[Webkit-unassigned] [Bug 21680] queryCommandValue("BackColor") always returns rgb(0, 0, 0)

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=21680


Ryosuke Niwa <rniwa at webkit.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
  Attachment #35075|0                           |1
        is obsolete|                            |




--- Comment #6 from Ryosuke Niwa <rniwa at webkit.org>  2010-03-21 03:19:55 PST ---
Created an attachment (id=51244)
 --> (https://bugs.webkit.org/attachment.cgi?id=51244)
more tests

More test cases:
1. The effective background color is of a parent element
2. The background color is explicitly specified
3. An element is placed with position:absolute and inherits the document's
default background.
4. An element is placed with position:absolute on top of another element with
an explicit background color.

Sadly, all major browsers give different results, and MSIE gives the most
reasonable result (passes 1 & 2 and returns white for 3 & 4).  But I'm not sure
of the expected results for the last two tests (3 & 4).  From implementors'
perspective, they are quite unreasonable because we need to extract the color
value from the rendered image.  We need CSS / rendering specialists on deciding
what is the most appropriate behavior here.

Also, we probably need a test with various alpha values.  Consider a situation
where an element has alpha=50%.  Then the background color will be mixed with
that of the parent element's background color.  Should we return the effective
value? or should we just return the value without alpha?

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.