feat: Support segment metadata (#264)

Author: khvn26

.gitmodules

@@ -1,4 +1,4 @@
 [submodule "tests/engine_tests/engine-test-data"]
 	path = tests/engine_tests/engine-test-data
 	url = https://github.com/flagsmith/engine-test-data.git
-	tag = v2.0.0
+	tag = v2.1.0

flag_engine/context/types.py

@@ -1,6 +1,6 @@
 # generated by datamodel-codegen:
-#   filename:  https://raw.githubusercontent.com/Flagsmith/flagsmith/main/sdk/evaluation-context.json  # noqa: E501
-#   timestamp: 2025-08-25T11:10:31+00:00
+#   filename:  https://raw.githubusercontent.com/Flagsmith/flagsmith/main/sdk/evaluation-context.json
+#   timestamp: 2025-10-10T16:01:57+00:00
 
 from __future__ import annotations
 
@@ -63,6 +63,7 @@ class SegmentContext(TypedDict):
     name: str
     rules: List[SegmentRule]
     overrides: NotRequired[List[FeatureContext]]
+    metadata: NotRequired[Dict[str, Any]]
 
 
 class EvaluationContext(TypedDict):

flag_engine/result/types.py

@@ -1,11 +1,13 @@
 # generated by datamodel-codegen:
-#   filename:  https://raw.githubusercontent.com/Flagsmith/flagsmith/refs/heads/chore/improve-evaluation-result/sdk/evaluation-result.json  # noqa: E501
-#   timestamp: 2025-10-01T11:31:34+00:00
+#   filename:  https://raw.githubusercontent.com/Flagsmith/flagsmith/refs/heads/main/sdk/evaluation-result.json
+#   timestamp: 2025-10-10T16:01:29+00:00
 
 from __future__ import annotations
 
 from typing import Any, Dict, List, TypedDict
 
+from typing_extensions import NotRequired
+
 
 class FlagResult(TypedDict):
     feature_key: str
@@ -18,6 +20,7 @@ class FlagResult(TypedDict):
 class SegmentResult(TypedDict):
     key: str
     name: str
+    metadata: NotRequired[Dict[str, Any]]
 
 
 class EvaluationResult(TypedDict):

flag_engine/segments/evaluator.py

@@ -48,12 +48,13 @@ def get_evaluation_result(context: EvaluationContext) -> EvaluationResult:
         if not is_context_in_segment(context, segment_context):
             continue
 
-        segments.append(
-            {
-                "key": segment_context["key"],
-                "name": segment_context["name"],
-            }
-        )
+        segment_result: SegmentResult = {
+            "key": segment_context["key"],
+            "name": segment_context["name"],
+        }
+        if metadata := segment_context.get("metadata"):
+            segment_result["metadata"] = metadata
+        segments.append(segment_result)
 
         if overrides := segment_context.get("overrides"):
             for override_feature_context in overrides:

tests/engine_tests/engine-test-data

@@ -40,7 +40,10 @@ def _extract_benchmark_contexts(
         yield pyjson5.loads((test_cases_dir_path / file_path).read_text())["context"]
 
 
-TEST_CASES = list(_extract_test_cases(TEST_CASES_PATH))
+TEST_CASES = sorted(
+    _extract_test_cases(TEST_CASES_PATH),
+    key=lambda param: str(param.id),
+)
 BENCHMARK_CONTEXTS = list(_extract_benchmark_contexts(TEST_CASES_PATH))