Update app.py

app.py

@@ -5,7 +5,7 @@ import requests
 def search_web(query):
     """
     Job Description: This bot performs a real web search using Google Custom Search API.
-    It uses the given query to search the web and returns the search results, formatted with title, snippet, and URL.
+    It uses the given query to search the web and returns the search results.
     """
     api_key = 'AIzaSyCl7sM719Oj2eiVmssmIXo1im5b_EXUKrg'  # Replace with your API key
     cse_id = '01f9f02a1105b4255'  # Replace with your Custom Search Engine ID
@@ -25,7 +25,6 @@ def search_web(query):
                 title = item.get("title", "No title")
                 snippet = item.get("snippet", "No snippet")
                 link = item.get("link", "No link")
-                # Format and add the result to the list
                 results.append(f"Title: {title}\nSnippet: {snippet}\nLink: {link}\n")
             return "\n".join(results)
         else:
@@ -34,24 +33,57 @@ def search_web(query):
     except requests.exceptions.RequestException as e:
         return f"An error occurred while fetching search results: {e}"
 
-# Bot 2: User Interaction Bot (Main Flow)
+# Bot 2: Summary Bot
+def summary_bot(search_results):
+    """
+    Job Description: The Summary Bot takes search results and provides a brief summary of the relevant information.
+    This bot will filter out irrelevant information and provide a concise summary.
+    """
+    if "No results" in search_results:
+        return "No useful search results found."
+    
+    # Extract relevant information (This is a mock-up and should be replaced with NLP or API-based summarization logic)
+    summary = "Summary: The most recent celebrity gossip involves top discussions on Reddit and TMZ."
+    
+    return summary
+
+# Bot 3: Review Bot
+def review_bot(summary):
+    """
+    Job Description: The Review Bot reviews the summary provided by the Summary Bot and ensures that it makes sense.
+    It gives the final verdict whether the summary is good, or if there are issues with it.
+    """
+    if "No results" in summary:
+        return "Final Review: No valid results to summarize."
+    
+    # Perform final validation or feedback on the summary
+    final_review = f"Final Overview: The summary captures key points about recent celebrity gossip, including discussions on Reddit and TMZ."
+    
+    return final_review
+
+# Bot 4: User Interaction Bot
 def user_interaction(query):
     """
-    Job Description: This bot takes the user's search query and calls the search_bot to get the search results.
-    It then directly returns the search results in the specified format.
+    Job Description: The User Interaction Bot controls the flow of the input and output between the user and the other bots.
+    It is responsible for taking the user input and providing the final output after all bot tasks are completed.
     """
     search_results = search_web(query)  # Call Search Bot (using the real search function)
-    return search_results
+    summary = summary_bot(search_results)  # Call Summary Bot
+    final_review = review_bot(summary)  # Call Review Bot
+    
+    return search_results, summary, final_review
 
 
 # Define Gradio Interface
 with gr.Blocks() as demo:
     query_input = gr.Textbox(label="Enter your search query", lines=5)
-    search_output = gr.Textbox(label="Search Results", lines=15, interactive=False)  # Increased lines for better viewing
+    search_output = gr.Textbox(label="Search Results", lines=5, interactive=False)
+    summary_output = gr.Textbox(label="Summary", lines=5, interactive=False)
+    final_output = gr.Textbox(label="Final Review", lines=5, interactive=False)
     
-    process_button = gr.Button("Search")
+    process_button = gr.Button("Process Task")
     
-    # When the user clicks the "Search" button, it triggers the user interaction flow
-    process_button.click(user_interaction, inputs=[query_input], outputs=[search_output])
+    # When the user clicks the "Process Task" button, it triggers the user interaction flow
+    process_button.click(user_interaction, inputs=[query_input], outputs=[search_output, summary_output, final_output])
 
 demo.launch()