这是一个简单的关于Rails Rspec的简单的介绍
1 安装Rspec
在Rails的配置文件Gemfile配置文件中,配置下面信息
group :development, :test do
gem 'rspec-rails', '2.13.1'
end
我们没有必要单独的安装RSpec ,因为它是rspec-rails的依赖件
会被自动安装.
执行bundle install 或者 bundle install --without production 来安装使用的gem.
rails generate rspec:install
设置让Rails使用RSpec 而不用Test::Unit. 该命令执行完毕之后,会产生一个文件夹spec,该文件加下面有spec/spec_helper.rb 这个文件,spec_helper.rb用来设置测试的配置信息.
下面是spec的固定的规范,固定的格式.
describe XXX do
it XXX do
......
end
end
包含了一个 describe 块以及其中的一个测试用例（sample），以 it "..." do 开头的代码块就是一个用例.
2 pending 测试大纲
describe xxxController do
describe 'GET #show' do
it "assigns the requested xxx to @xxx"
it "renders the :show template"
end
describe 'GET #new' do
it "assigns a new xxx to @xxx"
it "renders the :new template"
end
describe 'GET #edit' do
it "assigns the requested xxx to @xxx"
it "renders the :edit template"
end
describe "POST #create" do
context "with valid attributes" do
it "saves the new xxx in the database"
it "redirects to xxx#show"
end
context "with invalid attributes" do
it "does not save the new xxx in the database"
it "re-renders the :new template"
end
end
describe 'PUT #update' do
context "with valid attributes" do
it "updates the xxx in the database"
it "redirects to the xxx"
end
context "with invalid attributes" do
it "does not update the xxx"
it "re-renders the #edit template"
end
end
describe 'DELETE #destroy' do
it "deletes the xxx from the database"
it "redirects to users#index"
end
describe 'GET #index' do
it "response.status eq 200"
it "renders the :index view"
end
end
3 简单的测试
测试代码的结构
describe AA do
it 'should do something' do
something.should
end
end
使用 http://railstutorial-china.org/ 里面的列子,来做一个简单的测试演示.
rails generate controller StaticPages home help --no-test-framework
该命令会在app/controller文件下生成static_pages_controller.rb,并且会在路由里面生成help,home对应的路由.
match '/help', to: 'static_pages#help', via: 'get'
match '/home', to: 'static_pages#home', via: 'get'
之后运行下面命令,产生spec/requests/static_pages_spec.rb
rails generate integration_test static_pages
或者直接手动创建spec/requests/static_pages_spec.rb这个文件.
下面做一个简单的测试
describe "Home page" do
it "should have the content 'Sample App'" do
visit '/static_pages/home'
expect(page).to have_content('Sample App')
end
end
运行bundle exec rspec spec/requests/static_pages_spec.rb, 运行失败!
因为在app/views/static_pages/home.html.erb文件中并没有"Sample App"关键字.在该文件中,添加上"Sample App"测试就能通过.
4 let和let!的用法
Use let to define a memoized helper method. The value will be cached
across multiple calls in the same example but not across examples.
Note that let is lazy-evaluated: it is not evaluated until the first time
the method it defines is invoked. You can use let! to force the method's
invocation before each example.
http://stackoverflow.com/questions/10173097/rails-rspec-difference-between-let-and-let
使用let来定义一个memoized helper方法。该值可以在多个例子中使用，但不能跨多个实例调用。
-
spec/helpers/let_spec.rb *
describe "let" do
other_count = 0
invocation_order = []let(:count) do
invocation_order << :let!
other_count += 1
endit "calls the helper method in a before hook" do
# count
# invocation_order << :example
# expect(invocation_order).to eq([:let!,:example])invocation_order << :example expect(invocation_order).to eq([:example]) expect(other_count).to eq(0) other_count +=1 expect(other_count).to eq(1)end
it "calls the helper method in a before hook again" do
# count
# invocation_order << :example
# expect(invocation_order).to eq([:let!,:example])
invocation_order << :example
expect(invocation_order).to eq([:example,:example])
expect(other_count).to eq(1)
other_count +=1
expect(other_count).to eq(2)
end
end
describe "let test" do
let(:count) { $count += 1 }
# 可以使用count
end
describe "no let test" do
# 不能使用count
end
下面是let!的使用方法
-
spec/helpers/let_bang_spec.rb *
describe "let!" do
count = 0
invocation_order = []let!(:count) do
invocation_order << :let!
count += 1
endit "calls the helper method in a before hook" do
invocation_order << :example
expect(invocation_order).to eq([:let!, :example])
expect(count).to eq(1)
endit "calls the helper method in a before hook again" do
invocation_order << :example
expect(invocation_order).to eq([:let!, :example, :let!, :example])
expect(count).to eq(2)
end
end
5 before用法
http://stackoverflow.com/questions/5974360/rspec-difference-between-let-and-before-block
http://stackoverflow.com/questions/5359558/when-to-use-rspec-let?lq=1
let 和 before(:each)的区别, let不会自动初始化变量,而before(:each)会自动初始化变量.如果我其中的某一个测试用力不需要这些变量,依然需要初始化,如初始化变量需要很多时间,对这个测试的初始化就是浪费的时间和资源.
before(:each)的用法
-
spec/helpers/before_each_spec.rb *
require "rspec/expectations"
class Thing
def widgets
@widgets ||= []
end
end
describe Thing do
before(:each) do
@thing = Thing.new
end
describe "initialized in before(:each)" do
it "has 0 widgets" do
@thing.should have(0).widgets
end
it "can get accept new widgets" do
@thing.widgets << Object.new
end
it "does not share state across examples" do
@thing.should have(0).widgets
end
it "does not have 1 count" do
@thing.should_not have(1).widgets
end
end
end
before(:all)用法
-
spec/helpers/before_all_spec.rb *
require "rspec/expectations"
class Thing
def widgets
@widgets ||= []
end
end
describe Thing do
before(:all) do
@thing = Thing.new
end
describe "initialized in before(:all)" do
it "has 0 widgets" do
@thing.should have(0).widgets
end
it "can get accept new widgets" do
@thing.widgets << Object.new
end
it "shares state across examples" do
@thing.should have(1).widgets
end
it "should not have 0 widgets" do
@thing.should_not have(0).widgets
end
end
end
6 model 测试
spec/models/widget_spec.rb
describe Widget do
context "test fields" do
before do
@widget = create(:widget)
end
subject{ @widget }
# respond_to 用来判断属性有没有
it { should respond_to(:name) }
it { should respond_to(:email) }
it { should respond_to(:address) }
it { should respond_to(:lat) }
it { should respond_to(:lng) }
it { should be_valid }
# 简约的写法
describe "when name is not present" do
before { @widget.name = " " }
it { should_not be_valid }
end
# 繁琐的写法
describe "when name is not present" do
before { @widget.name = " " }
# 将it的单行的方式,改写成多行的方式
it "should not be valid" do
expect(@widget).not_to be_valid
end
end
describe "when name is invalid" do
before { @widget.name = nil }
it { should have(1).errors_on(:name) }
end
describe "when name length > 50 " do
before do
@widget.name = "a"*51
end
it { should_not be_valid }
end
describe "when email is invalid" do
before { @widget.email = nil }
it do
# Email can't be blank, Email is invalid
should have(2).errors_on(:email)
end
end
describe "when email is invalid" do
it "should be invalid" do
# 使用each 循环,遍历非法的email
addresses = %w[user@foo,com user_at_foo.org example.user@foo.
foo@bar_baz.com foo@bar+baz.com]
addresses.each do |addr|
@widget.email = addr
expect(@widget).not_to be_valid
end
end
end
describe "when email is valid" do
it "should be valid" do
# 使用each 循环,遍历可用的email
addresses = %w[[email protected] [email protected]
[email protected] [email protected]]
addresses.each do |addr|
@widget.email = addr
expect(@widget).to be_valid
end
end
end
# ========== 上述验证邮箱的可该写成如下
describe "when email" do
context "is valid" do
it "should be valid" do
addresses = %w[[email protected] [email protected]
[email protected] [email protected]]
addresses.each do |addr|
@widget.email = addr
expect(@widget).to be_valid
end
end
end
context "is invalid" do
it "should be invalid" do
addresses = %w[user@foo,com user_at_foo.org example.user@foo.
foo@bar_baz.com foo@bar+baz.com]
addresses.each do |addr|
@widget.email = addr
expect(@widget).not_to be_valid
end
end
end
end
end
context "when initialized" do
# 创建默认的共享变量
subject(:widget) { Widget.new(:name => Faker::Name.name,
:email => Faker::Internet.email) }
it "should increment the Relationship count" do
expect do
Widget.create!(:name => "first comment", :email => Faker::Internet.email)
Widget.create!(:name => "second comment", :email => Faker::Internet.email)
end.to change(Widget, :count).by(2) # from(1).to(3)
# https://www.relishapp.com/rspec/rspec-expectations/v/2-0/docs/matchers/expect-change
end
# 第一种写法
it "is a new widget" do
expect(widget).to be_a_new(Widget)
end
# 第二种写法
it { should be_a_new(Widget) }
it "is not a new string" do
expect(widget).not_to be_a_new(String)
end
end
context "when saved" do
subject(:widget) { create(:widget) }
it "is not a new widget" do
expect(widget).not_to be_a_new(Widget)
end
it "is not a new string" do
expect(widget).not_to be_a_new(String)
end
end
end
7 controller测试
spec/controllers/widgets_controller_spec.rb
describe WidgetsController do
# show ============
describe "GET #show" do
#• 和控制器动作交互的基本 DSL 句法:每个 HTTP 请求方法都对应于一个方法(本例中的方法
#是 get) ,其后跟着动作的 Symbol 形式(:show) ,然后是传入的请求参数(id: contact) 。
#• 控制器动作实例化的变量可以通过 assigns(:variable_name) 方法获取。
#• 控制器动作的返回结果可以通过 response 获取。
it "assigns the required 1 to @1" do
widget = create(:widget)
get :show, id: widget
expect(assigns(:widget)).to eq widget
# http://stackoverflow.com/questions/8616508/what-does-assigns
#
end
it "renders the :show template " do
widget = create(:widget)
get :show, id: widget
expect(response).to render_template :show
end
end
# new ============
describe "GET #new" do
it "assigns a new Widget to @widget" do
get :new
expect(assigns(:widget)).to be_a_new(Widget)
end
it "renders the :new template" do
get :new
expect(response).to render_template :new
end
end
# edit ============
describe "GET #edit" do
it "assigns the request widget to @widget" do
widget = create(:widget)
get :edit, id: widget
expect(assigns(:widget)).to eq widget
end
it "renders the :edit template" do
widget = create(:widget)
get :edit, id: widget
expect(response).to render_template :edit
end
end
# create ============
describe "POST #create" do
context "with invalid attributes" do
# create动作, 在符合 REST 架构的程序中,这个动作可以响应 POST 请求。
# create 动作和响应 GET 请求的动作最主要的不同点是,不能像 GET 请求那样只传入:id 参数,而
# 要传入 params[:widget] 对应的值,这个值就是用户可能在表单中输入的内容。
it "does not save the new widget in the database" do
expect{
post :create,
widget: attributes_for(:invalid_widget)
}.to change(Widget, :count).by(0)
end
# expect的另一种写法
it "does save the new widget in the database again" do
expect do
post :create,
widget: attributes_for(:widget)
end.to change(Widget, :count).by(1)
end
it "re-renders the :new template" do
post :create,
widget: attributes_for(:invalid_widget)
expect(response).to render_template :new
end
end
context "with valid attributes" do
it "does save the new widget in the database" do
expect{
post :create,
widget: attributes_for(:widget)
}.to change(Widget, :count).by(1)
end
it "re-renders the assigns[:widget] template" do
post :create,
widget: attributes_for(:widget)
expect(response).to redirect_to widget_path(assigns[:widget])
end
end
# update ============
describe "PATH #update" do
before :each do
@widget = create(:widget, name: 'lisi', email: '[email protected]')
end
context "valid attribute" do
# 先测试请求
it "located the request @widget " do
patch :update, id: @widget, widget: attributes_for(:widget)
expect(assigns(:widget)).to eq(@widget)
end
# 其次测试上传的数据,是否发生了变化
it "changes @widget`s attributes" do
patch :update, id: @widget,
widget: attributes_for(:widget, name: "lol", email: "[email protected]")
@widget.reload
expect(@widget.name).to eq("lol")
expect(@widget.email).to eq("[email protected]")
end
# 最后测试,返回结果
it "redirect to the updated widget" do
patch :update, id: @widget, widget: attributes_for(:widget)
expect(response).to redirect_to @widget
end
end
context "with invalid attributes" do
it "does not change the contact`s attribues" do
patch :update, id: @widget,
widget: attributes_for(:widget, name: nil, email: "[email protected]")
@widget.reload
expect(@widget.name).to eq("lisi")
expect(@widget.email).not_to eq("[email protected]")
end
it "re-renders the edit template" do
patch :update, id: @widget,
widget: attributes_for(:invalid_widget)
expect(response).to render_template :edit
end
end
end
# delete ============
describe "DELETE destroy" do
before :each do
@widget = create(:widget)
end
it "delete the widget" do
expect{
delete :destroy, id: @widget
}.to change(Widget, :count).by(-1)
end
it "redirect to widget#index" do
delete :destroy, id: @widget
expect(response).to redirect_to widgets_path
end
end
end
describe "GET index" do
it "has a 200 status code" do
get :index
expect(response.status).to eq(200)
end
it "renders the index template" do
get :index
expect(response).to render_template("index")
expect(response.body).to eq ""
end
it "renders the widgets/index template" do
get :index
expect(response).to render_template("widgets/index")
expect(response.body).to eq ""
end
it "not renders the 'new' template" do
get :index
expect(response).not_to render_template("new")
end
end
end
8 route测试
首先运行
rails generate controller admin/Accounts index
用这个来创建一个namespace为admin的account_controller.rb
-
spec/routing/admin_routing_spec.rb*
require "spec_helper"
describe "routes for Widgets" do
it "routes /admin/accounts to the admin/accounts controller" do
expect(get("/admin/accounts")).
to route_to("admin/accounts#index")
end
it "routes /admin/accounts to the admin/accounts controller again" do
expect(get("/admin/accounts")).
to route_to(:controller => "admin/accounts", :action => "index")
end
end
spec/routing/widgets_routing_spec.rb
describe "routes for Widgets" do
it "route to widgets" do
expect(:get => "/widgets").to be_routable
end
it "does not route to widgets/foo/bar" do
expect(:get => "/widgets/foo/bar").not_to be_routable
end
end
*spec/routing/widgets_routing_spec.rb *
it "routes a named route" do
expect(:get => new_widget_path).
to route_to(:controller => "widgets", :action => "new")
end
9 factory_girl 的使用
1.修改Gemfile配置
group :development, :test do
gem 'factory_girl_rails', '4.2.1'
gem 'ffaker', '~> 1.21.0'
gem 'rspec-rails', '2.13.1'
end
并且使用 gem 'jbuilder', '1.5.0'
配置完成之后,执行bundle install
2.在 * app/views/widgets/index.json.jbuilder * 文件中
json.widgets @widgets do |widget|
json.id widget.id
json.name widget.name
end
在 * app/views/widgets/show.json.jbuilder*文件中
json.widget do
json.id @widget.id
json.name @widget.name
end
3.在 spec/factories/widget.rb文件中
FactoryGirl.define do
factory :widget do
name { Faker::Name.name }
end
end
4.在* spec/requests/widgets_spec.rb *文件中
describe "Widgets" do
describe "GET /widgets" do
let(:widgets) { FactoryGirl.create_list(:widget, 10) }
let(:url) { "/widgets.json" }
before do
WidgetsController.stub(:index).and_return(widgets)
get(url)
end
describe "return JSON" do
it { MultiJson.decode(response.body)["widgets"].should have_at_most(10).items }
it "should be correct" do
expect(MultiJson.decode(response.body)["widgets"].first["id"]).to eq(widgets.first.id)
expect(MultiJson.decode(response.body)["widgets"].first["name"]).to eq(widgets.first.name)
end
end
end
describe "specify widget" do
let(:widget) { FactoryGirl.create(:widget) }
let(:url) {"/widgets/#{widget.id}.json"}
before { get(url) }
it "should be correct" do
expect(MultiJson.decode(response.body)["widget"]).to include("id")
expect(MultiJson.decode(response.body)["widget"]).to include("name")
end
it "should render the deal in JSON format" do
expect(MultiJson.decode(response.body)["widget"]["id"]).to eq(widget.id)
expect(MultiJson.decode(response.body)["widget"]["name"]).to eq(widget.name)
end
end
end
5.在spec/spec_helper.rb文件中配置如下
每次在测试中,都必须使用FactoryGirl.create(xxx) FactoryGirl.create_list(xxx)
在Factory Girl 3.0 开始, 只要做一个设置,Rails就会变得简洁,美好
可以加在spec_helper.rb文件Rspec.configure块中的任何地方
RSpec.configure do |config|
config.include FactoryGirl::Syntax::Methods
end
然后在测试的代码中,就能使用比较简洁的语法create(xxx) create_list(xxx)
还能使用attributes_for(xxx)等
6.Factory生成文件存放的位置
通过Factory Girl生成器生成的文件都会放在 spec/factories文件夹
中,文件的名字会试用对应模型名字的复数形式(所以Contact
模型的预构件文件是spec/factories/contacts.rb)
10 ffaker的使用
faker在用来创建比较真实的假数据.
ffaker是faker的重写,速度比faker块一些.
修改Gemfile配置
require 'ffaker'
之后执行bundle installl给应用程序安装ffaker gem包.
下面是常用的示例.
Faker::Name.name #=> "Christophe Bartell"
Faker::Internet.email #=> "[email protected]"
Faker::Geolocation.lat #=> 38.4685363769531
Faker::Geolocation.lng #=> -87.888795
Faker::Address.street_address #=> "95519 Evans Oval"
Faker::Lorem.sentence #=> "Aut recusandae quam cum omnis aut dolores."
Faker::Lorem.paragraph #=> "Earum et est qui id totam adipisci sint. In qui officia velit veritatis rerum consequuntur. Aut earum eaque velit. Numquam minima autem at."
11 mock的使用
为什么要使用mock?
系统总是很复杂,不同的模块功能耦合在一起,A调用B, B调用C甚至A.但是我们在测试A
的某个方法的时候,应该把注意力集中在A这个方法功能上,而没有必要把这个方法中需要的其他方法
(A的或者B,C)都测试一遍,虽然这些其他方法的正确是A的这个方法正确的保证!B,C的方法应该在各自的测试中独立进行.
如一个方法，读取rss，然后解析，生成view
url = "http://forum.rccchina.com/api/posts/meetings.xml"
require 'net/http'
require 'uri'
xml = Net::HTTP.get_print URI.parse(url)
这段代码核心在后面解析生成view的功能，测试的时候，完全不需要真的去从forum.rccchina.com上面获取xml的数据，覆盖Net::HTTP.get_print方法即可。
xml = "... "
Net::HTTP.stub!(:get_print).and_return(xml)
下面是一个mock的例子
`spec/factories/widget.rb`
FactoryGirl.define do
factory :widget do
name { Faker::Name.name }
email { Faker::Internet.email }
tel { 123456 }
factory :invalid_widget do
name nil
email nil
end
end
end
`spec/first/mock_spec.rb`
require 'spec_helper'
class Hello
def say
"hello world"
end
end
describe Hello do
context "factory girl spec " do
let(:widget) { FactoryGirl.create(:widget) }
subject { widget }
it { should respond_to(:name) }
it { should respond_to(:email) }
it { should respond_to(:tel) }
end
context "mock saying hello " do
before(:each) do
@hello = mock(Hello)
@hello.stub!(:say).and_return("hello world")
@hello.stub!(:sleep).and_return("sleep")
end
subject { @hello }
it { should respond_to(:say) }
it { should respond_to(:sleep) }
end
context "saying hello" do
before(:each) do
@hello = mock(Hello)
@hello.stub!(:say).and_return("hello world")
@hello.stub!(:sleep).and_return("sleep")
end
it "#sleep should return sleep" do
@hello.should_receive(:sleep).and_return("sleep")
answer = @hello.sleep
answer.should match("sleep")
end
it "#say should return hello world" do
@hello.should_receive(:say).and_return("hello world")
answer = @hello.say
answer.should match("hello world")
end
it "#say should not return zzz" do
@hello.should_receive(:say).and_return("hello world")
answer = @hello.say
answer.should_not match("zzz")
end
end
end
从上述的例子中可以看出,FactoryGirl 是根据模型来模拟真实的字段属性.
而mock可以模拟出模型中不存在的字段属性.mock可以用来虚拟比较复杂的属性
12 shared_examples 的使用
shared_examples 解决在测试用需要复用的代码
包含shared groups 的文件必须在使用前首先被加载,
一种方法是把所有包含shared examples 的文件放到 spec/support目录下
并且在spec/spec_helper.rb中导入他们
Dir["./spec/support/**/*.rb"].each { |f| require f }
spec/helpers/shared_examples_spec.rb
shared_examples 'a collection' do
let(:collection) { described_class.new([7,2,4]) }
context 'initialized with 3 items' do
it "says it has three items" do
expect(collection.size).to eq 3
end
end
describe "#include?" do
context 'with an item that is in the collection' do
it "returns true" do
expect(collection).to include(7)
end
end
context "with an item that is not in the collection" do
it "returns false" do
expect(collection).not_to include(9)
end
end
end
end
describe Array do
it_behaves_like 'a collection'
end
describe Set do
it_behaves_like 'a collection'
end
spec/helpers/shared_examples_block_spec.rb
shared_examples 'a collection' do
context 'initialized with 3 items' do
it "says it has three items" do
expect(collection.size).to eq 3
end
end
describe "#include?" do
context 'with an item that is in the collection' do
it "returns true" do
expect(collection).to include(7)
end
end
context "with an item that is not in the collection" do
it "returns false" do
expect(collection).not_to include(9)
end
end
end
end
describe Array do
it_behaves_like 'a collection' do
let(:collection) { Array.new([7,2,4]) }
end
end
describe Set do
it_behaves_like 'a collection' do
let(:collection) { Set.new([7,2,4]) }
end
end
spec/helpers/shared_examples_block_params_spec.rb
require 'set'
shared_examples "a measurable object" do |measurement, measurement_methods|
measurement_methods.each do |measurement_method|
it "should return #{measurement} from ##{measurement_method}" do
subject.send(measurement_method).should == measurement
end
end
end
describe Array, "with 3 items" do
subject { [1,2,3] }
it_should_behave_like "a measurable object", 3, [:size, :length]
end
describe String , "of 6 characters" do
subject { "FooBar" }
it_should_behave_like "a measurable object", 6, [:size, :length]
end
13 注意
1
describe User do
before { @user = User.new(name: "Example User", email: "[email protected]") }
subject { @user }
it { should respond_to(:name) }
it { should respond_to(:email) }
end
ruby代码中的写法
@user.respond_to?(:name)
在RSpec中可以写成
it "should respond to 'name'" do
expect(@user).to respond_to(:name)
end
因为指定了subject{ @user } 我们还可以写成单行的形式
it { should respond_to(:name) }
2
require 'spec_helper'
describe Contact do
describe "filter last name by letter" do
context "matching letters" do
end
context "non-matching letters" do
end
end
end
严格上来说describe 和 context 是可以互换的
但是describe 用来表示需要实现的功能,而context针对该功能不同的情况。
3
2012年6月,Rspec开发团队宣布,在v2.11中使用了新句法
来替代传统的should式句法,如
it "is true when true" do
true.should be_true
end
新句法会把要测试的值传递给 expect() 方法,然后和匹配器比较:
it "is true when true" do
expect(true).to be_true
end
4
describe "Home page" do
before { visit root_path }
it "should have the content 'Sample App'" do
expect(page).to have_content('Sample App')
end
it "should have the base title" do
expect(page).to have_title("Ruby on Rails Tutorial Sample App")
end
it "should not have a custom page title" do
expect(page).not_to have_title('| Home')
end
end
使用最新的rspec特性, 将重复的东西删除掉
在代码中的每一个用例都出现下面的东西
it "should have the content 'Sample App'" do
同时还使用了
expect(page).to have_content('Sample App')
二者虽然形式不同,要表达的意思却是相同的.而且两个用例都引用了page变量
使用it方法的另一种形式,把测试代码和描述文本合二为一
我们可以告诉Rspec page就是要测试的对象(subject), 这样就可以避免多次使用page
结果修改变为:
subject { page }
describe "Home page" do
before { visit root_path }
it { should have_content('Sample App') }
it { should have_content('Ruby ...')}
it { shoule have_content('| Home') }
end
参考:
http://ibugs.github.com/rspec_test
http://www.relishapp.com/rspec
https://github.com/rspec/rspec-expectations
https://github.com/thoughtbot/factory_girl
http://betterspecs.org/