shoulda Google Group

Re: [shoulda] Re: Does shoulda support ".should ==" syntax

2010-01-03T15:49:34Z

Use test/spec for that nice syntax:
[link]

Re: Does shoulda support ".should ==" syntax

2010-01-01T10:20:54Z

Yes, you're right. It is in rspec and not shoulda. Shame cos it's a
nice way to write tests.
Thanks
Nick

Re: [shoulda] Does shoulda support ".should ==" syntax

2009-12-31T17:09:26Z

I believe that's rspec, not shoulda.

Does shoulda support ".should ==" syntax

2009-12-31T16:59:58Z

Does shoulda support a syntax like this:
should "return weekly hours" do
hours = @project.recent_weekly_hours
hours.size.should == 52
end
I was expecting it to - and thought I had seen it somewhere - but I
get "undefined method 'should' "
thanks
nick

Re: [shoulda] macros don't work: undefined method `should_assign_to' for FoodEntryControllerTest::Subclass_1:Class (NoMethodError)

2009-12-24T14:15:06Z

I'm pretty sure that Shoulda won't load the rails macros if Spec is defined. Check out rails/init.rb:
if RAILS_ENV == 'test'
if defined? Spec
require 'shoulda/rspec'
else
require 'shoulda/rails'
end
end
so it looks like it's one or the other.
Best.
Mike

macros don't work: undefined method `should_assign_to' for FoodEntryControllerTest::Subclass_1:Class (NoMethodError)

2009-12-24T07:38:20Z

ruby 1.9.1p243, rails 2.3.5, thoughtbot-shoulda 2.10.2.
my spec, using macros:
class FoodEntryControllerTest < ActionController::TestCase
context "on GET new" do
setup do
@food_entry = FoodEntry.new
get :new
end
should_assign_to :food_entry
end
end
the error:
/home/oren/weighmyplate/spec/c ontrollers/food_entry_controll er_spec.rb:

Re: Cleaning db between tests

2009-12-22T18:30:08Z

I found another user with the same problem. My db had MyISAM tables,
so there was no transaction support to roll back.
[link]
Short answer: problem solved.

Cleaning db between tests

2009-12-22T15:18:25Z

Sorry if this is a noob question. A day of Google searching hasn't
turned up an answer.
I just started working on a project that has a test suite built with
shoulda. I'm new to shoulda, and I'm seeing a number of failures in
the test suite. I've tracked the errors down to records in the db
persisting between tests. If I put in explicit calls to clear out the

Re: stub helper for avoid twitter request

2009-11-29T01:02:53Z

Very thanks Michael, Fakeweb saves the day!

Re: [shoulda] stub helper for avoid twitter request

2009-11-28T14:05:59Z

You may want to think about stubbing the HTTP request to Twitter using webmock[1] or fakeweb[2] instead of stubbing your helper.

Best,
Mike

[1] [link]
[2] [link]

stub helper for avoid twitter request

2009-11-28T01:49:27Z

Hi folks, I'm using the twitter gem for get my last tweets and I've a
recent_tweets helper in my application_helper, this helper is used in
layout level, then I want to stub this helper and avoid the twitter
request because my tests are very slow, how to I can stub this helper
for all my functional tests?

Re: Validate uniqueness depends on record

2009-11-24T07:30:42Z

Maybe there's a better way to do it? What I'm trying to test is this:
validates_uniqueness_of :url, :unless => :local?
validates_uniqueness_of :url, :if => :local?, :scope => :user_id

Re: Validate uniqueness depends on record

2009-11-23T06:42:31Z

Sorry if I was unclear. There's no error, but the test fails with
this:
Expected errors to include "has already been taken" when url is set to
"[link]", got no errors

Your suggestion didn't work.

On Nov 23, 3:58 am, François Beausoleil

Re: [shoulda] Validate uniqueness depends on record

2009-11-23T03:58:00Z

You don't show any error messages, so I'm not too sure. I believe #subject is a class method, so you'd do it this way:

context "a global request" do
subject do
@subject ||= Factory(:request, :local => false)
end

should_validate_uniqueness_of :url
end

RDoc for #subject: [link]

Validate uniqueness depends on record

2009-11-22T16:42:17Z

Hi!

I have this Request model that I want to test a little bit different
depending on an attribute. The attribute is *local*. So if a Request
is local, I don't want to validate uniqueness of *url*. But I do want
that if the Request is global.

So I thought I would do it like this:
class RequestTest < ActiveSupport::TestCase