-
Notifications
You must be signed in to change notification settings - Fork 462
feat(api): add experimental feature flag update endpoints #6305
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: main
Are you sure you want to change the base?
Conversation
Adds two new experimental endpoints under /api/experiments/: - update-flag-v1: Update single feature state (environment default or segment override) - update-flag-v2: Update multiple feature states in a single request (environment default + segment overrides)
|
The latest updates on your projects. Learn more about Vercel for GitHub. 3 Skipped Deployments
|
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #6305 +/- ##
==========================================
+ Coverage 98.02% 98.05% +0.03%
==========================================
Files 1280 1289 +9
Lines 45406 46141 +735
==========================================
+ Hits 44508 45243 +735
Misses 898 898 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
- Parameterize test_update_flag_by_name for all value types (integer, string, boolean) - Remove unsupported float type from serializer (model doesn't support it) - Add test_serializers.py with unit tests for edge cases - Add test for duplicate segment_id validation - Add tests for updating existing segment overrides with priority - Add unit tests for boolean string conversion and replica path - Refactor _update_flag_for_versioning_v2 to separate segment vs environment default cases for cleaner code and full coverage
Replace SDK endpoint calls with direct database verification using get_environment_flags_list. This keeps the tests as proper unit tests rather than integration tests. - Remove sdk_client and sdk_client_factory fixtures - Remove get_identity_feature_state_from_sdk helper function - Remove Identity, Condition, SegmentRule imports and usage - Simplify segment override tests to verify database state directly
f280c7f to
b987697
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is the final PR Bugbot will review for you during this billing cycle
Your free Bugbot reviews will reset on December 10
Details
Your team is on the Bugbot Free tier. On this plan, Bugbot will review limited PRs each billing cycle for each member of your team.
To receive Bugbot reviews on all of your PRs, visit the Cursor dashboard to activate Pro and start your 14-day free trial.
…ning The previous filter Q(feature_segment__segment_id=None) relies on LEFT OUTER JOIN behavior to match NULL FKs. Using Q(feature_segment__isnull=True) is more explicit and robust as it directly checks the FK column.
The _update_feature_state_value function was setting new typed values without clearing old value fields when changing types. This left stale data in the database. Added set_value() method to AbstractBaseFeatureValueModel that centralizes validation, type conversion, and ensures unused fields are cleared.
The filter Q(feature_segment__isnull=True) doesn't exclude identity overrides, which also have feature_segment=None. Added identity_id__isnull=True to correctly retrieve only the environment default state.
6fb8e9b to
e998a78
Compare
e998a78 to
56443a0
Compare
…rides Creating a FeatureSegment with a priority value doesn't trigger reordering of existing segments. Use .to() after creation to properly reorder. Fixes both V1 and V2 endpoints.
56443a0 to
8d43e11
Compare
…ride Prevents cross-project segment references and provides clear error messages for non-existent segments.
emyller
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The experiment looks solid, despite the complexity brought by the 2x2 grid experiment x versioning version. Few suggestions.
|
|
||
| def update_traits( | ||
| self, | ||
| # TODO: this typing isn't true as it also supports regular dicts like {"trait_key": "foo", "trait_value": "bar"} |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Can we have an issue to cover this?
| @property | ||
| def environment(self) -> Environment: | ||
| environment: Environment | None = self.context.get("environment") | ||
| if not environment: | ||
| raise serializers.ValidationError("Environment context is required") | ||
| return environment |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I feel like this could be a mixin to reuse in many other views, though I appreciate it's out of scope.
| if value_type == "integer": | ||
| try: | ||
| int(string_val) | ||
| except ValueError: | ||
| raise serializers.ValidationError( | ||
| f"'{string_val}' is not a valid integer" | ||
| ) | ||
| elif value_type == "boolean": | ||
| if string_val.lower() not in ("true", "false"): | ||
| raise serializers.ValidationError( | ||
| f"'{string_val}' is not a valid boolean (use 'true' or 'false')" | ||
| ) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Can we have this logic abstracted? It is repeated but raising distinct exceptions, e.g. ValueError there, and ValidationError here.
| ) | ||
| serializer.is_valid() | ||
|
|
||
| with pytest.raises(Exception) as exc_info: |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is it possible to narrow down this exception type?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Missing Given/When/Then markers.
| ) | ||
|
|
||
|
|
||
| def update_flag( |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
These functions perform multiple database writes without @transaction.atomic. Can we wrap them?
|
|
||
| updated_segments = [] | ||
| for override in change_set.segment_overrides: | ||
| segment_states = get_environment_flags_dict( |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
get_environment_flags_dict is called once per segment override. Can we batch this into a single query?
| feature_name=feature.name, | ||
| additional_filters=Q(feature_segment__isnull=True, identity_id__isnull=True), | ||
| ) | ||
| assert len(env_default_states) == 1 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
assert provides no message when it fails. Can we raise an explicit exception instead?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Can we have a test for concurrent modifications? This is relevant since atomicity is not implemented yet (see other thread).
Thanks for submitting a PR! Please check the boxes below:
docs/if required so people know about the feature!Changes
Adds two new experimental endpoints under /api/experiments/:
Related issues: Create new experimental endpoint to update a single feature state for a given feature #6279 Create new experimental endpoints to update a given feature in an environment #6233
How did you test this code?
Unit tests