Bill Barker wrote:
----- Original Message ----- From: "Brent Worden"
<brent.wor...@gmail.com>
To: "Commons Developers List" <dev@commons.apache.org>
Sent: Wednesday, May 20, 2009 8:28 PM
Subject: [math] Re: [g...@vmgump]: Project commons-math (in module
apache-commons) failed
The test failed because it couldn't detect equality between two NaN
values. I modified the TestUtils.assertEquals method for double
arrays to use our TestUtils.assertEquals method for doubles instead
of Assert.assertEquals. Our method has the needed NaN checks. With
this changed, I enabled the test as well. This should make the test
pass but, if GUMP complains again, I'll disable the test.
I think this a Gump bug. [math] still uses the Ant build in gump, and
the Ant project still exports the junit3.6 jar (so there are two junit
jars on the classpath). Even with Phil's original patch, I could
build with Ant and junit4 (jdk 1.5).
It's a long weekend here, so I may look into how much damage changing
the Gump Ant to use junit4 causes.
Thanks, guys! I am very happy that it had nothing to do with different
behavior when using the same random seed, which was my first thought. I
should have used the TestUtils comparison. Thanks for fixing that, Brent.
Phil Steitz wrote:
The last failure was due to a test I introduced in r776311
(NaturalRankingTest.testNaNsFixedTiesRandom). I have disabled the
test for now. Sorry for the noise...
Phil
Luc
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
For additional commands, e-mail: dev-h...@commons.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
For additional commands, e-mail: dev-h...@commons.apache.org
--
Thanks,
Brent Worden
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
For additional commands, e-mail: dev-h...@commons.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
For additional commands, e-mail: dev-h...@commons.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
For additional commands, e-mail: dev-h...@commons.apache.org
