Update app.py

app.py

@@ -131,8 +131,137 @@ def init_agent():
     return agent
 
 def process_final_report(agent, file, chatbot_state: List[Dict[str, str]]) -> Tuple[List[Dict[str, str]], Union[str, None]]:
-    from app_backend import run_analysis_pipeline  # Or keep as is if internal
-    return run_analysis_pipeline(agent, file, chatbot_state)
+    """Process the Excel file and generate a final report."""
+    messages = chatbot_state if chatbot_state else []
+    report_path = None
+
+    if file is None or not hasattr(file, "name"):
+        messages.append({"role": "assistant", "content": "❌ Please upload a valid Excel file before analyzing."})
+        return messages, report_path
+
+    try:
+        messages.append({"role": "user", "content": f"Processing Excel file: {os.path.basename(file.name)}"})
+        messages.append({"role": "assistant", "content": "⏳ Extracting and analyzing data..."})
+
+        extracted_text = extract_text_from_excel(file.name)
+        chunks = split_text_into_chunks(extracted_text, max_tokens=MAX_CHUNK_TOKENS)
+        chunk_responses = []
+
+        for i, chunk in enumerate(chunks):
+            messages.append({"role": "assistant", "content": f"🔍 Analyzing chunk {i+1}/{len(chunks)}..."})
+            prompt = build_prompt_from_text(chunk)
+            prompt_tokens = estimate_tokens(prompt)
+
+            if prompt_tokens > MAX_MODEL_TOKENS:
+                messages.append({"role": "assistant", "content": f"❌ Chunk {i+1} prompt too long ({prompt_tokens} tokens). Skipping..."})
+                continue
+
+            response = ""
+            try:
+                for result in agent.run_gradio_chat(
+                    message=prompt,
+                    history=[],
+                    temperature=0.2,
+                    max_new_tokens=MAX_NEW_TOKENS,
+                    max_token=MAX_MODEL_TOKENS,
+                    call_agent=False,
+                    conversation=[],
+                ):
+                    if isinstance(result, str):
+                        response += result
+                    elif hasattr(result, "content"):
+                        response += result.content
+                    elif isinstance(result, list):
+                        for r in result:
+                            if hasattr(r, "content"):
+                                response += r.content
+            except Exception as e:
+                messages.append({"role": "assistant", "content": f"❌ Error analyzing chunk {i+1}: {str(e)}"})
+                continue
+
+            chunk_responses.append(clean_response(response))
+            messages.append({"role": "assistant", "content": f"✅ Chunk {i+1} analysis complete"})
+
+        if not chunk_responses:
+            messages.append({"role": "assistant", "content": "❌ No valid chunk responses to summarize."})
+            return messages, report_path
+
+        summary = ""
+        current_summary_tokens = 0
+        for i, response in enumerate(chunk_responses):
+            response_tokens = estimate_tokens(response)
+            if current_summary_tokens + response_tokens > MAX_MODEL_TOKENS - PROMPT_OVERHEAD - MAX_NEW_TOKENS:
+                summary_prompt = f"Summarize the following analysis:\n\n{summary}\n\nProvide a concise summary."
+                summary_response = ""
+                try:
+                    for result in agent.run_gradio_chat(
+                        message=summary_prompt,
+                        history=[],
+                        temperature=0.2,
+                        max_new_tokens=MAX_NEW_TOKENS,
+                        max_token=MAX_MODEL_TOKENS,
+                        call_agent=False,
+                        conversation=[],
+                    ):
+                        if isinstance(result, str):
+                            summary_response += result
+                        elif hasattr(result, "content"):
+                            summary_response += result.content
+                        elif isinstance(result, list):
+                            for r in result:
+                                if hasattr(r, "content"):
+                                    summary_response += r.content
+                    summary = clean_response(summary_response)
+                    current_summary_tokens = estimate_tokens(summary)
+                except Exception as e:
+                    messages.append({"role": "assistant", "content": f"❌ Error summarizing intermediate results: {str(e)}"})
+                    return messages, report_path
+
+            summary += f"\n\n### Chunk {i+1} Analysis\n{response}"
+            current_summary_tokens += response_tokens
+
+        final_prompt = f"Summarize the key findings from the following analyses:\n\n{summary}"
+        messages.append({"role": "assistant", "content": "📊 Generating final report..."})
+
+        final_report_text = ""
+        try:
+            for result in agent.run_gradio_chat(
+                message=final_prompt,
+                history=[],
+                temperature=0.2,
+                max_new_tokens=MAX_NEW_TOKENS,
+                max_token=MAX_MODEL_TOKENS,
+                call_agent=False,
+                conversation=[],
+            ):
+                if isinstance(result, str):
+                    final_report_text += result
+                elif hasattr(result, "content"):
+                    final_report_text += result.content
+                elif isinstance(result, list):
+                    for r in result:
+                        if hasattr(r, "content"):
+                            final_report_text += r.content
+        except Exception as e:
+            messages.append({"role": "assistant", "content": f"❌ Error generating final report: {str(e)}"})
+            return messages, report_path
+
+        final_report = f"# 🧠 Final Patient Report\n\n{clean_response(final_report_text)}"
+        messages[-1]["content"] = f"📊 Final Report:\n\n{clean_response(final_report_text)}"
+
+        timestamp = datetime.now().strftime('%Y%m%d_%H%M%S')
+        report_path = os.path.join(report_dir, f"report_{timestamp}.md")
+
+        with open(report_path, 'w') as f:
+            f.write(final_report)
+
+        messages.append({"role": "assistant", "content": f"✅ Report generated and saved: report_{timestamp}.md"})
+
+    except Exception as e:
+        messages.append({"role": "assistant", "content": f"❌ Error processing file: {str(e)}"})
+
+    return messages, report_path
+
 
 def create_ui(agent):
     """Create the Gradio UI for the patient history analysis tool."""