index.html

---
title: 'Better Specs'
---
<% content_for :menu do %>
<li><a href="/">Introduction to rspec guidelines</a></li>
<li class="little">&raquo; <a href="#describe">How to describe your methods</a></li>
<li class="little">&raquo; <a href="#contexts">Use contexts</a></li>
<li class="little">&raquo; <a href="#short">Keep your description short</a></li>
<li class="little">&raquo; <a href="#single">Single expectation</a></li>
<li class="little">&raquo; <a href="#all">Test all possible cases</a></li>
<li class="little">&raquo; <a href="#expect">Expect vs Should syntax</a></li>
<li class="little">&raquo; <a href="#subject">Use subject</a></li>
<li class="little">&raquo; <a href="#let">Use let and let!</a></li>
<li class="little">&raquo; <a href="#mock">Mock or not to mock</a></li>
<li class="little">&raquo; <a href="#data">Create only the data you need</a></li>
<li class="little">&raquo; <a href="#factories">Use factories and not fixtures</a></li>
<li class="little">&raquo; <a href="#matchers">Easy to use matchers</a></li>
<li class="little">&raquo; <a href="#sharedexamples">Shared examples</a></li>
<li class="little">&raquo; <a href="#integration">Test what you see</a></li>
<li class="little">&raquo; <a href="#should">Do not use should</a></li>
<li class="little">&raquo; <a href="#guard">Automatic tests with guard</a></li>
<li class="little">&raquo; <a href="#spork">Faster tests with spork</a></li>
<li class="little">&raquo; <a href="#http">Stubbing HTTP requests</a></li>
<li class="little">&raquo; <a href="#formatter">Useful formatter</a></li>
<li><a href="#books">Books</a></li>
<li><a href="#presentations">Presentations</a></li>
<li><a href="#resources">Resources on the web</a></li>
<li><a href="#screencasts">Screencasts</a></li>
<li><a href="#libraries">Libraries</a></li>
<li><a href="#styleguide">Styleguide</a></li>
<li><a href="#improving">Improving Better Specs</a></li>
<li><a href="#credits">Credits</a></li>
<% end %>

<div class="content">
<article>

<aside class="menu">

<h1>Affiliate books</h1>
<%= render "partials/books" %>

</aside>

</article>

<article>

<iframe src="http://ghbtns.com/github-btn.html?user=andreareginato&repo=betterspecs&type=watch&count=true&size=large"
  allowtransparency="true" frameborder="0" scrolling="0" width="170px" height="30px"></iframe>

<iframe src="http://ghbtns.com/github-btn.html?user=andreareginato&repo=betterspecs&type=fork&count=true&size=large"
  allowtransparency="true" frameborder="0" scrolling="0" width="170px" height="30px"></iframe>

<iframe src="http://ghbtns.com/github-btn.html?user=andreareginato&repo=betterspecs&type=follow&count=true&size=large"
  allowtransparency="true" frameborder="0" scrolling="0" width="300px" height="30px"></iframe>

</article>

</br>

<article>

<p>
<a href="https://www.relishapp.com/rspec/">RSpec</a> is a great tool in the behavior-driven development (BDD) process of writing human readable specifications that direct and validate
the development of your application.
</p>

<p>
On the web there are many resources that give complete overview of _what_ you can do with RSpec.
But there are fewer resources devoted to how to create a great RSpec test suite.
</p>

<p>
Better Specs tries to fill this gap by collecting most of the "best practices" that other
developers learn through years of experience.
</p>

</article>

<article>
  
<h1><a name="thanks">Thanks to Open Source</a></h1>

<p>
Betterspecs.org come to life from the experience I've made while working on Lelylan, an open source microservices 
architecture for the Internet of Things. If this project helped you in any way, think about giving us a 
<a href="https://github.com/lelylan/lelylan">Star on Github</a> or helping on 
<a href="https://github.com/lelylan/lelylan/blob/master/ROADMAP.md">improving the project</a>.
</p>

</p>
In Lelylan you can also find many of the tests who gave inspiration to these guidelines. They need a full
refactoring (to latest updates and Rails 5.0), but they could be of inspiration.
</p>

<a href="https://github.com/lelylan/lelylan">
<img src="https://raw.githubusercontent.com/lelylan/lelylan/master/public/logo-lelylan.png" data-canonical-src="https://raw.githubusercontent.com/lelylan/lelylan/master/public/logo-lelylan.png" width="300"/></a>


</article>

<article>

<h1><a name="describe">How to describe your methods</a></h1>

<p>
Be clear about what method you are describing. For instance, use the Ruby documentation convention of
<code>.</code> (or <code>::</code>) when referring to a class method's name and <code>#</code> when
referring to an instance method's name.
</p>

<p class="wrong">bad</p>

<div>
<pre><code class="ruby">describe 'the authenticate method for User' do
describe 'if the user is an admin' do
</code></pre>
</div>

<p class="correct">good</p>

<div>
<pre><code class="ruby">describe '.authenticate' do
describe '#admin?' do
</code></pre>
</div>

<p>
<a href="https://github.com/andreareginato/betterspecs/issues/2">Discuss this guideline &rarr;</a>
</p>

<%= render "partials/share", text: "betterspecs.org | how to describe your methods", url: "http://betterspecs.org/#describe" %>

</article>

<article>

<h1><a name="contexts">Use contexts</a></h1>

<p>
Contexts are a powerful method to make your tests clear and well organized.
In the long term this practice will keep tests easy to read.
</p>

<p class="wrong">bad</p>

<div>
<pre><code class="ruby">it 'has 200 status code if logged in' do
  expect(response).to respond_with 200
end
it 'has 401 status code if not logged in' do
  expect(response).to respond_with 401
end
</code></pre>
</div>

<p class="correct">good</p>

<div>
<pre><code class="ruby">context 'when logged in' do
  it { is_expected.to respond_with 200 }
end
context 'when logged out' do
  it { is_expected.to respond_with 401 }
end
</code></pre>
</div>

<p>
  When describing a context, start its description with "when", "with" or "without".
</p>

<p>
<a href="https://github.com/andreareginato/betterspecs/issues/3">Discuss this guideline &rarr;</a>
</p>

<%= render "partials/share", text: "betterspecs.org | use contexts", url: "http://betterspecs.org/#contexts" %>

</article>

<article>

<h1><a name="short">Keep your description short</a></h1>

<p>
A spec description should never be longer than 40 characters. If this happens
you should split it using a context.
</p>

<p class="wrong">bad</p>

<div>
<pre><code class="ruby">it 'has 422 status code if an unexpected params will be added' do
</code></pre>
</div>

<p class="correct">good</p>

<div>
<pre><code class="ruby">context 'when not valid' do
  it { is_expected.to respond_with 422 }
end
</code></pre>
</div>

<p>
In the example we removed the description related to the status code,
which has been replaced by the expectation <code>it { is_expected.to respond_with 422 }</code>.
If you run this test typing <code>rspec filename</code> you will obtain a readable output.
</p>

<p class="base">Formatted Output</p>

<div>
<pre><code>when not valid
  it should respond with 422
</code></pre>
</div>

<p>
<a href="https://github.com/andreareginato/betterspecs/issues/4">Discuss this guideline &rarr;</a>
</p>

<%= render "partials/share", text: "betterspecs.org | keep your description short", url: "http://betterspecs.org/#short" %>

</article>

<article>

<h1><a name="single">Single expectation test</a></h1>

<p>
The 'one expectation' tip is more broadly expressed as 'each test should make only one assertion'.
This helps you on finding possible errors, going directly to the failing test, and to make your
code readable.
</p>

<p>
In isolated unit specs, you want each example to specify one (and only one) behavior. Multiple
expectations in the same example are a signal that you may be specifying multiple behaviors.
</p>

<p>
Anyway, in tests that are not isolated (e.g. ones that integrate with a DB, an external webservice,
or end-to-end-tests), you take a massive performance hit to do the same setup over and over
again, just to set a different expectation in each test. In these sorts of slower tests, I
think it's fine to specify more than one isolated behavior.
</p>

<p class="correct">good (isolated)</p>

<div>
<pre><code class="ruby">it { is_expected.to respond_with_content_type(:json) }
it { is_expected.to assign_to(:resource) }
</code></pre>
</div>

<p>

<p class="correct">Good (not isolated)</p>

<div>
<pre><code class="ruby">it 'creates a resource' do
  expect(response).to respond_with_content_type(:json)
  expect(response).to assign_to(:resource)
end
</code></pre>
</div>

<a href="https://github.com/andreareginato/betterspecs/issues/5">Discuss this guideline &rarr;</a>
</p>

<%= render "partials/share", text: "betterspecs.org | single expectation test", url: "http://betterspecs.org/#single" %>

</article>

<article>

<h1><a name="all">Test all possible cases</a></h1>

<p>
Testing is a good practice, but if you do not test the edge cases, it will not
be useful. Test valid, edge and invalid case. For example, consider the following action.
</p>

<p class="base">Destroy action</p>

<div>
<pre><code class="ruby">before_filter :find_owned_resources
before_filter :find_resource

def destroy
  render 'show'
  @consumption.destroy
end</code></pre>
</div>

<p>
The error I usually see lies in testing only whether the resource has been removed.
But there are at least two edge cases: when the resource is not found and when it's not
owned. As a rule of thumb think of all the possible inputs and test them.
</p>

<p class="wrong">bad</p>

<div>
<pre><code class="ruby">it 'shows the resource'
</code></pre>
</div>

<p class="correct">good</p>

<div>
<div class="correct">
<pre><code class="ruby">describe '#destroy' do

  context 'when resource is found' do
    it 'responds with 200'
    it 'shows the resource'
  end

  context 'when resource is not found' do
    it 'responds with 404'
  end

  context 'when resource is not owned' do
    it 'responds with 404'
  end
end
</code></pre>
</div>

<p>
<a href="https://github.com/andreareginato/betterspecs/issues/6">Discuss this guideline &rarr;</a>
</p>

<%= render "partials/share", text: "betterspecs.org | test all possible cases", url: "http://betterspecs.org/#all" %>

</article>

<article>

<h1><a name="expect">Expect vs Should syntax</a></h1>

<p>
On new projects always use the <code>expect</code> syntax.
</p>

<p class="wrong">bad</p>

<div>
<pre><code class="ruby">it 'creates a resource' do
  response.should respond_with_content_type(:json)
end</code></pre>
</div>

<p class="correct">good</p>

<div>
<pre><code class="ruby">it 'creates a resource' do
  expect(response).to respond_with_content_type(:json)
end</code></pre>
</div>

<p>Configure the RSpec to only accept the new syntax on new projects, to avoid having the 2 syntax all over the place.</p>

<p class="correct">good</p>

<div>
<pre><code class="ruby"># spec_helper.rb
RSpec.configure do |config|
  # ...
  config.expect_with :rspec do |c|
    c.syntax = :expect
  end
end</code></pre>
</div>

<p>On one line expectations or with implicit subject we should use <code>is_expected.to</code>.</p>

<p class="wrong">bad</p>

<div>
<pre><code class="ruby">context 'when not valid' do
  it { should respond_with 422 }
end
</code></pre>
</div>

<p class="correct">good</p>

<div>
<pre><code class="ruby">context 'when not valid' do
  it { is_expected.to respond_with 422 }
end
</code></pre>
</div>

<p>On old projects you can use the <a href="https://github.com/yujinakayama/transpec">transpec</a> to convert them to the new syntax.</p>

<p>More about the new RSpec expectation syntax can be found <a href="http://myronmars.to/n/dev-blog/2012/06/rspecs-new-expectation-syntax">here</a> and <a href="http://myronmars.to/n/dev-blog/2013/07/the-plan-for-rspec-3#what_about_the_old_expectationmock_syntax">here</a>.</p>

<p>
<a href="https://github.com/andreareginato/betterspecs/issues/83">Discuss this guideline &rarr;</a>
</p>

<%= render "partials/share", text: "betterspecs.org | Expect vs Should syntax", url: "http://betterspecs.org/#expect" %>

</article>

<article>

<h1><a name="subject">Use subject</a></h1>

<p>
If you have several tests related to the same subject use
<code>subject{}</code> to DRY them up.
</p>

<p class="wrong">bad</p>

<div>
<pre><code class="ruby">it { expect(assigns('message')).to match /it was born in Belville/ }
</code></pre>
</div>

<p class="correct">good</p>

<div>
<pre><code class="ruby">subject { assigns('message') }
it { is_expected.to match /it was born in Billville/ }
</code></pre>
</div>

<p>
RSpec has also the ability to use a named subject.
</p>

<p class="correct">Good</p>

<div>
<pre><code class="ruby">subject(:hero) { Hero.first }
it "carries a sword" do
  expect(hero.equipment).to include "sword"
end
</code></pre>
</div>

<p>
Learn more about <a href="https://www.relishapp.com/rspec/rspec-core/v/2-11/docs/subject">rspec subject</a>.
</p>

<p>
<a href="https://github.com/andreareginato/betterspecs/issues/7">Discuss this guideline &rarr;</a>
</p>

<%= render "partials/share", text: "betterspecs.org | use subject", url: "http://betterspecs.org/#subject" %>

</article>

<article>

<h1><a name="let">Use let and let!</a></h1>

<p>
When you have to assign a variable instead of using a <code>before</code> block to create
an instance variable, use <code>let</code>. Using <code>let</code> the variable lazy loads
only when it is used the first time in the test and get cached until that specific test is
finished. A really good and deep description of what <code>let</code> does can be found in this
<a href="http://stackoverflow.com/questions/5359558/when-to-use-rspec-let/5359979#5359979">stackoverflow answer</a>.
</p>

<p class="wrong">bad</p>

<div>
<pre><code class="ruby">describe '#type_id' do
  before { @resource = FactoryGirl.create :device }
  before { @type     = Type.find @resource.type_id }

  it 'sets the type_id field' do
    expect(@resource.type_id).to equal(@type.id)
  end
end
</code></pre>
</div>

<p class="correct">good</p>

<div>
<pre><code class="ruby">describe '#type_id' do
  let(:resource) { FactoryGirl.create :device }
  let(:type)     { Type.find resource.type_id }

  it 'sets the type_id field' do
    expect(resource.type_id).to equal(type.id)
  end
end
</code></pre>
</div>

<p>
Use <code>let</code> to initialize actions that are lazy loaded to test your specs.
</p>

<p class="correct">good</p>

<div>
<pre><code class="ruby">context 'when updates a not existing property value' do
  let(:properties) { { id: Settings.resource_id, value: 'on'} }

  def update
    resource.properties = properties
  end

  it 'raises a not found error' do
    expect { update }.to raise_error Mongoid::Errors::DocumentNotFound
  end
end
</code></pre>
</div>

<p>
Use <code>let!</code> if you want to define the variable when the block is defined.
This can be useful to populate your database to test queries or scopes.
</p>

<p>
Here an example of what let actually is.
</p>

<p class="correct">good</p>

<div>
<pre><code class="ruby"># this:
let(:foo) { Foo.new }

# is very nearly equivalent to this:
def foo
  @foo ||= Foo.new
end
</code></pre>
</div>

<p>
Learn more about <a href="https://www.relishapp.com/rspec/rspec-core/v/2-11/docs/helper-methods/let-and-let">rspec let</a>.
</p>

<p>
<a href="https://github.com/andreareginato/betterspecs/issues/8">Discuss this guideline &rarr;</a>
</p>

<%= render "partials/share", text: "betterspecs.org | use let and let!", url: "http://betterspecs.org/#let" %>

</article>

<article>

<h1><a name="mock">Mock or not to mock</a></h1>

<p>
There's a debate going on. Do not (over)use mocks and test real behavior when possible.
Testing real cases are useful when updating your application flow.
</p>

<p class="correct">good</p>

<div>
<pre><code class="ruby"># simulate a not found resource
context "when not found" do
  before do
    allow(Resource).to receive(:where).with(created_from: params[:id])
      .and_return(false)
  end
  it { is_expected.to respond_with 404 }
end
</code></pre>
</div>

<p>
Mocking makes your specs faster but they are difficult to use. You need to understand them
well to use them well. Read more <a href="http://myronmars.to/n/dev-blog/2012/06/thoughts-on-mocking">about</a>.
</p>

<p>
<a href="https://github.com/andreareginato/betterspecs/issues/9">Discuss this guideline &rarr;</a>
</p>

<%= render "partials/share", text: "betterspecs.org | mock or not to mock", url: "http://betterspecs.org/#mock" %>

</article>

<article>

<h1><a name="data">Create only the data you need</a></h1>

<p>
If you have ever worked in a medium size project (but also in small ones), test suites
can be heavy to run. To solve this problem, it's important not to load more data than needed.
Also, if you think you need dozens of records, you are probably wrong.
</p>

<p class="correct">good</p>

<div>
<pre><code class="ruby">describe "User" do
  describe ".top" do
    before { FactoryGirl.create_list(:user, 3) }
    it { expect(User.top(2)).to have(2).item }
  end
end
</code></pre>
</div>

<p>
<a href="https://github.com/andreareginato/betterspecs/issues/10">Discuss this guideline &rarr;</a>
</p>

<%= render "partials/share", text: "betterspecs.org | create only tha data you need", url: "http://betterspecs.org/#data" %>

</article>

<article>

<h1><a name="factories">Use factories and not fixtures</a></h1>

<p>
This is an old topic, but it's still good to remember it. Do not use fixtures because
they are difficult to control, use factories instead. Use them to reduce the verbosity
on creating new data.
</p>

<p class="wrong">bad</p>

<div>
<pre><code class="ruby">user = User.create(
  name: 'Genoveffa',
  surname: 'Piccolina',
  city: 'Billyville',
  birth: '17 Agoust 1982',
  active: true
)
</code></pre>
</div>

<p class="correct">good</p>

<div>
<pre><code class="ruby">user = FactoryGirl.create :user
</code></pre>
</div>

<p>
One important note. When talking about unit tests the best practice would be to use neither
fixtures or factories. Put as much of your domain logic in libraries that can be tested without
needing complex, time consuming setup with either factories or fixtures. Read more in
<a href="http://blog.steveklabnik.com/posts/2012-07-14-why-i-don-t-like-factory_girl">this article</a>
</p>

<p>
Learn more about <a href="https://github.com/thoughtbot/factory_girl">Factory Girl</a>.
</p>

<p>
<a href="https://github.com/andreareginato/betterspecs/issues/11">Discuss this guideline &rarr;</a>
</p>

<%= render "partials/share", text: "betterspecs.org | use factories and not fixtures", url: "http://betterspecs.org/#factories" %>

</article>

<article>

<h1><a name="matchers">Easy to read matcher</a></h1>

<p>
Use readable matchers and double check the available <a href="https://www.relishapp.com/rspec/rspec-expectations/docs/built-in-matchers">rspec matchers</a>.
</p>

<p class="wrong">bad</p>

<div>
<pre><code class="ruby">lambda { model.save! }.to raise_error Mongoid::Errors::DocumentNotFound
</code></pre>
</div>

<p class="correct">good</p>

<div>
<pre><code class="ruby">expect { model.save! }.to raise_error Mongoid::Errors::DocumentNotFound
</code></pre>
</div>

<p>
<a href="https://github.com/andreareginato/betterspecs/issues/12">Discuss this guideline &rarr;</a>
</p>

<%= render "partials/share", text: "betterspecs.org | easy to read matcher", url: "http://betterspecs.org/#matchers" %>

</article>

<article>

<h1><a name="sharedexamples">Shared Examples</a></h1>

<p>
Making tests is great and you get more confident day after day. But in the end you will start to see code duplication coming up everywhere.
Use shared examples to DRY your test suite up.
</p>

<p class="wrong">bad</p>

<div>
<pre><code class="ruby">describe 'GET /devices' do
  let!(:resource) { FactoryGirl.create :device, created_from: user.id }
  let(:uri) { '/devices' }

  context 'when shows all resources' do
    let!(:not_owned) { FactoryGirl.create factory }

    it 'shows all owned resources' do
      page.driver.get uri
      expect(page.status_code).to be(200)
      contains_owned_resource resource
      does_not_contain_resource not_owned
    end
  end

  describe '?start=:uri' do
    it 'shows the next page' do
      page.driver.get uri, start: resource.uri
      expect(page.status_code).to be(200)
      contains_resource resources.first
      expect(page).to_not have_content resource.id.to_s
    end
  end
end
</code></pre>
</div>

<p class="correct">good</p>

<div>
<pre><code class="ruby">describe 'GET /devices' do

  let!(:resource) { FactoryGirl.create :device, created_from: user.id }
  let(:uri)       { '/devices' }

  it_behaves_like 'a listable resource'
  it_behaves_like 'a paginable resource'
  it_behaves_like 'a searchable resource'
  it_behaves_like 'a filterable list'
end
</code></pre>
</div>

<p>
In our experience, shared examples are used mainly for controllers. Since models are pretty
different from each other, they (usually) do not share much logic.
</p>

<p>
Learn more about <a href="https://www.relishapp.com/rspec/rspec-core/v/3-2/docs/example-groups/shared-examples">rspec shared examples</a>.
</p>

<p>
<a href="https://github.com/andreareginato/betterspecs/issues/13">Discuss this guideline &rarr;</a>
</p>

<%= render "partials/share", text: "betterspecs.org | shared examples", url: "http://betterspecs.org/#sharedexamples" %>

</article>

<article>

<h1><a name="integration">Test what you see</a></h1>

<p>
<strong>Deeply test your models and your application behaviour (integration tests).
Do not add useless complexity testing controllers.</strong>
</p>

<p>
When I first started testing my apps I was testing controllers, now I don't.
Now I only create integration tests using RSpec and Capybara. Why? Because I truly
believe that you should test what you see and because testing controllers is an
extra step you don't need. You'll find out that most of your tests go into the
models and that integration tests can be easily grouped into shared examples,
building a clear and readable test suite.
</p>

<p>
This is an open debate in the Ruby community and both sides have good arguments
supporting their idea. People supporting the need of testing controllers will tell
you that your integration tests don't cover all use cases and that they are slow.
</p>

<p>
Both are wrong. You can easily cover all use cases (why shouldn't you?) and you
can run single file specs using automated tools like Guard. In this way you will run
only the specs you need to test blazing fast without stopping your flow.
</p>

<p>
<a href="https://github.com/andreareginato/betterspecs/issues/14">Discuss this guideline &rarr;</a>
</p>

<%= render "partials/share", text: "betterspecs.org | test what you see", url: "http://betterspecs.org/#integration" %>

</article>

<article>

<h1><a name="should">Don't use should</a></h1>

<p>
Do not use should when describing your tests. Use the third person in the present tense.
Even better start using the new
<a href="http://myronmars.to/n/dev-blog/2012/06/rspecs-new-expectation-syntax">expectation</a>
syntax.
</p>

<p class="wrong">bad</p>

<div>
<pre><code class="ruby">it 'should not change timings' do
  consumption.occur_at.should == valid.occur_at
end
</code></pre>
</div>

<p class="correct">good</p>

<div>
<pre><code class="ruby">it 'does not change timings' do
  expect(consumption.occur_at).to equal(valid.occur_at)
end
</code></pre>
</div>

<p>
See <a href="https://github.com/should-not/should_not">the should_not gem</a> for a way to enforce this in RSpec
and <a href="https://github.com/siyelo/should_clean">the should_clean</a> gem for a way to clean up existing RSpec examples that begin with "should."
</p>

<p>
<a href="https://github.com/andreareginato/betterspecs/issues/15">Discuss this guideline &rarr;</a>
</p>

<%= render "partials/share", text: "betterspecs.org | don't use should", url: "http://betterspecs.org/#should" %>

</article>

<article>

<h1><a name="guard">Automatic tests with guard</a></h1>

<p>
Running all the test suite every time you change your app can be cumbersome. It takes a lot of
time and it can break your flow. With Guard you can automate your test suite running only the
tests related to the updated spec, model, controller or file you are working at.
</p>

<p class="correct">good</p>

<div>
<pre><code class="ruby">bundle exec guard
</code></pre>
</div>

Here you can see a sample Guardfile with some basic reloading rules.

<p class="correct">good</p>

<div>
<pre><code class="ruby">guard 'rspec', cli: '--drb --format Fuubar --color', version: 2 do
  # run every updated spec file
  watch(%r{^spec/.+_spec\.rb$})
  # run the lib specs when a file in lib/ changes
  watch(%r{^lib/(.+)\.rb$}) { |m| "spec/lib/#{m[1]}_spec.rb" }
  # run the model specs related to the changed model
  watch(%r{^app/(.+)\.rb$}) { |m| "spec/#{m[1]}_spec.rb" }
  # run the view specs related to the changed view
  watch(%r{^app/(.*)(\.erb|\.haml)$}) { |m| "spec/#{m[1]}#{m[2]}_spec.rb" }
  # run the integration specs related to the changed controller
  watch(%r{^app/controllers/(.+)\.rb}) { |m| "spec/requests/#{m[1]}_spec.rb" }
  # run all integration tests when application controller change
  watch('app/controllers/application_controller.rb') { "spec/requests" }
end
</code></pre>
</div>

<p>
Guard is a fine tool but as usual it doesn't fit all of your needs. Sometimes your TDD workflow
works best with a keybinding that makes it easy to run just the examples you want when you
want to. Then, you can use a rake task to run the entire suite before pushing code.
Find an example vim keybinding <a href="https://github.com/myronmarston/vim_files/blob/5bd4faad7c020ebcbf62dcbc59985262b4eacb53/vimrc.after#L61-103">here</a>.
</p>

<p>
Learn more about <a href="https://github.com/guard/guard-rspec">guard-rspec</a>.
</p>

<p>
<a href="https://github.com/andreareginato/betterspecs/issues/16">Discuss this guideline &rarr;</a>
</p>

<%= render "partials/share", text: "betterspecs.org | automatic tests with guard", url: "http://betterspecs.org/#guard" %>

</article>

<article>

<h1><a name="spork">Faster tests (preloading Rails)</a></h1>

<p>
When running a test on Rails the whole Rails app is loaded. This can take time and
it can break your development flow. To solve this problem use solutions like
<a href="https://github.com/burke/zeus">Zeus</a>,
<a href="https://github.com/jstorimer/spin">Spin</a> or
<a href="https://github.com/sporkrb/spork">Spork</a>.
Those solutions will preload all libraries you (usually) do not change and reload
controllers, models, view, factories and all the files you change most often.
</p>

<p>
Here you can find a <a href="https://gist.github.com/3821012">spec helper</a> and a
<a href="https://gist.github.com/3821031">Guardfile</a> configuration based on Spork.
With this configuration you will reload the whole app if a preloaded file (like initializers)
change and you will run the single tests really, really fast.
</p>

<p>
The drawback of using Spork is that it aggressively monkey-patches your code and you
could lose some hours trying to understand why a file is not reloaded. If you have some
code examples using Spin or any other solution
<a href="https://github.com/andreareginato/betterspecs/issues/17">let us know</a>.
</p>

<p>
Here you can find a <a href="https://gist.github.com/HuffMoody/5912373">Guardfile</a>
configuration for using Zeus.  The spec_helper does not need to be modified, however,
you will have to run `zeus start` in a console to start the zeus server before running
your tests.
</p>

<p>
Although Zeus takes a less aggressive approach than Spork, one major drawback is the fairly
strict usage requirements; Ruby 1.9.3+ (recommended using backported GC from Ruby 2.0)
as well as an operating system that supports FSEvents or inotify is required.
</p>

<p>
Many criticisms are moved to those solutions. Those libraries are a band aid on a problem
that is better solved through better design, and being intentional about only loading the
dependencies that you need. Learn more reading the
<a href="https://github.com/andreareginato/betterspecs/issues/17">related discussion</a>.
</p>

<p>
<a href="https://github.com/andreareginato/betterspecs/issues/17">Discuss this guideline &rarr;</a>
</p>

<%= render "partials/share", text: "betterspecs.org | faster tests with spork", url: "http://betterspecs.org/#spork" %>

</article>

<article>

<h1><a name="http">Stubbing HTTP requests</a></h1>

<p>
Sometimes you need to access external services. In these cases you can't
rely on the real service but you should stub it with solutions like webmock.
</p>

<p class="correct">good</p>
<div>
<pre><code class="ruby">context "with unauthorized access" do
  let(:uri) { 'http://api.lelylan.com/types' }
  before    { stub_request(:get, uri).to_return(status: 401, body: fixture('401.json')) }
  it "gets a not authorized notification" do
    page.driver.get uri
    expect(page).to have_content 'Access denied'
  end
end
</code></pre>
</div>

<p>
Learn more about <a href="https://github.com/bblimke/webmock">webmock</a> and
<a href="https://github.com/vcr/vcr">VCR</a>. Here is a
<a href="http://marnen.github.com/webmock-presentation/webmock.html">nice presentation</a>
explaining how to mix them together.
</p>

<p>
<a href="https://github.com/andreareginato/betterspecs/issues/18">Discuss this guideline &rarr;</a>
</p>

<%= render "partials/share", text: "betterspecs.org | mocking http requests", url: "http://betterspecs.org/#http" %>

</article>

<article>

<h1><a name="formatter">Useful formatter</a></h1>

<p>
Use a formatter that can give you useful information about the test suite.
I personally find fuubar really nice. To make it work add the gem and set
fuubar as default formatter in your Guardfile.
</p>

<p class="correct">good</p>

<div>
<pre><code class="ruby"># Gemfile
group :development, :test do
  gem 'fuubar'
</code></pre>
</div>

<p class="correct">good</p>

<div>
<pre><code class="ruby"># .rspec
--drb
--format Fuubar
--color
</code></pre>
</div>

<p>
Learn more about <a href="http://jeffkreeftmeijer.com/2010/fuubar-the-instafailing-rspec-progress-bar-formatter/">fuubar</a>.
</p>

<p>
<a href="https://github.com/andreareginato/betterspecs/issues/19">Discuss this guideline &rarr;</a>
</p>

<%= render "partials/share", text: "betterspecs.org | useful formatter", url: "http://betterspecs.org/#formatter" %>

</article>

<article>

<h1><a name="books">Books</a></h1>
  <%= render "partials/books" %>

</article>

<article>

<h1><a name="presentations">Presentations</a></h1>

<iframe src="http://www.slideshare.net/slideshow/embed_code/7050468" width="427" height="356" frameborder="0" marginwidth="0" marginheight="0" scrolling="no" style="border:1px solid #CCC;border-width:1px 1px 0;margin-bottom:5px" allowfullscreen> </iframe> <div style="margin-bottom:5px"> <strong> <a href="http://www.slideshare.net/andrea.reginato/rspec-2-best-practices" title="RSpec 2 Best practices" target="_blank">RSpec 2 Best practices</a> </strong> from <strong><a href="http://www.slideshare.net/andrea.reginato" target="_blank">Andrea Reginato</a></strong> </div>

</article>

<article>
  <h1><a name="resources">Resources on the web</a></h1>
  <%= render "partials/links" %>
</article>

<article>
  <h1><a name="screencasts">Screencasts</a></h1>
  <%= render "partials/screencasts" %>
</article>

<article>
  <h1><a name="libraries">Libraries (documentation)</a></h1>
  <%= render "partials/libraries" %>
</article>

<article>

<h1><a name="styleguide">Styleguide</a></h1>

<p>
We are seeking for the best guidelines to write "nice to read" specs. Right now a good
starting point is for sure the
<a class="see" href="https://github.com/mongoid/mongoid/tree/master/spec/mongoid">Mongoid test suite</a>.
It uses a clean style and easy-to-read specs, following most of the guidelines
described here.
</p>

</article>

<article>

<h1><a name="improving">Improving Better Specs</a></h1>

<p>
This is an open source project. If something is missing or incorrect just file an
<a class="see" href="https://github.com/andreareginato/betterspecs/issues">issue</a>
to discuss the topic. Also check the following issues:

<ul>
  <li><a href="https://github.com/andreareginato/betterspecs/issues/20">Multilanguage</a> (file an issue if you want to translate this guide)</li>
</ul>

</p>

</article>

<article>

<h1><a name="credits">Credits</a></h1>

<p>
This project was created by <a href="http://twitter.com/andreareginato"></a> and released as
open-source thanks to <a href="http://lelylan.com">Lelylan</a>, a new platform to monitor and
control your devices through a simple, open and robust REST API.
</p>

<p>
If you like what <a href="http://twitter.com/andreareginato">I'm</a> doing offer me a coffe
<script data-gittip-username="andreareginato"
        data-gittip-widget="button"
        src="//gttp.co/v1.js"></script>
</p>

</article>

<br><br><hr><br><br>

<iframe src="http://ghbtns.com/github-btn.html?user=andreareginato&repo=betterspecs&type=watch&count=true&size=large"
  allowtransparency="true" frameborder="0" scrolling="0" width="170px" height="30px"></iframe>

<iframe src="http://ghbtns.com/github-btn.html?user=andreareginato&repo=betterspecs&type=fork&count=true&size=large"
  allowtransparency="true" frameborder="0" scrolling="0" width="170px" height="30px"></iframe>

<iframe src="http://ghbtns.com/github-btn.html?user=andreareginato&repo=betterspecs&type=follow&count=true&size=large"
  allowtransparency="true" frameborder="0" scrolling="0" width="300px" height="30px"></iframe>

</div>