How to test the response code with Capybara + Selenium

Question

I have the following spec:

it "deletes post", :js => true do 
...
...
page.status_code.should = '404'

end

The page.status_code line is giving me this error:

Capybara::NotSupportedByDriverError

How do I check the page's status code?

related: https://groups.google.com/forum/?fromgroups=#!topic/ruby-capybara/KdMvhjcjm8E — tokland, Oct 12 '12 at 10:29
Possible duplicate of [How to get HTTP Response Code using Selenium WebDriver with Java?](https://stackoverflow.com/questions/6509628/how-to-get-http-response-code-using-selenium-webdriver-with-java) — David Moles, Jun 06 '18 at 21:42

score 47 · Answer 1 · answered Mar 06 '13 at 10:27

47

As an aside. This line

page.status_code.should = '404'

Should be

page.status_code.should == 404

This worked for me with capybara-webkit.

answered Mar 06 '13 at 10:27

Apie

6,091
6
23
24

11

another capybara-webkit user is here but `expect(page.status_code).to be(200)` returns `Capybara::NotSupportedByDriverError`. Do you have any idea why? – scaryguy Sep 10 '16 at 02:43

score 19 · Accepted Answer · answered Oct 27 '11 at 07:52

19

status_code is not currently supported by the Selenium driver. You will need to write a different test to check the response status code.

answered Oct 27 '11 at 07:52

eugen

8,170
11
54
63

score 3 · Answer 3 · answered Oct 23 '14 at 14:00

Selenium web driver doest not implement status_code and there is no direct way to test response_code with selenium (developer's choice).

To test it I added in my layout/application.html.erb:

<html code="<%= response.try(:code) if defined?(response) %>">[...]</html>

And then in my test:

def no_error?
  response_code = page.first('html')[:code]
  assert (response_code == '200'), "Response code should be 200 : got #{response_code}"
end

score 3 · Answer 4 · answered Oct 27 '11 at 08:13

3

Either switch to another driver (like rack-test) for that test, or test that the displayed page is the 404 page (should have content 'Not Found' in h1).

As @eugen said, Selenium doesn't support status codes.

answered Oct 27 '11 at 08:13

Leonid Shevtsov

13,290
8
48
80

score 2 · Answer 5 · edited Nov 30 '16 at 12:15

2

Try it out

expect(page).to have_http_status(200)

edited Nov 30 '16 at 12:15

Ingo Karkat

154,018
15
205
275

answered Nov 30 '16 at 11:47

Ajay

97
3

7

Hi Ajay; your code might be correct, but with some context it would make a better answer; for example, you could explain how and why this proposed change would resolve the questioner's problem, perhaps including a link to the relevant documentation. That would make it more useful to them, and also more useful to other site readers who are looking for solutions to similar problems. – Vince Bowdren Nov 30 '16 at 13:33
1

I get `expected # to respond to 'has_http_status?' (RSpec::Expectations::ExpectationNotMetError)` – lafeber Jun 17 '19 at 13:43
Note: This isn't provided by `Capybara`, but by `RSpec`. https://relishapp.com/rspec/rspec-rails/docs/matchers/have-http-status-matcher – XtraSimplicity Jul 09 '19 at 21:42

score 0 · Answer 6 · answered Jan 12 '17 at 01:38

Use js to make a request and get status as below:

js_script = <<JSS
xhr = new XMLHttpRequest();
xhr.open('GET', '#{src}', true);
xhr.send();
JSS
actual.execute_script(js_script)
status = actual.evaluate_script('xhr.status') # get js variable value

Check https://gist.github.com/yovasx2/1c767114f2e003474a546c89ab4f90db for more details

How to test the response code with Capybara + Selenium

6 Answers6