Rspec should have_many raises wrong number of arguments (0 for 1)

Question

When I run a file of rspec tests on my Rails application, everything passes except the tests for has_many associations:

it { should have_many(:notes) } # ERROR
it { should belong_to(:agent_field_set) } # OK
it { should have_and_belong_to_many(:profiles) } # OK

The should have_many(:notes) line raises ArgumentError: wrong number of arguments (0 for 1). I've checked my database and my model file to verify the existance of a has_many association, and both appear fine:

has_many :notes, class_name: "NoteNs::Note", foreign_key: :user_id

Can anyone tell me where to look for the problem?

Specs: Ruby 1.9.3, Rails 3.2.11

The subject is a class named Usage::Users. (But how does that make a difference?) No, I'm not using the shoulda_helpers. — JellicleCat, Feb 12 '13 at 19:30
Oops. It appears that the shoulda gem is included. (I'm developing as part of a team, so I was unaware.) What does that indicate? — JellicleCat, Feb 12 '13 at 19:37
You can have a look at this: http://stackoverflow.com/questions/9123218/rails-rspec-writing-spec-for-class-name-of-belongs-to-association — siekfried, Feb 12 '13 at 20:41
Well, thanks. A bit of a dead-end since I'm trying to run a test that was written by a teammate. — JellicleCat, Feb 12 '13 at 21:11

score 0 · Answer 1 · answered Feb 12 '13 at 23:48

At long last, what fixed the problem was renaming one of my named scopes. My named scope was called parent, and my model has a belongs_to :parent association.

I can't guess why a conflict arises between a named scope, being on the class, and an association, being on an instance. +1 to anyone who can explain that one to me.

Rspec should have_many raises wrong number of arguments (0 for 1)

1 Answers1