The failing test here[1] does two things:
1. makes a request to /sign-out
2. calls the index route, without actually making a request
This means that when the `login_manager.unauthorized_handler`[2] looks
at Flask’s `request` object it gets the request context from 1., not 2.,
because 2. isn’t actually a request. The means it sets the value of the
`next` parameter to that of the request, not of the index route.
Basically at some point Flask has changed and decided that 2. isn’t a
proper request, so won’t set new request context.
This isn’t a realistic test because nothing would call the index
function directly, it would always be as part of a request to that page.
But to make the minimal change to fix the breaking tests this commit
makes the check a bit more general, i.e. that the redirect is to the
sign in page with any `next` parameter, not a specific `next` parameter.
1. 9111a7fc86/tests/app/utils/test_user.py (L130-L138)
2. 9111a7fc86/app/main/views/sign_in.py (L86)
We have a `client_request` fixture which does a bunch of useful stuff
like:
- checking the status code of the response
- returning a `BeautifulSoup` object
Lots of our tests still use an older fixture called `client`. This is
not as good because it:
- returns a raw `Response` object
- doesn’t do the additional checks
- means our tests contain a lot of repetetive boilerplate like `page = BeautifulSoup(response.data.decode('utf-8'), 'html.parser')`
This commit converts all the tests which had a `client.login(…)`
statement to use `client_request` (which is already logged in by
default).
Subsequent commits will remove uses of `client` in other tests, but
doing it this way means the work can be broken up into more manageable
chunks.
Previously these were lumped together with integration-level tests
for specific endpoints, which test the decorator was applied to the
endpoint in question.
