Hi,
That's interesting info, thanks for sharing it. I would like to track
down the root cause of this, it would be much appreciated if either of
you or anyone else could provide a sample project where this is
reproducible. It might boil down to differences between the console test
runner and the test runner that the IDE uses.
Thanks,
Erno
Fjan wrote:
> Hi, I had the same problem. I think there are a few methods getting added
> somewhere that trip things up. In my case, the start method got overwritten
> by Netbeans, and this is what fixed it for me:
>
> class Slot < ActiveRecord::Base
> def start;self[:start];end
> end
>
> I'm not at all sure if this is something particular to my situation though.
>
>
> Chris Nelson-5 wrote:
>
>>> Chris Nelson wrote:
>>>
>>> I have the same problem on the new builds. I was finally able to
>>> narrow down what the problem is, and hack in a fix, but don't know
>>> *why* it happened. It appears to be some freaky combination of my
>>> particular project and NB. The issue is that a nil gets passed down
>>> into the Fixtures.create_fixtures method for the table_names param,
>>> but only when I run them in NB. When I run on command line it gets
>>> []. I fixed by hacking Fixtures to check for nil and set it to [].
>>> But I have no idea how this param gets different values based on
>>> whether I run from NB or not. Also, when created a clean project with
>>> a simple AR test it didn't exhibit this behaviour, so I guess it is
>>> some weird interaction between NB and my specific project.
>>>
>>> --Chris
>>>
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail:
users-unsubscribe@...
For additional commands, e-mail:
users-help@...
