Right. The option is a bit cryptic. I first thought of knowntofail-to-failures, but that was way too long.
--- On Wed, 7/22/09, Happy-melon happy-melon@live.com wrote:
From: Happy-melon happy-melon@live.com Subject: Re: [Wikitech-l] Known to fail interactions with compare and record To: wikitech-l@lists.wikimedia.org Date: Wednesday, July 22, 2009, 3:44 PM I assume "ktf" is short for Known To Fail? You've got a bit of RAS syndrome going on there... :-D
--HM
"dan nessett" dnessett@yahoo.com wrote in message news:842292.50373.qm@web32504.mail.mud.yahoo.com...
I have added an application option ktf-to-fail that
when specified
accumulates tests with known-to-fail status as if they
failed. When this
option is missing, failure statistics do not include
known-to-fail results
and there is a summary at the end of parserTests that
specifies how many
known-to-fail tests were run (unless that number is
zero). I also have
modified parserTests to indicate the known-to-fail
status when that option
is specified.
But, there is still an issue. How should the per-test
known-to-fail option
interact with the compare and record application
options? Should
parserTests be modified to record and compare
known-to-fail results? Or,
should these results be silent for recording purposes
and treated as
failures only if the ktf-to-fail application options
is specified?
Wikitech-l mailing list Wikitech-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikitech-l