maas-devel team mailing list archive

[John Arbash Meinel <john@xxxxxxxxxxxxxxxxx>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 9/12/2012 3:49 PM, Gavin Panella wrote:
> On 11 September 2012 15:34, Kapil Thangavelu 
> <kapil.thangavelu@xxxxxxxxxxxxx> wrote:
>> fwiw.. i had put together a nose plugin for integrating mongo
>> into the nose test runner. 
>> http://pypi.python.org/pypi/mongonose/0.5.0
> 
> Neat.
> 
> Fwiw, we're using nose, and its okay, but I'd like to avoid
> further entrenchment of its position. We're using django_nose,
> which is a hacky abomination afaict, but I'm not sure that nose can
> run Django tests without it. I don't want additional obstructions
> in the way of an exit from django_nose.

- From my experience, of all the test runners I've come across, nose
seems to be the furthest from the standard 'unittest' infrastructure.
So getting unittest stuff (like testscenarios, etc) working with nose
is quite difficult.

> 
> Generally, tests in Django seem to rely heavily on the Django test 
> runner, and I think this is a bit of an anti-pattern. I'd like to 
> avoid anything that relies on the test runner beyond the basic 
> contracts laid down by unittest.
> 
> I guess this code could be readily re-purposed into a Fixture or 
> TestResource, neither of which require support from the test
> runner.
> 

+1 from me.

John
=:->

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (Cygwin)
Comment: Using GnuPG with Mozilla - http://www.enigmail.net/

iEYEARECAAYFAlBQecAACgkQJdeBCYSNAAPRRgCgkoHRDOtaPVnWgF9JzXJ0MF7d
5UQAoMWp0pY9oTlNO444eIhMbQWZVfgS
=47bu
-----END PGP SIGNATURE-----