When to switch from cucumber to rspec in the BDD cycle for a login procedure

Hope that helps I've figured out a way, if you place this in your features/support/env.rb:

ActionController::Base.class_eval do
  private

  def begin_open_id_authentication(identity_url, options = {})
    yield OpenIdAuthentication::Result.new(:successful), identity_url, nil
  end 
end

Given /^I am logged in as "(.*)"$/ do |name|
  user = User.find_by_name(user)
  post '/session', :openid_url => user.identity_url
  # Some assertions just to make sure our hack in env.rb is still working
  response.should redirect_to('/')
  flash[:notice].should eql('Logged in successfully')
end

To fix this issue If you haven't already, you might want to check out Dan North's excellent article, What's in a Story? as a starting point.
We have two main uses for Cucumber stories. First, because the story form is very specific it helps focus the product owner's articulation of the features he wants built. This is the "token for a conversation" use of stories, and would be valuable whether or not we implemented the stories in code. Second, when the process is working well enough that we have complete stories before we begin writing the feature (more of an ideal that we strive for than a daily reality), you have your acceptance criteria spelled out clearly and you know exactly what and how much to build.

fixed the issue. Will look into that further I'm impressed at the brevity and usefulness of Cucumber's Scenarios, they're a great way to test a load of different cases. , Try the following approach. I like the way it turned out.

describe StateDateMethods do
  before :each do
    @product = OpenStruct.new
    @product.extend StateDateMethods
  end

  def parse_date(unparsed_date_value)
    unless unparsed_date_value.nil?
      DateTime.strptime(unparsed_date_value, '%m/%d/%Y')
    end
  end

  context '#pre_order?' do
    examples = [
      # [visible_on, pre_order_on, for_sale_on] => method_result
      { :inputs => [nil, nil, nil], :expected => false },
      { :inputs => ['1/1/2001', nil, nil], :expected => false },
      { :inputs => ['1/1/2001', '1/1/2001', nil], :expected => true },
      { :inputs => ['1/1/2001', '1/2/2001', nil], :expected => true },
      { :inputs => ['1/1/2001', '1/1/2001', '1/2/2001'], :expected => false },
      { :inputs => ['1/1/2001', '1/1/2001', '1/1/3001'], :expected => true },
      { :inputs => ['1/1/2001', '1/1/3001', '1/2/3001'], :expected => false },
      { :inputs => ['1/1/3001', '1/1/3001', '1/2/3001'], :expected => false },
      { :inputs => ['1/1/2001', nil, '1/1/2001'], :expected => false },
      { :inputs => ['1/1/2001', nil, '1/1/3001'], :expected => false }
    ]
    examples.each do |example|
      inputs = example[:inputs]

      it "should return #{example[:expected].inspect} when visible_on == #{inputs[0].inspect}, pre_order_on == #{inputs[1].inspect}, for_sale_on == #{inputs[2].inspect}" do
        @product.visible_on = parse_date(inputs[0])
        @product.pre_order_on = parse_date(inputs[1])
        @product.for_sale_on = parse_date(inputs[2])

        @product.pre_order?.should == example[:expected]
      end
    end
  end
end

....F.....

Failures:

  1) StateDateMethods#pre_order? should return false when visible_on == "1/1/2001", pre_order_on == "1/1/2001", for_sale_on == "1/2/2001"
     Failure/Error: @product.pre_order?.should == example[:expected]
       expected: false
            got: true (using ==)
     # ./spec_no_rails/state_date_methods_spec.rb:40:in `block (4 levels) in <top (required)>'

Finished in 0.38933 seconds
10 examples, 1 failure

Failed examples:

rspec ./spec_no_rails/state_date_methods_spec.rb:35 # StateDateMethods#pre_order? should return false when visible_on == "1/1/2001", pre_order_on == "1/1/2001", for_sale_on == "1/2/2001"

..........

Finished in 0.3889 seconds
10 examples, 0 failures

should help you out I ran into this same problem. After hunting down all binary rspec commands (several locations), I belatedly realized that the 1.3.2 branch of RSPEC titles its binary "spec" and not "rspec".
So the correct commands was always "bundle exec spec spec/models/..." Putting "rspec" meant that bundler failed to find it in my current rvm/gemset and went searching through my entire path, eventually coming up with an /opt/local/bin copy that belonged to an rspec2.x install.

around this issue You must disable Cucumber in Spork in the RSpec group and the other way around:

specs :specs    
  guard :spork, :cucumber => false do
    # ...
  end
end

specs :features    
  guard 'spork', :rspec => false do
    # ...
  end
end