W3C home > Mailing lists > Public > public-css-testsuite@w3.org > November 2010

Re: direction-unicode-bidi-028 probably invalid

From: fantasai <fantasai.lists@inkedblade.net>
Date: Mon, 22 Nov 2010 22:45:46 -0800
Message-ID: <4CEB631A.1040400@inkedblade.net>
To: "L. David Baron" <dbaron@dbaron.org>
CC: public-css-testsuite@w3.org, Richard Ishida <ishida@w3.org>
On 10/18/2010 01:46 PM, L. David Baron wrote:
> I believe these tests:
> http://test.csswg.org/suites/css2.1/20101001/html4/direction-unicode-bidi-028.htm
> http://test.csswg.org/suites/css2.1/20101001/xhtml1/direction-unicode-bidi-028.xht
> are invalid because they assume that the user-agent style sheet does
> not have the rule:
>    div { unicode-bidi: embed }
>
> In particular, this rule applies to the inner div in the test line,
> and makes it a non-overriding rtl embedding rather than being part
> of the containing overriding one.
>
> This rule is a recommended part of the default style sheet as a
> result of http://wiki.csswg.org/spec/css2.1#issue-148 .

You are correct. I've added
   .test div { unicode-bidi: normal; }
to the test. (In some ways, this defeats the original intention of
the test, so I'm not sure the test is particularly useful due to
this requirement.)

Richard, you will need to update i18n's copy.

~fantasai
Received on Tuesday, 23 November 2010 06:46:22 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 23 November 2010 06:46:34 GMT