[prev in list] [next in list] [prev in thread] [next in thread] 

List:       jakarta-commons-dev
Subject:    Re: [logging] LoadTest
From:       Simon Kitching <skitching () apache ! org>
Date:       2005-05-31 6:11:32
Message-ID: 1117519892.3792.42.camel () localhost ! localdomain
[Download RAW message or body]

On Tue, 2005-05-31 at 17:59 +1200, Simon Kitching wrote:
> The test doesn't set up its own clean classpath?! I thought that was
> what the AppClassLoader private class was all about - though of course
> it's hard to tell as there aren't any comments on it.
> 
> If the unit test doesn't properly set up a clean classpath before doing
> classpath-related unit tests, then that really sucks.
> 
> Ah well, I don't think any of us have enough enthusiasm to fix the unit
> tests as well as the code. So however you get the test to run is ok by
> me.

Hmm. I just tried your patch to include LoadTest in TestAll. And the
test LoadTest fails for me with some problem in the "reload" method.
That's without having applied any of your patches, ie the current HEAD
version of the test fails against the current HEAD version of JCL.

I suspect the reason the test isn't in the standard suite is that it has
never actually worked properly at all. The custom ClassLoader user
probably doesn't properly isolate the test from the surrounding
environment meaning that the test will work for some people but not
others (eg you, but not me).

The subversion history shows that while license text etc. have been
updated in it, no-one has actually committed a *meaningful* change to it
since it was created at 2003-03-01.

I might try to find time to have a look at this, but as the test was
never part of the standard suite, I'm now generally in favour of just
ignoring it.

Regards,

Simon


---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org

[prev in list] [next in list] [prev in thread] [next in thread]