W3C home > Mailing lists > Public > public-css-testsuite@w3.org > November 2009

Feedback on 2 Microsoft submitted tests on outline

From: Gérard Talbot <css21testsuite@gtalbot.org>
Date: Thu, 5 Nov 2009 10:32:36 -0500
Message-ID: <80930016b7a0cb9170914fa204e03aa4.squirrel@69.59.28.71>
To: public-css-testsuite@w3.org
Hello,

There may be a problem with a few Microsoft submitted tests on outline.

Outline border is not necessarly *immediately* surrounding the border-box.
The spec unfortunately says

"
The outline may be drawn starting just outside the border edge.
"
http://www.w3.org/TR/CSS2/ui.html#dynamic-outlines

while it used to say

"
The outline is drawn starting just outside the border edge.
"
http://www.w3.org/TR/2008/REC-CSS2-20080411/ui.html#dynamic-outlines

The difference means that an user agent may have a default, arbitrary
outline-offset of "n" pixels.

This makes some tests not entirely correct or not entirely reliable, like

http://test.csswg.org/source/contributors/microsoft/submitted/Chapter_18/outline-layout-001.xht
(the complete overlapping of border-bottom: 10px solid red; is not
assured, is not certain/guaranteed)

and

the assert
<meta name="assert" content="Outline is drawn directly outside of the
border." />
in this test
http://test.csswg.org/source/contributors/microsoft/submitted/Chapter_18/outline-layout-003.xht
may not be perfectly exact, not perfectly true. Right now, the word
"directly" is somewhat questionable here. Am I nitpicking too far here?

I wish the CSS 2.1 spec would intentionally and deliberately say
"
The outline is drawn starting just outside the border edge.
"
meaning/suggesting that outline-offset is assumed to be 0 by default.

regards, Gérard
Received on Thursday, 5 November 2009 15:35:33 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 20 September 2010 17:52:01 GMT