why is rspec telling me that this user is valid - ruby-on-rails

Question

I have the following classes in Rails and am writing some rspec tests (any critiques are more than welcome as I'm a nOOb at rspec).
class User.rb
class User < ActiveRecord::Base
email_regex = /\A[\w+\-.]+#[a-z\d\-.]+\.[a-z]+\z/i
validates :email, :presence => true ,
:format => { :with => email_regex },
:uniqueness => { :case_sensitive => true },
:on => :create
end
and in factories.rb
FactoryGirl.define do
factory :user do
sequence(:name) { |n| "my-name#{n}" }
sequence(:email) { |n| "blue#{n}#12blue.com" }
end
end
and in my rspec (users_spec.rb):
require 'spec_helper'
describe User do
let(:user) { FactoryGirl.build(:user) }
it { user.should be_valid }
it { user.should be_a(User) }
it { user.should respond_to(:email) }
it { user.email = " " }
it { user.should_not be_valid } # this is causing the error
end
and get
1) User
Failure/Error: it { user.should_not be_valid }
expected valid? to return false, got true
But based upon the validates, user should be not be valid. What is going on here? What am I not getting (and I know it's my fault)?
thx

Answer 1

I assume that the test failure surprises you because you think the user email should be " ".
In rspec every example is independent. This means that anything you did in a previous example is forgotten.
In your case your second to last example runs, builds a new, valid activerecord user whose email is "blue4#12blue.com", overwrites that email with " " and then passes since it makes no assertions.
Then your last example runs, builds a new, valid activerecord user who's email is "blue5#12blue.com" and fails because the user is valid, it's email has not been overwritten.
You probably want something like this:
it 'should validate the email' do
user.email = " "
user.should_not be_valid
end