Update app.py

app.py

@@ -633,31 +633,71 @@ async def resources_endpoint(profile: MedicalProfile):
 @app.post("/api/recipes")
 async def recipes_endpoint(profile: MedicalProfile):
     try:
-        recipe_query = f"Recipes and meals suitable for someone with: {', '.join(profile.chronic_conditions + profile.food_restrictions)}"
-        query_text = recipe_query      
-        query_embedding = embed_query_text(query_text)  # Embed the query text
-        embeddings_data = load_embeddings ()
+        # Build the query text for recipes
+        recipe_query = (
+            f"Recipes and meals suitable for someone with: "
+            f"{profile.conditions} and experiencing {profile.daily_symptoms}"
+        )
+        query_text = recipe_query
+
+        # Generate the query embedding
+        query_embedding = embed_query_text(query_text)
+        if query_embedding is None:
+            raise ValueError("Failed to generate query embedding.")
+
+        # Load embeddings and retrieve initial results
+        embeddings_data = load_embeddings()
         folder_path = 'downloaded_articles/downloaded_articles'
         initial_results = query_embeddings(query_embedding, embeddings_data, n_results=5)
+        if not initial_results:
+            raise ValueError("No relevant recipes found.")
+
+        # Extract document IDs
         document_ids = [doc_id for doc_id, _ in initial_results]
+
+        # Retrieve document texts
         document_texts = retrieve_document_texts(document_ids, folder_path)
+        if not document_texts:
+            raise ValueError("Failed to retrieve document texts.")
+
+        # Perform re-ranking with cross-encoder
         cross_encoder = models['cross_encoder']
         scores = cross_encoder.predict([(query_text, doc) for doc in document_texts])
+        scores = [float(score) for score in scores]  # Convert scores to native floats
+
+        # Combine document data
         scored_documents = list(zip(scores, document_ids, document_texts))
-        ranked_docs = scored_documents.sort(key=lambda x: x[0], reverse=True)
+        scored_documents.sort(key=lambda x: x[0], reverse=True)  # Sort by score
+
+        # Load recipe metadata from DataFrame
+        file_path = 'finalcleaned_excel_file.xlsx'
+        df = pd.read_excel(file_path)
+
+        # Prepare the final recipes list
         recipes = []
-        for (doc_id, _), score, text in ranked_docs[:10]:
-            doc_info = data['df'][data['df']['id'] == doc_id].iloc[0]
-            if 'recipe' in text.lower() or 'meal' in text.lower():
-                recipes.append({
-                    "id": doc_id,
-                    "title": doc_info['title'],
-                    "content": text[:200],
-                    "score": float(score)
-                })
+        for score, doc_id, text in scored_documents:
+            # Retrieve metadata for the document
+            doc_info = df[df['Unnamed: 0'] == doc_id]
+            if not doc_info.empty:
+                title = doc_info.iloc[0]['title'] if 'title' in doc_info.columns else "Unknown Title"
+                if 'recipe' in text.lower() or 'meal' in text.lower():
+                    recipes.append({
+                        "id": doc_id,
+                        "title": title,
+                        "content_preview": text[:200],  # First 200 characters of text
+                        "score": score,
+                    })
+
+        # Limit the response to top 5 recipes
         return {"recipes": recipes[:5], "success": True}
+
+    except ValueError as ve:
+        # Handle expected errors
+        raise HTTPException(status_code=400, detail=str(ve))
     except Exception as e:
-        raise HTTPException(status_code=500, detail=str(e))
+        # Handle unexpected errors
+        print(f"Unexpected error: {e}")
+        raise HTTPException(status_code=500, detail="An unexpected error occurred.")
 
 if not init_success:
     print("Warning: Application initialized with partial functionality")