Update app.py

app.py

@@ -128,19 +128,28 @@ class PredictResponse(BaseModel):
     soreness_conf: float
     exercises:     List[ExerciseResponse]
 
+def format_pg_array(values: list) -> str:
+    """Convert a Python list to PostgreSQL array literal format"""
+    return '{' + ','.join(str(v) for v in values) + '}'
+
+    
 # ── Supabase Helper ───────────────────────────────────────────────────────────
 def get_suitable_exercises(workout_type: int, mood: int, soreness: int) -> List[ExerciseResponse]:
     try:
+        logger.info(f"Querying exercises — workout_type: {workout_type}, mood: {mood}, soreness: {soreness}")
+
         print("DATA")
-        print(workout_type,mood,soreness)
         response = (
             state.supabase.table('exerciseai')
             .select('*')
             .eq('workout_type', workout_type)
-            .contains('suitable_moods',    [string(mood)])
-            .contains('suitable_soreness', [string(soreness)])
+            .contains('suitable_moods',    format_pg_array([mood]))
+            .contains('suitable_soreness', format_pg_array([soreness]))
             .execute()
         )
+
+        logger.info(f"Supabase returned {len(response.data)} exercises")
+
         return [ExerciseResponse(**exercise) for exercise in response.data]
 
     except Exception as e:
@@ -159,7 +168,7 @@ def health_check():
 # ── Predict Endpoint ──────────────────────────────────────────────────────────
 @app.post("/predict", response_model=PredictResponse)
 def predict(request: PredictRequest):
-
+    print("HERE")
     # ── Input validation ──────────────────────────────────────────────────────
     if not request.user_input.strip():
         raise HTTPException(status_code=400, detail="user_input cannot be empty")
@@ -182,7 +191,7 @@ def predict(request: PredictRequest):
             workout_logits, mood_logits, soreness_logits = state.model(
                 input_ids, attention_mask
             )
-
+        print("HERE2")
         # ── Softmax + confidence ──────────────────────────────────────────────
         workout_probs  = torch.softmax(workout_logits,  dim=1)
         mood_probs     = torch.softmax(mood_logits,     dim=1)