rspec users - Apr 2010 - update_attributes fails in rake spec but not in script/spec???

Hi,

So I just wrote a spec, and tested it by running script/spec
models/result_list_spec.rb

I get:
-----------------------------
.....

Finished in 28.558841 seconds

5 examples, 0 failures

-----------------------------
So then I did:  rake spec

and I get a bunch of errors on that same file "wrong number of arguments (0
for 3)"

I then threw a debugger into the code to try to see where it was failing, and
it was failing when I was doing self.update_attributes in the model...

I tried assigning this myself in the debugger console:

(rdb:1) self.update_attribute(:job_item_count, 4)
ArgumentError Exception: wrong number of arguments (0 for 3)

(rdb:1) self
#<ResultList id: 1, table_id: nil, newsletter_id: 3, type: nil, job_start:
"2010-04-30 10:25:17", job_finish: "2010-04-30 10:25:17",
job_item_count: 0,
job_action_count: -5, job_error: nil, attachment_file_name: nil,
attachment_content_type: nil, attachment_file_size: nil, attachment_updated_at:
nil, created_at: "2010-04-30 18:01:41", updated_at: "2010-04-30
18:01:41">

... ????????????

Can anyone please explain to me why this is behaving this way?



Patrick J. Collins
http://collinatorstudios.com

Well, after some investigating, I discovered than another one of my spec files
was causing this behavior.  When removing this other spec file, the one that
was failing via rake spec no longer failed..  Totally bizarre.

This is the spec that fails:
http://gist.github.com/386129

And this is the spec that when I delete, the other one that failed now passes:
http://gist.github.com/386125

This is so weird, I would really love it if someone could shed some light on
this.

Patrick J. Collins
http://collinatorstudios.com

On May 1, 2010, at 2:29 AM, Patrick J. Collins wrote:
> Well, after some investigating, I discovered than another one of my spec
files
> was causing this behavior.  When removing this other spec file, the one
that
> was failing via rake spec no longer failed..  Totally bizarre.
> 
> This is the spec that fails:
> http://gist.github.com/386129
> 
> And this is the spec that when I delete, the other one that failed now
passes:
> http://gist.github.com/386125
> 
> This is so weird, I would really love it if someone could shed some light
on
> this.
Hey Patrick,

This is just a guess, but on line 3 of http://gist.github.com/386125,
ActionView::Helpers is being mixed in to the main object, which means
it''s methods are made accessible to every single object in the system.
It''s very likely that it is overriding methods that are deeper down the
stack. Try moving that include statement inside the ExampleGroup (i.e. inside
the describe block).

If that doesn''t work, please run the specs with the -b flag, which
produces a full backtrace, and then gist the output for us.

HTH,
David
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://rubyforge.org/pipermail/rspec-users/attachments/20100501/a7bbc1bb/attachment.html>

> ActionView::Helpers is being mixed in to the main object, which means
it''s
> methods are made accessible to every single object in the system.
It''s very
> likely that it is overriding methods
Hi David,

Ahh.. Thank you very much, that was indeed the problem.  rake spec now passes
all test.

Patrick J. Collins
http://collinatorstudios.com