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

Re: Tests and http

From: Gérard Talbot <css21testsuite@gtalbot.org>
Date: Sun, 19 Dec 2010 12:10:30 -0800
Message-ID: <76117d2281159fa17eb8fff7ed5b0d09.squirrel@cp3.shieldhost.com>
To: "Linss, Peter" <peter.linss@hp.com>
Cc: "public-css-testsuite@w3.org" <public-css-testsuite@w3.org>
Le Ven 17 décembre 2010 13:38, Linss, Peter a écrit :
> Elika and I recently discovered that due to a configuration glitch on
the test suite server, a number of tests requiring specific http
headers
> weren't being served with those headers (this happened with the move
to
> the new server in July).
> Therefore I'm going to remove all results for those tests that were
gathered through the harness interface for RC4. The full list of
affected tests is below.
> The other problem I found with these, is that only five of these tests
have the 'http' flag in their metadata (this error is still in RC4).

Peter,

The files

http://test.csswg.org/suites/css2.1/20101210/html4/content-type-000.htm

http://test.csswg.org/suites/css2.1/20101210/xhtml1/content-type-000.xht

http://test.csswg.org/source/contributors/gtalbot/submitted/content-type-000.htm

http://test.csswg.org/source/contributors/gtalbot/submitted/content-type-000.xht

should have the http flag and they have.


OTOH, the files

http://test.csswg.org/suites/css2.1/20101210/html4/content-type-001.htm

http://test.csswg.org/suites/css2.1/20101210/xhtml1/content-type-001.xht

http://test.csswg.org/source/contributors/gtalbot/submitted/content-type-001.htm

http://test.csswg.org/source/contributors/gtalbot/submitted/content-type-001.xht

do NOT have the http flag and they should NOT have it.



I have gotten in the past (with Opera 10.63) and I still get today
totally different results when trying those same testcases on my
webhosting server with latest and recently released Opera 11.0 (build
1156) browser:

http://www.gtalbot.org/BrowserBugsSection/css21testsuite/content-type-000.html

http://www.gtalbot.org/BrowserBugsSection/css21testsuite/content-type-001.html

> Which leads me to question the validity of the results for those tests
that we received from implementation reports as they may not have been
tested with the proper headers.
> I'd like all browser vendors who submitted implementation reports for
these tests to confirm that they were tested properly despite the lack
of metadata, or I'll be removing those results for these tests as well
from RC4 (this will only affect results grandfathered in from previous
versions of the test suite).

As far as the Opera browser is involved, I think you should remove the
results for content-type-000 and content-type-001

With View/Panels (F4)/Info panel, I can see the content-type of the
http://www.gtalbot.org/BrowserBugsSection/css21testsuite/support/plaintext.css

and

http://www.gtalbot.org/BrowserBugsSection/css21testsuite/support/plaintext-css.txt


http://web-sniffer.net/
should report
Content-Type:	text/plain; charset=utf-8
for HTTP response header for
http://test.csswg.org/suites/css2.1/20101210/xhtml1/support/plaintext-css.txt
and
http://test.csswg.org/suites/css2.1/20101210/html4/support/plaintext-css.txt

and for

http://test.csswg.org/suites/css2.1/20101210/xhtml1/support/plaintext.css
and
http://test.csswg.org/suites/css2.1/20101210/html4/support/plaintext.css

The differences of rendering of the involved testcases in Opera 11.0 is
a mystery to me.

regards, Gérard
-- 
Contributions to the CSS 2.1 test suite:
http://www.gtalbot.org/BrowserBugsSection/css21testsuite/

CSS 2.1 test suite (RC4; December 10th 2010):
http://test.csswg.org/suites/css2.1/20101210/html4/toc.html

CSS 2.1 test suite contributors:
http://test.csswg.org/source/contributors/
Received on Sunday, 19 December 2010 20:11:06 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Sunday, 19 December 2010 20:11:14 GMT