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

Re: Improvements to the CSS2.1 test harness landed

From: Gérard Talbot <css21testsuite@gtalbot.org>
Date: Wed, 17 Nov 2010 18:37:58 -0800
Message-ID: <4a026a5c4933bade7995ba458ad2ab36.squirrel@cp3.shieldhost.com>
To: "Peter Linss" <peter.linss@hp.com>
Cc: "public-css-testsuite@w3.org" <public-css-testsuite@w3.org>, "www-style" <www-style@w3.org>

Le Mer 17 novembre 2010 18:16, Peter Linss a écrit :
> I just completed a significant change to the way the harness computes
> "least tested" test cases.
> It now does an intelligent sorting of the tests based on where passing
> results are needed most from the browser you are using in order to
> meet CR exit criteria. The tests are re-sorted every six hours based
> on new result data. This should allow us to get the maximum benefit
> from the minimum effort.
> The CSS Working Group still needs data for over 700 test cases. Please
> help us complete CSS2.1 by running the test harness on the latest
> version of your favorite browser(s).
> http://test.csswg.org/harness/
> Thanks,
> Peter


Sometimes, (rare but it happens), some testcases are not tried/tested
because they cause a hang bug (% of CPU activity gets maximized and
remains maximized due to a bug) caused by an infinite reflow-loop. This
happened to me in 5 testcases with Konqueror 4.5.3:





(now can't find the 5th one; it was triggered by resizing the window).

regards, Gérard
Contributions to the CSS 2.1 test suite:

CSS 2.1 test suite (RC3; October 27th 2010):

CSS 2.1 test suite contributors:
Received on Thursday, 18 November 2010 02:38:38 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:13:21 UTC