Refactor integration tests to use persistent buckets #379
Conversation
test/integration/helpers.py
Outdated
|
|
||
|
|
||
| def _clean_and_delete_bucket(raw_api, api_url, account_auth_token, account_id, bucket_id): | ||
| # Delete the files. This test never creates more than a few files, |
There was a problem hiding this comment.
It seems like this is still tightly tied with raw-api tests - not sure why it was moved to general helpers if its only useful in raw-api tests.
You seem to have created a test_bucket fixture - wouldn it make better sense for it to do a cleanup of the bucket in there i.e.
def test_bucket(...):
...
yield bucket
_clean_and_delete_bucket(..., bucket)There was a problem hiding this comment.
This test? which test?
Context got lost when I moved it from test_raw_api.py. Will fix.
It's also called, albeit indirectly, by cleanup_old_buckets nox session to remove stale buckets. This is what warranted the move in the first place.
test/integration/conftest.py
Outdated
| application_key_id = os.environ.get('B2_TEST_APPLICATION_KEY_ID') | ||
| application_key = os.environ.get('B2_TEST_APPLICATION_KEY') | ||
| if application_key_id is None or application_key is None: | ||
| pytest.fail('B2_TEST_APPLICATION_KEY_ID or B2_TEST_APPLICATION_KEY is not set.') |
There was a problem hiding this comment.
use b2_auth_data fixture instead
I understand you're referring to the persistent bucket retrieval code. Currently, this code differs slightly between the two repositories. In the SDK, the code for deriving the persistent bucket name takes into account the arguments used to create a bucket. However, I haven't observed a similar need in the CLI. |
Even if the one here is slightly different, it is duplicated. But I guess it is a more general problem with entire integration test suite case and perhaps best to dealt with as separate task. edited: I see #379 (comment) was not yet address, so please create PR after doing that |
test/integration/test_raw_api.pyto use persistent bucket where applicable