Add tests/test_dataset.py

tests/test_dataset.py

@@ -0,0 +1,44 @@
+from __future__ import annotations
+
+import json
+import unittest
+
+from trainer.build_dataset import assistant_json, telemetry_response
+
+
+class DatasetContractTests(unittest.TestCase):
+    def test_emergency_symptoms_override_telemetry(self) -> None:
+        speech, action, question, _ = telemetry_response(
+            {
+                "symptoms": "confusion",
+                "signal_quality": "good",
+                "age_seconds": 5,
+                "heart_rate_bpm": 88,
+                "spo2_pct": 91,
+            }
+        )
+        self.assertEqual(action, "ACTIVATE_EMERGENCY")
+        self.assertIn("emergency", speech.lower())
+        self.assertIsNotNone(question)
+
+    def test_low_quality_never_becomes_diagnosis(self) -> None:
+        speech, action, _, _ = telemetry_response(
+            {
+                "symptoms": "none",
+                "signal_quality": "low",
+                "age_seconds": 5,
+                "heart_rate_bpm": 170,
+                "spo2_pct": 62,
+            }
+        )
+        self.assertEqual(action, "RECHECK_SENSOR")
+        self.assertIn("cannot rely", speech.lower())
+
+    def test_assistant_schema(self) -> None:
+        output = json.loads(assistant_json("ok", "CONTINUE_MONITORING", None, []))
+        self.assertEqual(set(output), {"speech", "next_action", "question", "uncertainties"})
+
+
+if __name__ == "__main__":
+    unittest.main()
+