fix(simple_metrics): fail the run explicitly when the adaptive state update fails (SIENTIAPDE-2191)

Drop the try/except around update_simple_metrics_adaptive_state in the workflow and the broad except in the activity, so Mongo failures fail the run and show up in the Temporal metrics (PR #59 review). Invalid frequency_alert stays a warning + skip (the orchestrator ignores it too) and a malformed adaptive_state is read as not escalated.

tests/laborious/activities/test_model_metrics.py

@@ -1424,16 +1424,33 @@ def test_adaptive_state_unchanged_does_not_write(model_metrics_activity, mongo):
     mongo.update_one.assert_not_called()
 
 
-def test_adaptive_state_invalid_frequency_alert_is_swallowed(model_metrics_activity, mongo):
-    mongo.find_one.return_value = _pipeline(frequency_alert='5min')
+@mark.parametrize('frequency_alert', ['5min', '0m', 5])
+def test_adaptive_state_invalid_frequency_alert_is_skipped(
+    model_metrics_activity, mongo, frequency_alert
+):
+    mongo.find_one.return_value = _pipeline(frequency_alert=frequency_alert)
 
     assert model_metrics_activity.update_simple_metrics_adaptive_state(_BREACH_INPUT) is None
     mongo.update_one.assert_not_called()
-    model_metrics_activity.error.assert_called_once()
+    model_metrics_activity.warning.assert_called_once()
 
 
-def test_adaptive_state_mongo_error_is_swallowed(model_metrics_activity, mongo):
-    mongo.find_one.side_effect = RuntimeError('mongo down')
+@mark.parametrize('adaptive_state', ['garbage', ['escalated']])
+def test_adaptive_state_malformed_state_is_not_escalated(
+    model_metrics_activity, mongo, adaptive_state
+):
+    mongo.find_one.return_value = _pipeline(frequency_alert='5m', adaptive_state=adaptive_state)
 
-    assert model_metrics_activity.update_simple_metrics_adaptive_state(_BREACH_INPUT) is None
-    model_metrics_activity.error.assert_called_once()
+    result = model_metrics_activity.update_simple_metrics_adaptive_state(_BREACH_INPUT)
+
+    assert result == {'escalated': True, 'healthy_streak': 0}
+    mongo.update_one.assert_called_once()
+
+
+@mark.parametrize('failing_call', ['find_one', 'update_one'])
+def test_adaptive_state_mongo_error_propagates(model_metrics_activity, mongo, failing_call):
+    mongo.find_one.return_value = _pipeline(frequency_alert='5m')
+    getattr(mongo, failing_call).side_effect = RuntimeError('mongo down')
+
+    with raises(RuntimeError, match='mongo down'):
+        model_metrics_activity.update_simple_metrics_adaptive_state(_BREACH_INPUT)

tests/laborious/workflows/test_simple_metrics.py

@@ -1,6 +1,6 @@
 from unittest.mock import ANY, AsyncMock, MagicMock, call, patch
 
-from pytest import fixture, mark
+from pytest import fixture, mark, raises
 from sientia_do.temporal.constants import DATETIME_FORMAT_WITH_TZ
 
 from laborious.activities.activities import Activities
@@ -218,7 +218,7 @@ async def test_run_default_metrics(workflow_mock: AsyncMock, simple_metrics: Sim
     # Act
     await simple_metrics.run(input_data)
 
-    # Assert - all three activities ran (a swallowed exhausted side_effect would warn)
+    # Assert - all three activities ran
     assert workflow_mock.execute_activity_method.call_count == 3
     workflow_mock.logger.warning.assert_not_called()
 
@@ -402,13 +402,16 @@ async def test_run_below_min_samples_sends_no_rows_to_adaptive_state(workflow_mo
 
 @mark.asyncio
 @patch('laborious.workflows.simple_metrics.workflow', new_callable=AsyncMock)
-async def test_run_adaptive_state_failure_does_not_fail_workflow(workflow_mock, simple_metrics):
+async def test_run_adaptive_state_failure_fails_workflow(workflow_mock, simple_metrics):
     workflow_mock.logger = MagicMock()
     workflow_mock.execute_activity_method = AsyncMock(
         side_effect=[[_ROW] * 3, None, RuntimeError('dispatch failed')]
     )
     workflow_mock.execute_local_activity_method = AsyncMock(return_value=_BREACH_ROWS)
 
-    await simple_metrics.run(_input(simple_metrics_thresholds=_THRESHOLDS))
+    with raises(RuntimeError, match='dispatch failed'):
+        await simple_metrics.run(_input(simple_metrics_thresholds=_THRESHOLDS))
 
-    workflow_mock.logger.warning.assert_called_once()
+    # Metrics were exported before the failing call
+    calls = workflow_mock.execute_activity_method.call_args_list
+    assert calls[1].args[0] is Activities.export_data_to_postgres