think we should still go with a "should" for auto being treated as currentColor, and the provide explicit may language about adjustment. I'll add some text accordingly. > 2) The property is marked is not inherited > > It should be inherited, similarly to the 'color' property, Agreed, not sure why we didn't do that originally. > Both Bloomberg's implementation in chromium[1] and Daniel's implementation for Gecko[2] treat it as inherited. Even better knowing that. > [1] https://github.com/bloomberg/chromium.bb/commit/4a0f4d723a14c0634734da94eef383884488b99f > [2] https://bugzilla.mozilla.org/show_bug.cgi?id=1063162#c12 > Thanks, Tantek
Received on Wednesday, 22 April 2015 03:34:35 UTC