Why isn't my test picking up environment variables?
For your app to use environment variables from .env.test, make sure you are accessing them AFTER the test has a chance to read the variables out of .env.test.
If your application code defines your_constant = os.getenv('YOUR_ENV_VAR') outside of any function, it will only evaluate once, when the file is first imported (before your test has set up the environment).
You should either read the values during create_app, or only read the environment variables right before you need them.
Why do my tests pass in VS Code's test runner, but not when running "pytest" in a terminal?
VS Code actually reads your .env file automatically, so you may be relying on that fact. Check that you are loading environment variables during the test (fixture) and that your code accesses the environment variables afterwards. See the above question and answer.
Why am I getting an import error or "Starting path not found" error?
Your tests folder needs to be recognised as a Python package: add an empty __init__.py file to any test directories.
Why isn't my integration test using my mock function? I'm still seeing a JSON decode error.
Say you have a file, "trello_items.py", which imports requests in this way: from requests import get.
Patching requests.get (e.g. monkeypatch.setattr(requests, 'get', stub)) will have no effect - trello_items already has a reference to the real "get" function.
Either import the whole module instead (use import requests in trello_items.py) or patch the function you added to your trello_items module (monkeypatch.setattr(trello_items, 'get', stub)).
Why is my stub function giving me a TypeError about the arguments?
The exact usage of your "stub" function depends on how your application code is trying to use requests.
If your application has a line of code like requests.get(my_url, headers=my_headers), then it must be valid to call stub(my_url, headers=my_headers). Check that your definition of stub matches its usage.
Why is my selenium test is failing?
-
If your app depends on environment variables for each list ID, then your selenium test will also need to override those variables with the IDs for the temporary test board's lists.
-
Check that your app is using environment variables correctly (not too early and no IDs are hardcoded strings).