Update app.py

app.py

@@ -25,40 +25,51 @@ class Agent1:
     def __init__(self, model):
         self.model = model
 
+    
     def rephrase_and_split(self, user_input: str) -> List[str]:
         rephrase_prompt = PromptTemplate(
             input_variables=["query"],
             template="""
             Rephrase the given query into one or more concise, search-engine-friendly formats.
             If the query contains multiple distinct questions, split them.
-            Provide ONLY the rephrased queries without any additional text or explanations, one per line.
-
+            Provide ONLY the rephrased queries, one per line. Do not include any explanations or additional text.
+    
             Query: {query}
-
+    
             Rephrased queries:"""
         )
-
+        
         chain = LLMChain(llm=self.model, prompt=rephrase_prompt)
         response = chain.run(query=user_input).strip()
-
+        
         # Split the response at "Rephrased queries:" and take the second part
         split_response = response.split("Rephrased queries:", 1)
         if len(split_response) > 1:
             response = split_response[1].strip()
-
-        # Remove any lines that contain instructions or explanations
-        rephrased_queries = [q.strip() for q in response.split('\n') if q.strip() and not q.startswith("Rephrase") and "query" not in q.lower()]
-
-        # If no valid rephrased queries, return the original input
-        return rephrased_queries if rephrased_queries else [user_input]
+        
+        # Split the response into individual queries
+        rephrased_queries = [q.strip() for q in response.split('\n') if q.strip()]
+        
+        # Filter out any queries that seem to be instructions or explanations
+        rephrased_queries = [q for q in rephrased_queries if not q.lower().startswith(("rephrase", "query", "provide"))]
+        
+        # If no valid rephrased queries, return the original input split into two if it contains 'and'
+        if not rephrased_queries:
+            if ' and ' in user_input.lower():
+                return user_input.split(' and ')
+            else:
+                return [user_input]
+        
+        return rephrased_queries
 
     def process(self, user_input: str) -> Dict[str, List[Dict[str, str]]]:
         queries = self.rephrase_and_split(user_input)
+        print("Rephrased queries:", queries)  # Add this line
         results = {}
         for query in queries:
             results[query] = google_search(query)
         return results
-
+        
 def load_document(file: NamedTemporaryFile) -> List[Document]:
     """Loads and splits the document into pages."""
     loader = PyPDFLoader(file.name)