I have a Rails 3 background job (delayed_job) which sends a hipchat / Campfire message to their API and I want to check the response in my Cucumber feature. Is there a way to get the last HTTP response(s) which VCR have recorded?
The feature looks like this
#vcr
Scenario: Send hipchat message when task created
Given an hipchat_sample integration exists with app: app "teamway"
When I create an "ActionMailer::Error" task to "Teamway"
And all jobs are worked off # invoke Delayed::Worker.new.work_off
Then a hipchat message should be sent "ActionMailer::Error"
In my step definition I want to check the response body:
Then /^a hipchat message should be sent "(.*?)"$/ do |arg1|
# Like this:
# VCR::Response.body.should == arg1
end
VCR already records the request and response, but I do not know how to take them. I think of something similar to catching the emails sent with Pickle's steps. Does anybody have an idea how to do this?
I use rails 3.2.8, cucumber-rails 1.3 and vcr 2.2.4 (with webmock).
Best regards
Torsten
You can use VCR.current_cassette to get the current cassette, and then interrogate that to get the [VCR::HTTPInteraction][1] object you're looking for, but it'll be a bit complex--the VCR cassette stores the newly recorded HTTP interactions separately from the ones it has available for playback and from the ones it has already played back...so you'll need some complex conditionals to ensure things work properly both when your tests are recording and when they are playing back.
Instead, I recommend you use an after_http_request hook:
module HipmunkHelpers
extend self
attr_accessor :last_http_response
end
Before { HipmunkHelpers.last_http_response = nil }
VCR.configure do |c|
c.after_http_request(lambda { |req| URI(req.uri).host == 'hipmunk.com' }) do |request, response|
HipmunkHelpers.last_http_response = response
end
end
Then, in your cucumber step, you can access HipmunkHelpers.last_http_response.
For more details on the after_http_request hook, check out the relish docs.
Related
Consider a simple application where a user fills a form to divide two numbers, in the routes the form data is proceeded [made into float] and then passed as parameters to a python script's function that has the division logic.
The logic fails due to division by 0 is handled as a custom message in the terminal. How does one send this custom message back to the front end UI along with a 500 error message? Trying to make a restful flask app here.
So far I can abort and show a custom message but not the one that propagated from the backend. Also looked into custom error handling but I want to writer of the external python script to be able to write the custom message.
You can Flask errorhandler(errorcode) to manage your errors and display those on the frontend.
#app.errorhandler(500)
def code_500(error):
return render_template("errors/500.html", error=error), 500
You can put whatever else you want in the html template.
You can also call the code_500(error) func directly.
Same principle applies for any other HTTP error code if you want to customize the page and the message (401, 403, 404, etc...).
If you're inside a blueprint, you can use app_errorhandler instead.
You could use the abort() function. From the docs:
When using Flask for web APIs, you can use the same techniques as above to return JSON responses to API errors. abort() is called with a description parameter. The errorhandler() will use that as the JSON error message, and set the status code to 404.
You could implement it like this
#app.route("/div")
def divide():
x, y = request.form['x'], request.form['y']
try:
result = x / y
except ZeroDivisionError:
abort(400, description="Your message here")
else:
# Proper response
From there, the important step is properly catching that message on your frontend.
I run some load tests (all endpoints) and we do have a known issue in our code: if multiple POST requests are sent in the same time we do get a duplicate error based on a timestamp field in our database.
All I want to do is to count timeouts (based on the message received "Service is not available. Request timeout") in a variable and accept this as a normal behavior (don't fail the tests).
For now I've added a Response Assertion for this (in order to keep the tests running) but I cannot tell if or how many timeout actually happen.
How can I count this?
Thank you
I would recommend doing this as follows:
Add JSR223 Listener to your Test Plan
Put the following code into "Script" area:
if (prev.getResponseDataAsString().contains('Service is not available. Request timeout')) {
prev.setSampleLabel('False negative')
}
That's it, if sampler will contain Service is not available. Request timeout in the response body - JMeter will change its title to False negative.
You can even mark it as passed by adding prev.setSuccessful(false) line to your script. See Apache Groovy - Why and How You Should Use It article fore more information on what else you can do with Groovy in JMeter tests
If you just need to find out the count based on the response message then you can save the performance results in a csv file using simple data writer (configure for csv only) and then filter csv based on the response message to get the required count. Or you can use Display only "errors" option to get all the errors and then filter out based on the expected error message.
If you need to find out at the runtime then you can use aggregate report listener and use "Errors" checkbox to get the count of failure but this will include other failures also.
But, if you need to get the count at the run time to use it later then it is a different case. I am assuming that it is not the case.
Thanks,
I have three different API calls happening when a record is created.
1) Call to create a bitly URL for the record.
2) Call to post to facebook with the bitly URL
3) Call to post to twitter with the bitly URL
My current process is as follows:
On record Create and Update
respond_to do |format|
if #dealer.save
call_bitly_api
end
end
in my model:
after_save :social_media_posting
def social_media_posting
if (self.bitly_url.present? && self.posted_to_facebook == false)
call_facebook_post
end
if (self.bitly_url.present? && self.posted_to_twitter == false)
call_twitter_post
end
end
The issue that I am facing is that the facebook and twitter post is being called on first save where the bitly_url is not yet created.
Need help is figuring out how to add these calls that they can still happen yet they happen in order and wait for the bitly_url to be present before the call the facebook and twitter is made? Major thing worth mentioning I am using sidekiq to make the calls and send the actual call to the sidekiq worker to work in the background.
In controller:
CallBitlyWorker.perform_async(dealer.id)
In your worker:
class CallBitlyWorker
include Sidekiq::Worker
def perform(dealer_id)
dealer = Dealer.find(dealer_id)
# make bitly call
dealer.update_attribute(:bitly_url, some_url)
# make the social media calls in parallel
CallFacebook.perform_async(dealer_id)
CallTwitter.perform_async(dealer_id)
end
end
ActiveRecord callbacks should be avoided where possible. They just make your code more opaque.
I am trying to integrate SEOMOZ API with VCR.
As the API request to SEOMOZ contains parameters that change for the same request over time, I need to implement a custom matcher.
Here is what my API call looks like :
http://lsapi.seomoz.com/linkscape/url-metrics/#{self.url}?Cols=#{cols}&AccessID=#{moz_id}&Expires=#{expires}&Signature=#{url_safe_signature}
I also make calls to other endpoints such as Twitter,Facebook etc etc. For which the default matcher does the job well.
How can I override the matcher behavior just for SEOMOZ. Also on what parameters should I best match the request in this case.
You'll want to match on all parameters except Signature and Expires.
Another option you might consider (we use it internally when using VCR with this sort of API) is to record the time of the test in a file with the cassettes, and use Timecop or something equivalent to ensure you're re-running the recorded test at the "same time" every time you run it.
You can use the VCR.request_matchers.uri_without_params custom matcher, see https://relishapp.com/vcr/vcr/v/2-5-0/docs/request-matching/uri-without-param-s
You would use it like this in your case:
VCR.configure do |c|
# ...
c.default_cassette_options = {
# ...
# the default is: match_requests_on: [:method :uri]
match_requests_on: [:method, VCR.request_matchers.uri_without_params(:Signature, :Expires)]
}
end
I'd like to test if an email is delivered if I call a controller method with :post. I'll use email_spec so I tried this snipped here: http://rubydoc.info/gems/email_spec/1.2.1/file/README.rdoc#Testing_In_Isolation
But it doesn't work, because I pass an instance of the model-object to the delivery-method and the instance is saved before the delivery.
I tried to create an other instance of the model-object, but then the id isn't the same.
My controller-method looks like this:
def create
#params = params[:reservation]
#reservation = Reservation.new(#params)
if #reservation.save
ReservationMailer.confirm_email(#reservation).deliver
redirect_to success_path
else
#title = "Reservation"
render 'new'
end
end
Do you have any idea to solve this?
Assuming your test environment is set up in the usual fashion (that is, you have config.action_mailer.delivery_method = :test), then delivered emails are inserted into the global array ActionMailer::Base.deliveries as Mail::Message instances. You can read that from your test case and ensure the email is as expected. See here.
Configure your test environment to accumulate sent mails in ActionMailer::Base.deliveries.
# config/environments/test.rb
config.action_mailer.delivery_method = :test
Then something like this should allow you to test that the mail was sent.
# Sample parameters you would expect for POST #create.
def reservation_params
{ "reservation" => "Drinks for two at 8pm" }
end
describe MyController do
describe "#create" do
context "when a reservation is saved" do
it "sends a confirmation email" do
expect { post :create, reservation_params }.to change { ActionMailer::Base.deliveries.count }.by(1)
end
end
end
end
Note that my example uses RSpec 3 syntax.
I know I'm late to the party with this one, but for future Googlers...
I think a better solution to this problem is answered here
The previously accepted answer is testing the Mailer itself (inside the controller spec). All you should be testing for here is that the Mailer gets told to deliver something with the right parameters.
You can then test the Mailer elsewhere to make sure it responds to those parameters correctly.
ReservationMailer.should_receive(:confirm_email).with(an_instance_of(Reservation))
This is way how to test that Mailer is called with right arguments. You can use this code in feature, controller or mailer spec:
delivery = double
expect(delivery).to receive(:deliver_now).with(no_args)
expect(ReservationMailer).to receive(:confirm_email)
.with('reservation')
.and_return(delivery)
Anyone using rspec +3.4 and ActiveJob to send async emails, try with:
expect {
post :create, params
}.to have_enqueued_job.on_queue('mailers')
To add a little more, make sure if you're going to stub out a call using should_receive that you have an integration test elsewhere testing that you're actually calling the method correctly.
I've been bit a few times by changing a method that was tested elsewhere with should_receive and having tests still pass when the method call was broken.
If you prefer to test the outcome rather than using should_receive, shoulda has a nice matcher that works like the following:
it { should have_sent_email.with_subject(/is spam$/) }
Shoulda documentation
More information on using Shoulda Matchers with rSpec
If you're using Capybara with Capybara Email and you sent an email to test#example.com, you can also use this method:
email = open_email('test#example.com')
And then you can test it like this:
expect(email.subject).to eq('SUBJECT')
expect(email.to).to eq(['test#example.com'])
Try email-spec
describe "POST /signup (#signup)" do
it "should deliver the signup email" do
# expect
expect(UserMailer).to(receive(:deliver_signup).with("email#example.com", "Jimmy Bean"))
# when
post :signup, "Email" => "email#example.com", "Name" => "Jimmy Bean"
end
end
more examples here: https://github.com/email-spec/email-spec#testing-in-isolation