W3C home > Mailing lists > Public > public-css-testsuite@w3.org > July 2011

Re: [RC6] before-after-dynamic-attr-001 : ought to have a "may" flag

From: Geoffrey Sneddon <gsneddon@opera.com>
Date: Mon, 04 Jul 2011 19:09:32 +0100
Message-ID: <4E1201DC.4010403@opera.com>
To: css21testsuite@gtalbot.org
CC: Public CSS test suite mailing list <public-css-testsuite@w3.org>
On 02/07/11 01:45, "Gérard Talbot" wrote:
> http://test.csswg.org/suites/css2.1/20110323/html4/before-after-dynamic-attr-001.htm
>
> http://test.csswg.org/suites/css2.1/nightly-unstable/html4/before-after-dynamic-attr-001.htm
>
> Konqueror 4.6.4 fails this testcase because, as far as I can tell, it
> can not execute this DOM instruction
>
> line 22:   document.body.setAttribute("my-attr", "before");
>
> ... which has nothing - absolutely nothing - to do with CSS 2.1 and with
> HTML 4.01.
>
> If only, at least, the targeted attribute in the test was an existing
> one in the DTD to begin with ...
>
> I say such testcase should have a "may" flag at minimum, otherwise it
> should not be in the CSS 2.1 test suite.

It has the DOM flag, so expecting DOM Core to work seems reasonable. 
 From your description, it appears it fails because Konqueror doesn't 
implement the DOM properly: there's nothing in DOM Core that says that 
setAttribute should fail if setting the attribute would make the 
document invalid, and hence it should succeed in adding that attribute.

As such, I'd say Konqueror fails a prerequisite of the tests with the 
"dom" flag: having a working DOM implementation.

Given a working DOM implementation, the test seems like a perfectly 
reasonable thing to test.

-- 
Geoffrey Sneddon — Opera Software
<http://gsnedders.com>
<http://opera.com>
Received on Monday, 4 July 2011 18:10:11 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 4 July 2011 18:10:16 GMT