RE: Conversion of MS CSS 2.1 tests to reftests

Running the test suite takes about 3 days. I very much doubt that
building an automation system, testing it and converting thousands
of testcases takes less time.

I don't disagree with the benefits of the approach though. But if that is
what WG members want to do, and if doing that means we'll miss the 
deadline we agreed to they should say so instead of arguing and giving
others ifs and buts. 

Last week David wondered whether Mozilla would submit an
implementation report 'at all'. Now you say it's not worth to try 
'rushing' to REC. (As if running a test suite and
publishing implementation report was a complete surprise 
after two years of submitting testcases ?)

To be continued on the telcon. 

From: Anne van Kesteren []
Sent: Tuesday, September 21, 2010 5:27 AM
To: L. David Baron; John Jansen; Sylvain Galineau
Cc: Geoffrey Sneddon; fantasai; Arron Eicholz;
Subject: Re: Conversion of MS CSS 2.1 tests to reftests

On Mon, 20 Sep 2010 21:31:39 +0200, Sylvain Galineau
<> wrote:
> If that were the case, it would be helpful for you to at least say so
> clearly and unambiguously.

I do think test automation is way more valuable than putting a lot of
effort into running this manually. Apparently we already did the work for
the HTML set of tests, but keeping those up to date is more work than it
would be if we had reftests.

Unfortunately I was not present at this F2F discussion everyone keeps
referring to, but I do not think it is worth it to try rushing to REC
while the test suite is in the state it is in. That does not at all seem
like a productive use of time of our collective resources.

Anne van Kesteren

Received on Tuesday, 21 September 2010 15:22:19 UTC