Update app.py

app.py

@@ -632,6 +632,7 @@ def create_ui() -> gr.Blocks:
                         value=True,
                         info="Automatically start analysis when repositories are found"
                     )
+                    smart_submit_btn = gr.Button("🔍 Find & Process Repositories", variant="primary", size="lg", scale=1)
                 
                 status_box_input = gr.Textbox(label="📊 Status", interactive=False, lines=2)
 
@@ -704,6 +705,9 @@ def create_ui() -> gr.Blocks:
                     )
                     send_btn = gr.Button("📤", variant="primary", scale=1)
 
+                with gr.Row():
+                    extract_analyze_btn = gr.Button("🎯 Extract Keywords & Analyze Now", variant="secondary", size="lg")
+
                 # Status and extracted info (auto-updated, no manual buttons needed)
                 with gr.Row():
                     with gr.Column():
@@ -857,6 +861,58 @@ def create_ui() -> gr.Blocks:
             
             return ""
 
+        def handle_extract_and_analyze(history: List[Dict[str, str]]) -> Tuple[str, str, str, List[str], int, pd.DataFrame, Any, pd.DataFrame, str, Any]:
+            """Extract keywords from chat, search repositories, and immediately start analysis."""
+            if not history:
+                return "❌ No conversation to extract from.", "", "", [], 0, pd.DataFrame(), gr.update(), pd.DataFrame(), "", gr.update(visible=False)
+            
+            # Convert the full, valid history for the extraction logic
+            tuple_history = convert_messages_to_tuples(history)
+            if not tuple_history:
+                return "❌ No completed conversations to analyze.", "", "", [], 0, pd.DataFrame(), gr.update(), pd.DataFrame(), "", gr.update(visible=False)
+                
+            # Get raw keywords string from the LLM
+            raw_keywords_str = extract_keywords_from_conversation(tuple_history)
+            
+            # Sanitize the LLM output to extract only keyword-like parts
+            cleaned_keywords = re.findall(r'[\w\s-]+', raw_keywords_str)
+            cleaned_keywords = [kw.strip() for kw in cleaned_keywords if kw.strip()]
+            
+            if not cleaned_keywords:
+                return f"❌ Could not extract valid keywords. Raw output: '{raw_keywords_str}'", "", "", [], 0, pd.DataFrame(), gr.update(), pd.DataFrame(), "", gr.update(visible=False)
+
+            # Join them into a clean, comma-separated string
+            final_keywords_str = ", ".join(cleaned_keywords)
+            
+            # Extract user requirements for analysis
+            user_requirements = extract_user_requirements_from_chat(history)
+            
+            # Auto-search repositories
+            repo_ids = []
+            for kw in cleaned_keywords[:3]:  # Use top 3 keywords to avoid too many results
+                repo_ids.extend(search_top_spaces(kw, limit=5))
+            
+            unique_repo_ids = list(dict.fromkeys(repo_ids))
+            
+            if not unique_repo_ids:
+                return f"❌ No repositories found for keywords: {final_keywords_str}", final_keywords_str, user_requirements, [], 0, pd.DataFrame(), gr.update(), pd.DataFrame(), "", gr.update(visible=False)
+            
+            write_repos_to_csv(unique_repo_ids)
+            df = format_dataframe_for_display(read_csv_to_dataframe())
+            
+            # Immediately start analysis
+            try:
+                analyzed_df, analysis_status, top_repos, top_section_update = handle_analyze_all_repos(unique_repo_ids, user_requirements)
+                
+                chat_status = f"🎉 Extracted keywords → Found {len(unique_repo_ids)} repositories → Analysis complete!"
+                
+                return chat_status, final_keywords_str, user_requirements, unique_repo_ids, 0, analyzed_df, gr.update(selected="analysis_tab"), top_repos, analysis_status, top_section_update
+                
+            except Exception as e:
+                logger.error(f"Error during extract and analyze: {e}")
+                error_status = f"✅ Found {len(unique_repo_ids)} repositories, but analysis failed: {e}"
+                return error_status, final_keywords_str, user_requirements, unique_repo_ids, 0, df, gr.update(selected="analysis_tab"), pd.DataFrame(), "", gr.update(visible=False)
+
         def extract_user_requirements_from_chat(history: List[Dict[str, str]]) -> str:
             """Extract user requirements from chatbot conversation."""
             if not history:
@@ -1071,6 +1127,18 @@ def create_ui() -> gr.Blocks:
             outputs=[df_output, status_box_input, top_repos_df, top_repos_section]
         )
         
+        # Smart Submit Button (same behavior as enter)
+        smart_submit_btn.click(
+            fn=handle_smart_input,
+            inputs=[smart_input, auto_analyze_checkbox],
+            outputs=[repo_ids_state, current_repo_idx_state, df_output, status_box_input, tabs, status_box_input]
+        ).then(
+            # If auto_analyze is enabled and we got repos, start analysis automatically
+            fn=lambda repo_ids, user_reqs, trigger: handle_analyze_all_repos(repo_ids, user_reqs) if trigger == "auto_analyze" and repo_ids else (pd.DataFrame(), "Ready for analysis.", pd.DataFrame(), gr.update(visible=False)),
+            inputs=[repo_ids_state, user_requirements_state, status_box_input],
+            outputs=[df_output, status_box_input, top_repos_df, top_repos_section]
+        )
+        
         # Auto-analyze checkbox toggle
         auto_analyze_checkbox.change(
             fn=handle_auto_analyze_toggle,
@@ -1126,6 +1194,18 @@ def create_ui() -> gr.Blocks:
             outputs=[df_output, chat_status, top_repos_df, top_repos_section]
         )
         
+        # Extract and Analyze Button (one-click solution for chatbot)
+        extract_analyze_btn.click(
+            fn=handle_extract_and_analyze,
+            inputs=[chatbot],
+            outputs=[chat_status, extracted_keywords_output, user_requirements_state, repo_ids_state, current_repo_idx_state, df_output, tabs, top_repos_df, status_box_analysis, top_repos_section]
+        ).then(
+            # Update requirements display when they change
+            fn=lambda req: req if req.strip() else "No specific requirements extracted from conversation.",
+            inputs=[user_requirements_state],
+            outputs=[current_requirements_display]
+        )
+        
         # Repo Explorer Tab
         setup_repo_explorer_events(repo_components, repo_states)