In my Rails 5 app I have this:
class InvoicesController < ApplicationController
def index
#invoices = current_account.invoices
respond_to do |format|
format.csv do
invoices_file(:csv)
end
format.xml do
invoices_file(:xml)
end
end
end
private
def invoices_file(type)
headers['Content-Disposition'] = "inline; filename=\"invoices.#{type.to_s}\""
end
end
describe InvoicesController, :type => :controller do
it "renders a csv attachment" do
get :index, :params => {:format => :csv}
expect(response.headers["Content-Type"]).to eq("text/csv; charset=utf-8")
expect(response).to have_http_status(200)
expect(response).to render_template :index
end
end
My problem is that my Spec always passes (!), even when I put a bunch of crap into my index.csv.erb file. It seems that the view file isn't even evaluated / tested by RSpec.
How is this possible? What am I missing here?
Controller tests/specs are these weird stubbed creations born out of the idea of unit testing controllers in isolation. That idea turned out to be pretty flawed and has really fallen out of vogue lately.
Controller specs don't actually make a real HTTP request to your application that passes through the routes. Rather they just kind of fake it and pass a fake request through.
To make the tests faster they also don't really render the views either. Thats why it does not error out as you have expected. And the response is not really a real rack response object either.
You can make RSpec render the views with render_views.
describe InvoicesController, :type => :controller do
render_views
it "renders a csv attachment" do
get :index, format: :csv
expect(response.headers["Content-Type"]).to eq("text/csv; charset=utf-8")
expect(response).to have_http_status(200)
expect(response).to render_template :index
end
end
But a better and more future proof option is using a request spec.
The official recommendation of the Rails team and the RSpec core team
is to write request specs instead. Request specs allow you to focus on
a single controller action, but unlike controller tests involve the
router, the middleware stack, and both rack requests and responses.
This adds realism to the test that you are writing, and helps avoid
many of the issues that are common in controller specs.
http://rspec.info/blog/2016/07/rspec-3-5-has-been-released/
# spec/requests/invoices
require 'rails_helper'
require 'csv'
RSpec.describe "Invoices", type: :request do
let(:csv) { response.body.parse_csv }
# Group by the route
describe "GET /invoices" do
it "renders a csv attachment" do
get invoices_path, format: :csv
expect(response.headers["Content-Type"]).to eq("text/csv; charset=utf-8")
expect(response).to have_http_status(200)
expect(csv).to eq ["foo", "bar"] # just an example
end
end
end
The format option should be specified outside of the params, i.e. get :index, params: {}, format: :csv}.
Regarding RSpec evaluating views, no, in controller tests, it doesn't, regardless of the format. However, it's possible to test views with RSpec: https://relishapp.com/rspec/rspec-rails/v/2-0/docs/view-specs/view-spec
Related
I am new to RSpec but here I am trying to create tests based on this code and I am keep on getting this error. Any suggestions?
CODE:
serialization_scope nil
before_action :set_list, only: [:show, :destroy, :update]
before_action :verify_user, only: :show
def create
#list = current_user.lists.build(list_params)
if #list.save
render json: {message: ['Success']}, status: 200
else
render json: {errors:[#list.errors.full_messages]}, status: 400
end
end
Here is the RSpec file that I started :
require "rails_helper"
RSpec.describe V1::ListsController, :type => :controller do
describe "POST create" do
it "returns HTTP status" do
expect(post :create).to change(#list, :count).by(+1)
expect(response).to have_http_status :success #200
end
end
describe 'GET status if its not created' do
it "return HTTP status - reports BAD REQUEST (HTTP status 400)" do
expect(response.status).to eq 400
end
end
end
And the error that I got is :
Failures:
1) V1::ListsController GET status if its created returns HTTP status
Failure/Error: expect(post :create).to change(#list, :count).by(+1)
expected #count to have changed by 1, but was not given a block
# ./spec/controllers/lists_controller_spec.rb:8:in `block (3 levels) in <top (required)>'
2) GET status if its not created return HTTP status - reports BAD REQUEST (HTTP status 400)
Failure/Error: expect(response.status).to eq 400
expected: 400
got: 200
(compared using ==)
Try this code.
require 'rails_helper'
RSpec.describe V1::ListsController, type: :request do
describe 'valid request' do
it 'returns HTTP status' do
post '/list', params: { list: { list_name: 'xyz' } }
expect(response.status).to eq 201
end
end
describe 'invalid request' do
it "should return unauthorized" do
post '/list'
assert_response :unauthorized
end
end
end
In params you need to pass your list_params.
Spec would look like:
describe "POST create" do
context 'valid request' do
it 'should increase #list item' do
expect { post :create }.to change(List, :count).by(1)
end
it "returns HTTP status" do
post :create
expect(response).to have_http_status :success #200
end
end
context 'invalid request' do
it "return HTTP status - reports BAD REQUEST (HTTP status 400)" do
get :create
expect(response.status).to eq 400
end
end
end
Cheers!
You can test an object not being created by intentionally causing some of its validations to fail e.g. you can pass a mandatory attribute as nil from the RSpec.
Sample request: post :create, { title: nil }.
But as per your RSpec code, it seems there are no validations on List model. So, lets try to stub save and return false for this particular test.
describe 'GET status if its not created' do
# Assuming your model name is `List`
before { allow_any_instance_of(List).to receive(:save) { false } }
it "return HTTP status - reports BAD REQUEST (HTTP status 400)" do
post :create
expect(response.status).to eq 400
end
end
Please post your model for list and i can update the answer with more appropriate test.
Ishika, let me see if I can help you :)
RSpec official documentation recommends you to use request specs instead of controller specs. That is recommended because Rails 5 deprecated some methods used on controller testings. You can read more about this here at RSpec blog
ps.: You can use controller tests so far, but it can be deprecated in a future major version of RSpec.
There are some notes I left after the code, please read them also.
I would write a request spec like this:
# spec/requests/v1/lists_controller_create_spec.rb
require "rails_helper"
RSpec.describe V1::ListsController do
describe 'success' do
it 'returns ok and creates a list', :aggregate_failures do # :aggregate_failures is available only for RSpec 3.3+
expect do
post '/list', title: 'foo' # This will also test your route, avoiding routing specs to be necessary
end.to change { List.count }.from(0).to(1)
expect(response).to have_http_status(:ok)
end
end
describe 'bad request' do
before do
# This is needed because your controller is not validating the object, but look at my
# comment below (out of the code), to think about this behavior, please.
allow_any_instance_of(List).to receive(:save).and_return(false)
end
it 'returns a bad request and does not create a list' do
expect do
post '/list', title: 'foo' # This will also test your route, avoiding routing specs to be necessary
end.not_to change { List.count }
expect(response).to have_http_status(:bad_request)
end
end
end
Notes:
I suggested using more than 1 expectation by example, that is ok in this spec because they are simple and because I'm using :aggregate_failures option. With this option, if the first expectation fails, the next expectations will also be executed, considering that in this case, the following expectations does not depend on the first one, it is ok to use more than 1 expectation for the example.Reference
You are returning a bad request if the object is not saved, but you are not validating it. If your model has validations that will validate the object there, please adjust the specs to fail the save (instead of using the mock I used) and consider rendering an error message in the response
If you think that making the post inside a expect block, you can do different: Store the count of Lists in a variable before making the post and after the post you test if the variable has changed or not, maybe you think it will be more clear and it will do exactly the same thing in the background.
I've got routes setup so that they work as expected in my controllers; I can use both room_path and rooms_path as expected.
However when I try to use the same routes in a controller spec for some reason then I get an error:
ActionController::UrlGenerationError:
No route matches {:action=>"/1", :controller=>"rooms"}
My routes.rb file:
root "rooms#index"
resources :rooms, :path => '/', only: [:index, :create, :show] do
resources :connections, only: [:create,:destroy]
end
And if I rake routes:
room_connections POST /:room_id/connections(.:format) connections#create
room_connection DELETE /:room_id/connections/:id(.:format) connections#destroy
rooms GET / rooms#index
POST / rooms#create
room GET /:id(.:format) rooms#show
However my test fails:
describe "GET room_path(room)" do
it "renders show" do
#room = Room.create
get room_path(#room)
expect(response.status).to eq(200)
expect(response).to render_template(:show)
end
end
While my controllers can use the same route helpers without issue:
class RoomsController < ApplicationController
def index
end
def create
#room = Room.create
redirect_to room_path(#room)
end
def show
#room = Room.find(params[:id])
end
end
I'm not sure why in my tests it seems to go looking for a "/1" action rather than rooms#show like I would expect.
Update
So continuing to play this I've been able to get the test green by changing to the following:
describe "GET room_path(room)" do
it "renders show" do
#room = Room.create
get :show, params: { id: #room.id }
expect(response.status).to eq(200)
expect(response).to render_template(:show)
end
end
I would still love to understand why my helpers aren't working though. Is this to be expected? Manually writing the Parameters hash is kind of a PITA.
I don't know what version of Rails and RSpec you're on. This works for me on Rails 4.2 and Rspec 3.4.4:
describe "my neat test description", type: :routing do
it "can use path helper" do
puts whatever_path
end
end
The type: :routing pulls in the path and url helpers.
I believe the reason you don't have that by default is that rspec is replicating a lot of the environment for the different types of tests. For controller tests, it's pulling in the various path and url helpers because generally speaking they're used there often enough to be worth pulling in. In model tests, for example, they aren't used there often so they aren't pulled in by default.
These helpers live on the app object.
I'm trying to write some rspec tests to check API endpoints for an API-only application.
Testing error
Failure/Error: expect( res ).to be_success
expected 200 to respond to `success?`
But if the same call (with full api url) is made from another application it works fine and returns a response.
Example from other application:
res = RestClient.get "site.io/api/v1/projects/1"
p JSON.parse(res)
Blog example I'm trying to follow: (http://matthewlehner.net/rails-api-testing-guidelines/).
# spec/requests/api/v1/messages_spec.rb
describe "Messages API" do
it 'sends a list of messages' do
FactoryGirl.create_list(:message, 10)
get '/api/v1/messages'
json = JSON.parse(response.body)
# test for the 200 status-code
expect(response).to be_success
# check to make sure the right amount of messages are returned
expect(json['messages'].length).to eq(10)
end
end
My Application
/requests/projects_spec.rb
require 'rails_helper'
RSpec.describe Project do
describe "show_project" do
before do
#project1 = create(:project)
end
it "Checks if responds successfully" do
res = get '/api/v1/projects/1'
expect( res ).to be_success
end
end
end
/factories/projects.rb
FactoryGirl.define do
factory :project do
name "Thing"
key "123123"
end
end
routes.rb
namespace :api, :defaults => { :format => 'json'} do
namespace :v1 do
resources :projects, only: [:create, :show]
end
end
end
I don't have much experience with testing, so if anyone can point me in the correct direction I would really really appreciate it.
When using Rspec Request Specs, your call to get '/api/v1/projects/1' doesn't need to captured by your res variable. Spec Request tests automatically set the value of response when get '/api/v1/projects/1' is run. The example you're following is correct, it just looks like your missing some knowledge about how much Rspec is handling for you behind the scenes. This makes your test simpler:
it "Checks if responds successfully" do
get '/api/v1/projects/1'
expect(response).to be_success
end
In Rspec Request tests, response is automatically setup by the call the get without you needing to do anything extra.
I've recently learned how to stub in rspec and found that some benefits of it are we can decouple the code (eg. controller and model), more efficient test execution (eg. stubbing database call).
However I figured that if we stub, the code can be tightly tied to a particular implementation which therefore sacrifice the way we refactor the code later.
Example:
UsersController
# /app/controllers/users_controller.rb
class UsersController < ApplicationController
def create
User.create(name: params[:name])
end
end
Controller spec
# /spec/controllers/users_controller_spec.rb
RSpec.describe UsersController, :type => :controller do
describe "POST 'create'" do
it 'saves new user' do
expect(User).to receive(:create)
post :create, :name => "abc"
end
end
end
By doing that didn't I just limit the implementation to only using User.create? So later if I change the code my test will fail even though the purpose of both code is the same which is to save the new user to database
# /app/controllers/users_controller.rb
class UsersController < ApplicationController
def create
#user = User.new
#user.name = params[:name]
#user.save!
end
end
Whereas if I test the controller without stubbing, I can create a real record and later check against the record in the database. As long as the controller is able to save the user Like so
RSpec.describe UsersController, :type => :controller do
describe "POST 'create'" do
it 'saves new user' do
post :create, :name => "abc"
user = User.first
expect(user.name).to eql("abc")
end
end
end
Really sorry if the codes don't look right or have errors, I didn't check the code but you get my point.
So my question is, can we mock/stub without having to be tied to a particular implementation? If so, would you please throw me an example in rspec
You should use mocking and stubbing to simulate services external to the code, which it uses, but you are not interested in them running in your test.
For example, say your code is using the twitter gem:
status = client.status(my_client)
In your test, you don't really want your code to go to twitter API and get your bogus client's status! Instead you stub that method:
expect(client).to receive(:status).with(my_client).and_return("this is my status!")
Now you can safely check your code, with deterministic, short running results!
This is one use case where stubs and mocks are useful, there are more. Of course, like any other tool, they may be abused, and cause pain later on.
Internally create calls save and new
def create(attributes = nil, options = {}, &block)
if attributes.is_a?(Array)
attributes.collect { |attr| create(attr, options, &block) }
else
object = new(attributes, options, &block)
object.save
object
end
end
So possibly your second test would cover both cases.
It is not straight forward to write tests which are implementation independent. That's why integration tests have a lot of value and are better suited than unit tests for testing the behavior of the application.
In the code you're presented, you're not exactly mocking or stubbing. Let's take a look at the first spec:
RSpec.describe UsersController, :type => :controller do
describe "POST 'create'" do
it 'saves new user' do
expect(User).to receive(:create)
post :create, :name => "abc"
end
end
end
Here, you're testing that User received the 'create' message. You're right that there's something wrong with this test because it's going to break if you change the implementation of the controllers 'create' action, which defeats the purpose of testing. Tests should be flexible to change and not a hinderance.
What you want to do is not test implementation, but side effects. What is the controller 'create' action supposed to do? It's supposed to create a user. Here's how I would test it
# /spec/controllers/users_controller_spec.rb
RSpec.describe UsersController, :type => :controller do
describe "POST 'create'" do
it 'saves new user' do
expect { post :create, name: 'abc' }.to change(User, :count).by(1)
end
end
end
As for mocking and stubbing, I try to stay away from too much stubbing. I think it's super useful when you're trying to test conditionals. Here's an example:
# /app/controllers/users_controller.rb
class UsersController < ApplicationController
def create
user = User.new(user_params)
if user.save
flash[:success] = 'User created'
redirect_to root_path
else
flash[:error] = 'Something went wrong'
render 'new'
end
end
# /spec/controllers/users_controller_spec.rb
RSpec.describe UsersController, :type => :controller do
describe "POST 'create'" do
it "renders new if didn't save" do
User.any_instance.stub(:save).and_return(false)
post :create, name: 'abc'
expect(response).to render_template('new')
end
end
end
Here I'm stubbing out 'save' and returning 'false' so I can test what's supposed to happen if the user fails to save.
Also, the other answers were correct in saying that you want to stub out external services so you don't call on their API every time you're running your test suite.
How do you test redirect_to :back in rspec?
I get
ActionController::RedirectBackError:
No HTTP_REFERER was set in the request to this action, so redirect_to :back could not be called successfully. If this is a test, make sure to specify request.env["HTTP_REFERER"].
How do I go about setting the HTTP_REFERER in my test?
Using RSpec, you can set the referer in a before block. When I tried to set the referer directly in the test, it didn't seem to work no matter where I put it, but the before block does the trick.
describe BackController < ApplicationController do
before(:each) do
request.env["HTTP_REFERER"] = "where_i_came_from"
end
describe "GET /goback" do
it "redirects back to the referring page" do
get 'goback'
response.should redirect_to "where_i_came_from"
end
end
end
From the rails guide when requesting the request with the new request style:
describe BackController < ApplicationController do
describe "GET /goback" do
it "redirects back to the referring page" do
get :show,
params: { id: 12 },
headers: { "HTTP_REFERER" => "http://example.com/home" }
expect(response).to redirect_to("http://example.com/home")
end
end
end
If someone stumbles upon this and they're using request specs, you'll need to explicitly set the headers on the request you're making. The format of the test request depends on which version of RSpec you're using and if you can use keyword arguments instead of positional arguments.
let(:headers){ { "HTTP_REFERER" => "/widgets" } }
it "redirects back to widgets" do
post "/widgets", params: {}, headers: headers # keyword (better)
post "/widgets", {}, headers # positional
expect(response).to redirect_to(widgets_path)
end
https://relishapp.com/rspec/rspec-rails/docs/request-specs/request-spec
IMHO the accepted answer is a bit hacky. A better alternative would be to set the HTTP_REFERER to an actual url in your application and then expect to be redirected back:
describe BackController, type: :controller do
before(:each) do
request.env['HTTP_REFERER'] = root_url
end
it 'redirects back' do
get :whatever
response.should redirect_to :back
end
end
Redirecting to a random string constant feels as if it works by accident
You take advantage of rspec's built in functionality to express exactly what you wanted
You don't introduce and repeat magic string values
For newer versions of rspec, you can use expectations instead:
expect(response).to redirect_to :back
In regard to testing :back links in integration tests, I first visit a deadend page, that I think is not likely to be ever used as a link, and then the page I am testing. So my code looks like this
before(:each) do
visit deadend_path
visit testpage_path
end
it "testpage Page should have a Back button going :back" do
response.should have_selector("a",:href => deadend_path,
:content => "Back")
end
However this does have the flaw that if the link is really to the deadend_path, then the test will incorrectly pass.
request.env['HTTP_REFERER'] = '/your_referring_url'