On Friday, April 27, 2012 at 3:54 AM, Christophe Porteneuve wrote:
> This is on a legacy project: Rails 2.3.14, rspec 1.3.2, rspec-rails 1.3.4 :-/
>
> I've got a before(:each) setting up AR instances, so that I end up
> with a @product assign.
>
> Then within a context, one of my "it" descriptions goes:
>
> money = Money.new(5_00, 'EUR')
> fd = double('FlashSaleDiscount', :sale_total_cents =>
> money.cents)
> Rails.logger.debug '1>' * 70
>
> @product.should_receive(:active_flash_sale_discount).and_return(fd)
> @product.min_money(:force).should == money
> Rails.logger.debug '<1' * 70
>
> But when I run this, @product's *actual* active_flash_sale_discount
> method is called, not the stubbed version used by the should_receive
> expectation.
>
>
There is a guideline that you should not stub or set message expectations on
the object you're testing. Doing so changes the runtime structure of the object
in a way that never happens in the running system so it's quite possible to end
up with false results. Not saying this is the source of your issue. Just saying
:)
Usually when you set an expectation on an ActiveRecord model and the real
message gets called it's because there is a finder involved that is building a
new instance of the same data e.g.
describe Thing do
before do
@thing_helper = ThingHelper.create!
@thing = Thing.create!(:thing_helper_id => @thing_helper.id)
end
it "delegates to the helper" do
@helper.should_receive(:help_me)
@thing.do_something_that_needs_help
end
end
class Thing < ActiveRecord::Base
belongs_to :thing_helper
def do_something_that_needs_help
thing_helper.help_me
end
end
Here the do_something_that_needs_help loads a new copy of the ThingHelper with
the same data as the @thing_helper set up in the before block, and sends
`help_me` to that object instead.
I'd need to see the code the in the before(:each) block to be sure, but that is
one place to look.
> Another thing that confuses me is that the before(:each) AR-creating
> code is, according to logs, triggered *after* the first
> Rails.logger.debug call. Is there some lazy evaluation stuff I'm not
> aware of in there? As for the final call, it doesn't show up in my
> logs since the should fails.
>
> This test code used to work. I have no idea why it suddenly went dead.
> I thought this might be due to my running it through spork and guard,
> but reverting back to a bare-bones "spec spec/models/product_spec.rb"
> CLI had the exact same results.
>
>
Did anything change since the code did work? Diff version of Ruby? Upgraded any
gems? Etc, etc.
David
--
You received this message because you are subscribed to the Google Groups
"rspec" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/rspec?hl=en.
