[TIP] Need help with IsolationPlugin and nose 0.10 API

[Brandon Corfman]

Yes, you've got it right -- I'm trying to re-run the test suite within
kNose after the code has changed. 

@#$%, I was really hoping I wouldn't have to run subprocess, since
that's what I'm doing now, and it's kind of brittle. (It's also billed
as being cross-platform, but it isn't really.) 

My initial testing with nose.run() using the isolation plugin is
noticeably faster for me than using subprocess. That's one reason I was
happy to find it, but I guess it isn't completely reliable yet.

If I don't hear from anyone else, then I'll probably take your
suggestion about submitting a feature request and try to work around the
problem in the meantime.

Thanks for your helpful suggestions.

Best regards,
Brandon 

Kumar McMillan wrote:
>Hi Brandon.
>from your Active State comment
>http://community.activestate.com/impatient-developers-guide-writing-python-nose-plugins#comment-4864
>it looks like what you are actually trying to do is re-run your suite
>after the code has changed, correct?

>The isolation plugin is kind of overkill for this because it uses a
>separate process for *each* test, thus slowing down the test suite
>considerably.  Another approach might be to drive a nosetests
>subprocess that calls nose.run() and exits [1].  This would mean you'd
>have to scrape the output but you'd probably have to do that anyway
>unless you are introspecting the unittest TestResult (which would be
>painful).  Alternatively, you could submit a feature request asking
>for a way to "clear" all test suites from memory:
>http://code.google.com/p/python-nose/issues/list
.>..this shouldn't be too hard (famous last words:)).