feat(flags): Add integration for custom tracking of flag evaluations#3860
Conversation
❌ 1 Tests Failed:
View the top 1 failed tests by shortest run time
To view more test analytics, go to the Test Analytics Dashboard |
320e49e to
4651b6a
Compare
|
|
||
|
|
||
| @pytest.fixture | ||
| def uninstall_integration(): |
There was a problem hiding this comment.
Note on this fixture: I found using the reset_integrations fixture in conftest.py caused issues with the default dedupe integration. We just want the integration being tested to be setup at the start of every new test.
Without using this, the error processor isn't added correctly. This isn't a concern for real use, since user apps only call init once
|
We also need to add docs for this, can you prepare a PR in the docs repo? |
|
Sounds good! I've opened getsentry/sentry-docs#12152, will do a last check in the morning to make sure it's up to date! |
|
Do you know anything about the failing ubuntu tests @antonpirker ? |
In Python 3.6 there is probably no Can you fix this somehow, or skip this test in Python 3.6 (then this integration will only be supported form Python 3.7 and up) |
Follow-up to
Adds an integration for buffering feature flags manually with an API (add_feature_flag), and automatically capturing them on error events.
Adds improved test coverage to launchdarkly and openfeature by testing end to end with
capture_eventsfixture, rather than asserting the flags are stored to scope.