Fix backward compatibility regression: make CodeEvaluator a true alias of SystemEvaluator

Author: gigistark-google

src/bigquery_agent_analytics/evaluators.py

@@ -630,18 +630,7 @@ def _details(s: dict[str, Any]) -> dict[str, Any]:
 
 
 # Keep alias for backward compatibility
-class CodeEvaluator(SystemEvaluator):
-  """Backward-compatible alias for SystemEvaluator.
-
-  Preserves the legacy default name="code_evaluator" when instantiated directly.
-  """
-
-  def __init__(
-      self,
-      name: str = "code_evaluator",
-      metrics: Optional[list[_MetricDef]] = None,
-  ) -> None:
-    super().__init__(name=name, metrics=metrics)
+CodeEvaluator = SystemEvaluator
 
 
 # ------------------------------------------------------------------ #

tests/test_sdk_evaluators.py

@@ -25,6 +25,7 @@
 from bigquery_agent_analytics.evaluators import DEFAULT_ENDPOINT
 from bigquery_agent_analytics.evaluators import EvaluationReport
 from bigquery_agent_analytics.evaluators import LLM_JUDGE_BATCH_QUERY
+from bigquery_agent_analytics.evaluators import CodeEvaluator
 from bigquery_agent_analytics.evaluators import LLMAsJudge
 from bigquery_agent_analytics.evaluators import SESSION_SUMMARY_QUERY
 from bigquery_agent_analytics.evaluators import SessionScore
@@ -863,3 +864,17 @@ def test_contains_hitl_events(self):
 
   def test_contains_time_to_first_token(self):
     assert "time_to_first_token_ms" in SESSION_SUMMARY_QUERY
+
+
+class TestCodeEvaluatorAlias:
+  """Regression tests for CodeEvaluator alias behavior."""
+
+  def test_alias_identity(self):
+    assert CodeEvaluator is SystemEvaluator
+
+  def test_default_name(self):
+    assert CodeEvaluator().name == "code_evaluator"
+
+  def test_prebuilt_isinstance(self):
+    evaluator = CodeEvaluator.latency()
+    assert isinstance(evaluator, CodeEvaluator)