Update app.py

app.py

@@ -35,23 +35,35 @@ def file_hash(path: str) -> str:
         return hashlib.md5(f.read()).hexdigest()
 
 def clean_response(text: str) -> str:
-    text = text.encode("utf-16", "surrogatepass").decode("utf-16", "ignore")
+    try:
+        # First try to encode/decode to handle any surrogate pairs
+        text = text.encode('utf-8', 'surrogatepass').decode('utf-8')
+    except UnicodeEncodeError:
+        # Fallback to replace strategy if there are invalid characters
+        text = text.encode('utf-8', 'replace').decode('utf-8')
+    
+    # Additional cleaning
     text = re.sub(r"\[.*?\]|\bNone\b", "", text, flags=re.DOTALL)
     text = re.sub(r"\n{3,}", "\n\n", text)
     text = re.sub(r"[^\n#\-\*\w\s\.,:\(\)]+", "", text)
     return text.strip()
 
 def parse_excel_to_prompts(file_path: str) -> List[str]:
-    xl = pd.ExcelFile(file_path)
-    df = xl.parse(xl.sheet_names[0], header=0).fillna("")
-    groups = df.groupby("Booking Number")
-    prompts = []
-    for booking, group in groups:
-        records = []
-        for _, row in group.iterrows():
-            records.append(f"- {row['Form Name']}: {row['Form Item']} = {row['Item Response']} ({row['Interview Date']} by {row['Interviewer']})\n{row['Description']}")
-        record_text = "\n".join(records)
-        prompt = f"""
+    try:
+        xl = pd.ExcelFile(file_path)
+        df = xl.parse(xl.sheet_names[0], header=0).fillna("")
+        groups = df.groupby("Booking Number")
+        prompts = []
+        
+        for booking, group in groups:
+            records = []
+            for _, row in group.iterrows():
+                record = f"- {row['Form Name']}: {row['Form Item']} = {row['Item Response']} ({row['Interview Date']} by {row['Interviewer']})\n{row['Description']}"
+                # Clean each record to prevent encoding issues
+                records.append(clean_response(record))
+            
+            record_text = "\n".join(records)
+            prompt = f"""
 Patient Booking Number: {booking}
 
 Instructions:
@@ -72,84 +84,181 @@ Data:
 ### Urgent Follow-up
 - ...
 """
-        prompts.append(prompt)
-    return prompts
+            prompts.append(prompt)
+        return prompts
+    except Exception as e:
+        raise ValueError(f"Error parsing Excel file: {str(e)}")
 
 def init_agent():
     default_tool_path = os.path.abspath("data/new_tool.json")
     target_tool_path = os.path.join(tool_cache_dir, "new_tool.json")
+    
     if not os.path.exists(target_tool_path):
-        shutil.copy(default_tool_path, target_tool_path)
-    agent = TxAgent(
-        model_name="mims-harvard/TxAgent-T1-Llama-3.1-8B",
-        rag_model_name="mims-harvard/ToolRAG-T1-GTE-Qwen2-1.5B",
-        tool_files_dict={"new_tool": target_tool_path},
-        force_finish=True,
-        enable_checker=True,
-        step_rag_num=4,
-        seed=100,
-        additional_default_tools=[],
-    )
-    agent.init_model()
-    return agent
+        try:
+            shutil.copy(default_tool_path, target_tool_path)
+        except Exception as e:
+            raise RuntimeError(f"Failed to copy tool file: {str(e)}")
+    
+    try:
+        agent = TxAgent(
+            model_name="mims-harvard/TxAgent-T1-Llama-3.1-8B",
+            rag_model_name="mims-harvard/ToolRAG-T1-GTE-Qwen2-1.5B",
+            tool_files_dict={"new_tool": target_tool_path},
+            force_finish=True,
+            enable_checker=True,
+            step_rag_num=4,
+            seed=100,
+            additional_default_tools=[],
+        )
+        agent.init_model()
+        return agent
+    except Exception as e:
+        raise RuntimeError(f"Failed to initialize agent: {str(e)}")
 
 def create_ui(agent):
     with gr.Blocks(theme=gr.themes.Soft()) as demo:
-        gr.Markdown("<h1 style='text-align: center;'>\ud83e\uddfa Clinical Oversight Assistant (Excel Optimized)</h1>")
-        chatbot = gr.Chatbot(label="Analysis", height=600, type="messages")
-        file_upload = gr.File(file_types=[".xlsx"], file_count="single")
-        msg_input = gr.Textbox(placeholder="Ask about patient history...", show_label=False)
-        send_btn = gr.Button("Analyze", variant="primary")
-        download_output = gr.File(label="Download Full Report")
-
+        gr.Markdown("# 🏥 Clinical Oversight Assistant (Excel Optimized)")
+        
+        with gr.Row():
+            with gr.Column(scale=1):
+                file_upload = gr.File(
+                    label="Upload Excel File",
+                    file_types=[".xlsx"], 
+                    file_count="single",
+                    interactive=True
+                )
+                msg_input = gr.Textbox(
+                    label="Additional Instructions",
+                    placeholder="Add any specific analysis requests...",
+                    lines=3
+                )
+                send_btn = gr.Button("Analyze", variant="primary")
+                
+            with gr.Column(scale=2):
+                chatbot = gr.Chatbot(
+                    label="Analysis Results",
+                    height=600,
+                    bubble_full_width=False,
+                    show_copy_button=True
+                )
+                download_output = gr.File(
+                    label="Download Full Report",
+                    interactive=False
+                )
+        
         def analyze(message: str, history: List[dict], file) -> tuple:
-            history.append({"role": "user", "content": message})
-            history.append({"role": "assistant", "content": "⏳ Processing Excel data..."})
-            yield history, None
-
-            prompts = parse_excel_to_prompts(file.name)
-            full_output = ""
-
-            for idx, prompt in enumerate(prompts, 1):
-                chunk_output = ""
-                for result in agent.run_gradio_chat(
-                    message=prompt,
-                    history=[],
-                    temperature=0.2,
-                    max_new_tokens=1024,
-                    max_token=4096,
-                    call_agent=False,
-                    conversation=[],
-                ):
-                    if isinstance(result, list):
-                        for r in result:
-                            if hasattr(r, 'content') and r.content:
-                                chunk_output += clean_response(r.content) + "\n"
-                    elif isinstance(result, str):
-                        chunk_output += clean_response(result) + "\n"
-                if chunk_output:
-                    output = f"--- Booking {idx} ---\n{chunk_output.strip()}\n"
-                    history.append({"role": "assistant", "content": output})
-                    full_output += output + "\n"
-                    yield history, None
-
-            file_hash_value = file_hash(file.name)
-            report_path = os.path.join(report_dir, f"{file_hash_value}_report.txt")
-            with open(report_path, "w", encoding="utf-8") as f:
-                f.write(full_output)
-            yield history, report_path if os.path.exists(report_path) else None
-
-        send_btn.click(analyze, inputs=[msg_input, gr.State([]), file_upload], outputs=[chatbot, download_output])
-        msg_input.submit(analyze, inputs=[msg_input, gr.State([]), file_upload], outputs=[chatbot, download_output])
+            if not file:
+                raise gr.Error("Please upload an Excel file first")
+            
+            try:
+                # Initial processing message
+                history.append({"role": "user", "content": message})
+                history.append({"role": "assistant", "content": "⏳ Processing Excel data..."})
+                yield history, None
+                
+                # Parse Excel file
+                prompts = parse_excel_to_prompts(file.name)
+                full_output = ""
+                
+                # Process each booking
+                for idx, prompt in enumerate(prompts, 1):
+                    chunk_output = ""
+                    try:
+                        for result in agent.run_gradio_chat(
+                            message=prompt,
+                            history=[],
+                            temperature=0.2,
+                            max_new_tokens=1024,
+                            max_token=4096,
+                            call_agent=False,
+                            conversation=[],
+                        ):
+                            if isinstance(result, list):
+                                for r in result:
+                                    if hasattr(r, 'content') and r.content:
+                                        cleaned = clean_response(r.content)
+                                        chunk_output += cleaned + "\n"
+                            elif isinstance(result, str):
+                                cleaned = clean_response(result)
+                                chunk_output += cleaned + "\n"
+                                
+                            # Yield intermediate results
+                            if chunk_output:
+                                output = f"--- Booking {idx} ---\n{chunk_output.strip()}\n"
+                                history[-1] = {"role": "assistant", "content": output}
+                                yield history, None
+                                
+                    except Exception as e:
+                        error_msg = f"⚠️ Error processing booking {idx}: {str(e)}"
+                        history.append({"role": "assistant", "content": error_msg})
+                        yield history, None
+                        continue
+                        
+                    if chunk_output:
+                        output = f"--- Booking {idx} ---\n{chunk_output.strip()}\n"
+                        history.append({"role": "assistant", "content": output})
+                        full_output += output + "\n"
+                        yield history, None
+                
+                # Save report
+                file_hash_value = file_hash(file.name)
+                report_path = os.path.join(report_dir, f"{file_hash_value}_report.txt")
+                with open(report_path, "w", encoding="utf-8") as f:
+                    f.write(full_output)
+                    
+                yield history, report_path if os.path.exists(report_path) else None
+                
+            except Exception as e:
+                history.append({"role": "assistant", "content": f"❌ Error: {str(e)}"})
+                yield history, None
+                raise gr.Error(f"Analysis failed: {str(e)}")
+
+        # Event handlers
+        send_btn.click(
+            analyze,
+            inputs=[msg_input, gr.State([]), file_upload],
+            outputs=[chatbot, download_output],
+            api_name="analyze"
+        )
+        
+        msg_input.submit(
+            analyze,
+            inputs=[msg_input, gr.State([]), file_upload],
+            outputs=[chatbot, download_output]
+        )
+        
+        # Additional UI elements
+        with gr.Accordion("Instructions", open=False):
+            gr.Markdown("""
+            **How to use:**
+            1. Upload an Excel file containing patient records
+            2. Optionally add specific analysis instructions
+            3. Click 'Analyze' to process the data
+            4. Review results and download the full report
+            
+            **Excel Format Requirements:**
+            - Must contain columns: Booking Number, Form Name, Form Item, Item Response, Interview Date, Interviewer, Description
+            - Each row represents one patient record item
+            """)
+    
     return demo
 
 if __name__ == "__main__":
-    agent = init_agent()
-    demo = create_ui(agent)
-    demo.queue(api_open=False).launch(
-        server_name="0.0.0.0",
-        server_port=7860,
-        show_error=True,
-        allowed_paths=[report_dir],
-        share=True
-    )
+    try:
+        agent = init_agent()
+        demo = create_ui(agent)
+        
+        # Launch with error handling
+        demo.queue(
+            api_open=False,
+            max_size=20
+        ).launch(
+            server_name="0.0.0.0",
+            server_port=7860,
+            show_error=True,
+            allowed_paths=[report_dir],
+            share=False  # Changed to False to avoid the warning
+        )
+    except Exception as e:
+        print(f"Failed to launch application: {str(e)}")
+        sys.exit(1)