feat(Segment Membership): Skip traitless identities when seeding

api/segment_membership/tasks.py

@@ -1,5 +1,5 @@
 from datetime import timedelta
-from typing import cast
+from typing import Any, cast
 
 import structlog
 from django.conf import settings
@@ -61,6 +61,9 @@ def seed_organisation_identities(organisation_id: int) -> None:
 
     Rows are versioned at scan start via `inserted_at`
     so writes arriving mid-scan win ReplacingMergeTree dedup over the seeded row.
+
+    Identities carrying no traits are skipped as they carry little to no value
+    for segment membership.
     """
     log = logger.bind(organisation__id=organisation_id)
     if not settings.CLICKHOUSE_ENABLED:
@@ -104,7 +107,10 @@ def seed_organisation_identities(organisation_id: int) -> None:
                                     scan_started_at,
                                 )
                                 for doc in batch
+                                if not _is_empty_identity(doc)
                             ]
+                            if not rows:
+                                continue
                             # Django's CursorWrapper stub forbids dicts in
                             # the params sequence; clickhouse-driver accepts
                             # them as JSON-column payloads.
@@ -269,3 +275,10 @@ def refresh_project_segment_counts(project_id: int) -> None:
             membership_counts__count=len(membership_counts),
             stale_counts__count=stale_deleted,
         )
+
+
+def _is_empty_identity(identity_document: dict[str, Any]) -> bool:
+    return not (
+        identity_document.get("identity_traits")
+        or identity_document.get("system_traits")
+    )

api/tests/integration/segment_membership/test_segment_membership_clickhouse.py

@@ -102,7 +102,7 @@ def test_seed_organisation_identities__happy_path__rows_land_in_clickhouse(
                 "composite_key": "k2",
                 "environment_api_key": environment_api_key,
                 "created_date": "2026-05-08T00:00:00Z",
-                "identity_traits": [],
+                "identity_traits": [{"trait_key": "foo2", "trait_value": "bar2"}],
             },
         ]
     )

api/tests/unit/segment_membership/test_unit_segment_membership_tasks.py

@@ -151,6 +151,54 @@ def test_seed_organisation_identities__insert_fails__logs_and_continues(
     ]
 
 
+def test_seed_organisation_identities__traitless_identities__are_not_mirrored(
+    mocker: MockerFixture,
+    settings: SettingsWrapper,
+    project: Project,
+    environment: Environment,
+    segment: Segment,
+    flagsmith_identities_table: Table,
+    dynamo_identities: None,
+    enable_features: EnableFeaturesFixture,
+) -> None:
+    # Given
+    enable_features("segment_membership_inspection")
+    settings.CLICKHOUSE_ENABLED = True
+    mocker.patch.object(tasks, "_INSERT_BATCH_SIZE", 1)
+
+    cursor = MagicMock()
+    open_cursor = mocker.patch.object(tasks, "open_clickhouse_cursor")
+    open_cursor.return_value.__enter__.return_value = cursor
+    mocker.patch.object(tasks, "enqueue_membership_refresh")
+
+    for identifier, extra in (
+        ("dave", {}),
+        ("erin", {"system_traits": {"flagsmith_cohort_e2b1": True}}),
+    ):
+        flagsmith_identities_table.put_item(
+            Item={
+                "composite_key": f"{environment.api_key}_{identifier}",
+                "environment_api_key": environment.api_key,
+                "identifier": identifier,
+                "identity_uuid": f"f47ac10b-58cc-4372-a567-0e02b2c3d4{identifier[:2]}",
+                "identity_traits": [],
+                **extra,
+            }
+        )
+
+    # When
+    seed_organisation_identities(project.organisation_id)
+
+    # Then
+    insert_identity_payloads = [
+        call.args[1][0] for call in cursor.executemany.call_args_list
+    ]
+    inserted_identifiers = sorted(payload[1] for payload in insert_identity_payloads)
+
+    # 'dave' is not included in the list of inserted identities
+    assert inserted_identifiers == ["alice", "carol", "erin"]
+
+
 @pytest.mark.clickhouse
 def test_seed_organisation_identities__matching_identities__inserts_rows_versioned_at_scan_start(
     mocker: MockerFixture,

docs/docs/deployment-self-hosting/observability/_events-catalogue.md

@@ -618,7 +618,7 @@ Attributes:
 ### `segment_membership.refresh.project.completed`
 
 Logged at `info` from:
- - `api/segment_membership/tasks.py:266`
+ - `api/segment_membership/tasks.py:272`
 
 Attributes:
  - `membership_counts.count`
@@ -628,16 +628,16 @@ Attributes:
 ### `segment_membership.refresh.project.failed`
 
 Logged at `exception` from:
- - `api/segment_membership/tasks.py:239`
+ - `api/segment_membership/tasks.py:245`
 
 Attributes:
  - `project.id`
 
 ### `segment_membership.refresh.project.skipped`
 
 Logged at `info` from:
- - `api/segment_membership/tasks.py:206`
- - `api/segment_membership/tasks.py:218`
+ - `api/segment_membership/tasks.py:212`
+ - `api/segment_membership/tasks.py:224`
 
 Attributes:
  - `project.id`
@@ -647,7 +647,7 @@ Attributes:
 ### `segment_membership.seed.environment.completed`
 
 Logged at `info` from:
- - `api/segment_membership/tasks.py:121`
+ - `api/segment_membership/tasks.py:127`
 
 Attributes:
  - `environment.id`
@@ -658,7 +658,7 @@ Attributes:
 ### `segment_membership.seed.environment.failed`
 
 Logged at `exception` from:
- - `api/segment_membership/tasks.py:114`
+ - `api/segment_membership/tasks.py:120`
 
 Attributes:
  - `environment.id`
@@ -668,9 +668,9 @@ Attributes:
 ### `segment_membership.seed.skipped`
 
 Logged at `warning` from:
- - `api/segment_membership/tasks.py:67`
- - `api/segment_membership/tasks.py:72`
- - `api/segment_membership/tasks.py:77`
+ - `api/segment_membership/tasks.py:70`
+ - `api/segment_membership/tasks.py:75`
+ - `api/segment_membership/tasks.py:80`
 
 Attributes:
  - `organisation.id`