Update app.py

app.py

@@ -329,51 +329,138 @@ def get_response_from_duckduckgo(query, model, context, num_calls=1, temperature
 
     yield full_response.strip()
 
-def respond(message, history, model, temperature, num_calls, selected_docs):
-    logging.info(f"User Query: {message}")
-    logging.info(f"Model Used: {model}")
-    logging.info(f"Selected Documents: {selected_docs}")
+class ConversationManager:
+    def __init__(self):
+        self.history = []
+        self.current_context = None
 
-    try:
-        embed = get_embeddings()
-        if os.path.exists("faiss_database"):
-            database = FAISS.load_local("faiss_database", embed, allow_dangerous_deserialization=True)
-            retriever = database.as_retriever(search_kwargs={"k": 20})
-            
-            all_relevant_docs = retriever.get_relevant_documents(message)
-            relevant_docs = [doc for doc in all_relevant_docs if doc.metadata["source"] in selected_docs]
-            
-            if not relevant_docs:
-                yield "No relevant information found in the selected documents. Please try selecting different documents or rephrasing your query."
-                return
+    def add_interaction(self, query, response):
+        self.history.append((query, response))
+        self.current_context = f"Previous query: {query}\nPrevious response summary: {response[:200]}..."
 
-            context_str = "\n".join([doc.page_content for doc in relevant_docs])
-            logging.info(f"Context length: {len(context_str)}")
-        else:
-            context_str = "No documents available."
-            yield "No documents available. Please upload PDF documents to answer questions."
-            return
+    def get_context(self):
+        return self.current_context
+
+conversation_manager = ConversationManager()
+
+def get_web_search_results(query: str, max_results: int = 10) -> List[Dict[str, str]]:
+    try:
+        results = list(DDGS().text(query, max_results=max_results))
+        if not results:
+            print(f"No results found for query: {query}")
+        return results
+    except Exception as e:
+        print(f"An error occurred during web search: {str(e)}")
+        return [{"error": f"An error occurred during web search: {str(e)}"}]
+
+def rephrase_query(original_query: str, conversation_manager: ConversationManager) -> str:
+    context = conversation_manager.get_context()
+    if context:
+        prompt = f"""You are a highly intelligent conversational chatbot. Your task is to analyze the given context and new query, then decide whether to rephrase the query with or without incorporating the context. Follow these steps:
+
+        1. Determine if the new query is a continuation of the previous conversation or an entirely new topic.
+        2. If it's a continuation, rephrase the query by incorporating relevant information from the context to make it more specific and contextual.
+        3. If it's a new topic, rephrase the query to make it more appropriate for a web search, focusing on clarity and accuracy without using the previous context.
+        4. Provide ONLY the rephrased query without any additional explanation or reasoning.
         
-        if model.startswith("duckduckgo/"):
-            # Use DuckDuckGo chat with context
-            for partial_response in get_response_from_duckduckgo(message, model, context_str, num_calls, temperature):
-                yield partial_response
-        elif model == "@cf/meta/llama-3.1-8b-instruct":
-            # Use Cloudflare API
-            for partial_response in get_response_from_cloudflare(prompt="", context=context_str, query=message, num_calls=num_calls, temperature=temperature, search_type="pdf"):
-                yield partial_response
-        else:
-            # Use Hugging Face API
-            for partial_response in get_response_from_pdf(message, model, selected_docs, num_calls=num_calls, temperature=temperature):
-                yield partial_response
+        Context: {context}
+        
+        New query: {original_query}
+        
+        Rephrased query:"""
+        response = DDGS().chat(prompt, model="llama-3.1-70b")
+        rephrased_query = response.split('\n')[0].strip()
+        return rephrased_query
+    return original_query
+
+def summarize_web_results(query: str, search_results: List[Dict[str, str]], conversation_manager: ConversationManager) -> str:
+    try:
+        context = conversation_manager.get_context()
+        search_context = "\n\n".join([f"Title: {result['title']}\nContent: {result['body']}" for result in search_results])
+
+        prompt = f"""You are a highly intelligent & expert analyst and your job is to skillfully articulate the web search results about '{query}' and considering the context: {context}, 
+        You have to create a comprehensive news summary FOCUSING on the context provided to you. 
+        Include key facts, relevant statistics, and expert opinions if available. 
+        Ensure the article is well-structured with an introduction, main body, and conclusion, IF NECESSARY. 
+        Address the query in the context of the ongoing conversation IF APPLICABLE.
+        Cite sources directly within the generated text and not at the end of the generated text, integrating URLs where appropriate to support the information provided:
+
+        {search_context}
+
+        Article:"""
+
+        summary = DDGS().chat(prompt, model="llama-3-70b")
+        return summary
     except Exception as e:
-        logging.error(f"Error with {model}: {str(e)}")
-        if "microsoft/Phi-3-mini-4k-instruct" in model:
-            logging.info("Falling back to Mistral model due to Phi-3 error")
-            fallback_model = "mistralai/Mistral-7B-Instruct-v0.3"
-            yield from respond(message, history, fallback_model, temperature, num_calls, selected_docs)
+        return f"An error occurred during summarization: {str(e)}"
+
+# Modify the existing respond function to handle both PDF and web search
+def respond(message, history, model, temperature, num_calls, selected_docs, use_web_search):
+    logging.info(f"User Query: {message}")
+    logging.info(f"Model Used: {model}")
+    logging.info(f"Selected Documents: {selected_docs}")
+    logging.info(f"Use Web Search: {use_web_search}")
+
+    if use_web_search:
+        original_query = message
+        rephrased_query = rephrase_query(message, conversation_manager)
+        logging.info(f"Original query: {original_query}")
+        logging.info(f"Rephrased query: {rephrased_query}")
+
+        final_summary = ""
+        for _ in range(num_calls):
+            search_results = get_web_search_results(rephrased_query)
+            if not search_results:
+                final_summary += f"No search results found for the query: {rephrased_query}\n\n"
+            elif "error" in search_results[0]:
+                final_summary += search_results[0]["error"] + "\n\n"
+            else:
+                summary = summarize_web_results(rephrased_query, search_results, conversation_manager)
+                final_summary += summary + "\n\n"
+
+        if final_summary:
+            conversation_manager.add_interaction(original_query, final_summary)
+            yield final_summary
         else:
-            yield f"An error occurred with the {model} model: {str(e)}. Please try again or select a different model."
+            yield "Unable to generate a response. Please try a different query."
+    else:
+        # Existing PDF search logic
+        try:
+            embed = get_embeddings()
+            if os.path.exists("faiss_database"):
+                database = FAISS.load_local("faiss_database", embed, allow_dangerous_deserialization=True)
+                retriever = database.as_retriever(search_kwargs={"k": 20})
+                
+                all_relevant_docs = retriever.get_relevant_documents(message)
+                relevant_docs = [doc for doc in all_relevant_docs if doc.metadata["source"] in selected_docs]
+                
+                if not relevant_docs:
+                    yield "No relevant information found in the selected documents. Please try selecting different documents or rephrasing your query."
+                    return
+
+                context_str = "\n".join([doc.page_content for doc in relevant_docs])
+                logging.info(f"Context length: {len(context_str)}")
+            else:
+                context_str = "No documents available."
+                yield "No documents available. Please upload PDF documents to answer questions."
+                return
+            
+            if model == "@cf/meta/llama-3.1-8b-instruct":
+                # Use Cloudflare API
+                for response in get_response_from_cloudflare(prompt="", context=context_str, query=message, num_calls=num_calls, temperature=temperature, search_type="pdf"):
+                    yield response
+            else:
+                # Use Hugging Face API
+                for response in get_response_from_pdf(message, model, selected_docs, num_calls=num_calls, temperature=temperature):
+                    yield response
+        except Exception as e:
+            logging.error(f"Error with {model}: {str(e)}")
+            if "microsoft/Phi-3-mini-4k-instruct" in model:
+                logging.info("Falling back to Mistral model due to Phi-3 error")
+                fallback_model = "mistralai/Mistral-7B-Instruct-v0.3"
+                yield from respond(message, history, fallback_model, temperature, num_calls, selected_docs, use_web_search)
+            else:
+                yield f"An error occurred with the {model} model: {str(e)}. Please try again or select a different model."
 
 logging.basicConfig(level=logging.DEBUG)
 
@@ -563,16 +650,18 @@ use_web_search = gr.Checkbox(label="Use Web Search", value=True)
 custom_placeholder = "Ask a question (Note: You can toggle between Web Search and PDF Chat in Additional Inputs below)"
 
 # Update the demo interface
+# Update the Gradio interface
 demo = gr.ChatInterface(
     respond,
     additional_inputs=[
         gr.Dropdown(choices=MODELS, label="Select Model", value=MODELS[3]),
         gr.Slider(minimum=0.1, maximum=1.0, value=0.2, step=0.1, label="Temperature"),
         gr.Slider(minimum=1, maximum=5, value=1, step=1, label="Number of API Calls"),
-        document_selector
+        gr.CheckboxGroup(label="Select documents to query", choices=[]),
+        gr.Checkbox(label="Use Web Search", value=False)
     ],
-    title="AI-powered PDF Chat and DuckDuckGo Chat Assistant",
-    description="Chat with your PDFs or use DuckDuckGo chat models to answer questions.",
+    title="AI-powered PDF Chat and Web Search Assistant",
+    description="Chat with your PDFs or use web search to answer questions.",
     theme=gr.themes.Soft(
         primary_hue="orange",
         secondary_hue="amber",