Update app.py

app.py

@@ -126,16 +126,22 @@ def rag_workflow(query):
     rag_chain = RAGChain(llm, vector_store)
 
     # Step 1: Predict which library usage is relevant
-    library_usage_prediction = rag_chain.predict_library_usage(query)
+    # predict which python library to search in: (standard) kadiAPY-library or kadiAPY-cli-library 
+    code_library_usage_prediction = rag_chain.predict_library_usage(query)
     print(f"Predicted library usage: {library_usage_prediction}")
 
+    
     # Step 2: Retrieve contexts (documents and code snippets)
-    doc_contexts, code_contexts = rag_chain.retrieve_contexts(query, library_usage_prediction)
+    kadiAPY_doc_documents, kadiAPY_code_documents = rag_chain.retrieve_contexts(query, library_usage_prediction)
     print("Retrieved Document Contexts:", doc_contexts)
     print("Retrieved Code Contexts:", code_contexts)
 
+    
+    formatted_doc_snippets = rag_chain.format_documents(kadiAPY_doc_documents)
+    formatted_code_snippets = rag_chain.format_documents(kadiAPY_code_documents)
+
     # Step 4: Generate the final response
-    response = rag_chain.generate_response(query, formatted_doc_context, formatted_code_context)
+    response = rag_chain.generate_response(query, formatted_doc_snippets, formatted_code_snippets)
     print("Generated Response:", response)
 
     return response