W3C home > Mailing lists > Public > public-qa-dev@w3.org > March 2008

hunting for CSS validator “Regressions” in latest code

From: olivier Thereaux <ot@w3.org>
Date: Mon, 31 Mar 2008 11:51:22 -0400
Message-Id: <67D6105F-E4F6-47DF-9A53-103BBDFEC292@w3.org>
To: public-qa-dev hacking list <public-qa-dev@w3.org>

Hello,

I ran the current test suite for the CSS validator today, trying to  
check if the latest, much cleaner code (especially in the grammar) was  
causing any regression to the known, fixed bugs.

http://qa-dev.w3.org:8001/css-validator/autotest/results/

So far I mostly found one issue with
http://www.w3.org/Bugs/Public/show_bug.cgi?id=576
(which I reopened)

In the "inherit" department, the test file
http://qa-dev.w3.org:8001/css-validator/autotest/testsuite/properties/css3/propertiesCSS3.css
yields a lot of errors when tested with css3 profile, but I think that  
test file may be bogus (gotta check the specs in greater detail).  
Ditto for a couple of test files in the css2 area.

Another little thing I noticed is that the warning for the absence of  
a generic font-family appears to now be "under the radar", ie hidden  
from view unless one selects "all warnings". Given that it is a  
recommendation in the spec prose, should we raise the warning back to  
default level?


-- 
olivier
Received on Monday, 31 March 2008 15:51:58 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 19 August 2010 18:12:48 GMT