app.py

import os
import sys
import random
import gradio as gr
from multiprocessing import freeze_support
import importlib
import inspect

# === Fix path to include src/txagent
sys.path.insert(0, os.path.join(os.path.dirname(__file__), "src"))

# === Import and reload to ensure correct file
import txagent.txagent
importlib.reload(txagent.txagent)
from txagent.txagent import TxAgent

# === Debug print
print(">>> TxAgent loaded from:", inspect.getfile(TxAgent))
print(">>> TxAgent has run_gradio_chat:", hasattr(TxAgent, "run_gradio_chat"))

# === Environment
current_dir = os.path.abspath(os.path.dirname(__file__))
os.environ["MKL_THREADING_LAYER"] = "GNU"
os.environ["TOKENIZERS_PARALLELISM"] = "false"

# === Model config
model_name = "mims-harvard/TxAgent-T1-Llama-3.1-8B"
rag_model_name = "mims-harvard/ToolRAG-T1-GTE-Qwen2-1.5B"
new_tool_files = {
    "new_tool": os.path.join(current_dir, "data", "new_tool.json")
}

# === Example prompts
question_examples = [
    ["Given a patient with WHIM syndrome on prophylactic antibiotics, is it advisable to co-administer Xolremdi with fluconazole?"],
    ["What treatment options exist for HER2+ breast cancer resistant to trastuzumab?"]
]

# === UI creation
def create_ui(agent):
    with gr.Blocks() as demo:
        gr.Markdown("<h1 style='text-align: center;'>TxAgent: Therapeutic Reasoning</h1>")
        gr.Markdown("Ask biomedical or therapeutic questions. Powered by step-by-step reasoning and tools.")

        temperature = gr.Slider(0, 1, value=0.3, label="Temperature")
        max_new_tokens = gr.Slider(128, 4096, value=1024, label="Max New Tokens")
        max_tokens = gr.Slider(128, 32000, value=8192, label="Max Total Tokens")
        max_round = gr.Slider(1, 50, value=30, label="Max Rounds")
        multi_agent = gr.Checkbox(label="Enable Multi-agent Reasoning", value=False)
        conversation_state = gr.State([])

        chatbot = gr.Chatbot(label="TxAgent", height=600, type="messages")
        message_input = gr.Textbox(placeholder="Ask your biomedical question...", show_label=False)
        send_button = gr.Button("Send", variant="primary")

        # === Core handler (streaming generator)
        def handle_chat(message, history, temperature, max_new_tokens, max_tokens, multi_agent, conversation, max_round):
            # Must yield a list of {"role": ..., "content": ...} dicts
            generator = agent.run_gradio_chat(
                message=message,
                history=history,
                temperature=temperature,
                max_new_tokens=max_new_tokens,
                max_token=max_tokens,
                call_agent=multi_agent,
                conversation=conversation,
                max_round=max_round
            )
            for update in generator:
                # Convert to list of dicts if not already
                formatted = [
                    {"role": m["role"], "content": m["content"]}
                    if isinstance(m, dict)
                    else {"role": m.role, "content": m.content}
                    for m in update
                ]
                yield formatted

        # === Trigger handlers
        send_button.click(
            fn=handle_chat,
            inputs=[message_input, chatbot, temperature, max_new_tokens, max_tokens, multi_agent, conversation_state, max_round],
            outputs=chatbot
        )

        message_input.submit(
            fn=handle_chat,
            inputs=[message_input, chatbot, temperature, max_new_tokens, max_tokens, multi_agent, conversation_state, max_round],
            outputs=chatbot
        )

        gr.Examples(examples=question_examples, inputs=message_input)

        gr.Markdown("**DISCLAIMER**: This demo is for research purposes only and does not provide medical advice.")

    return demo

# === Startup
if __name__ == "__main__":
    freeze_support()

    try:
        agent = TxAgent(
            model_name=model_name,
            rag_model_name=rag_model_name,
            tool_files_dict=new_tool_files,
            force_finish=True,
            enable_checker=True,
            step_rag_num=10,
            seed=100,
            additional_default_tools=["DirectResponse", "RequireClarification"]
        )
        agent.init_model()

        if not hasattr(agent, "run_gradio_chat"):
            raise AttributeError("❌ TxAgent is missing `run_gradio_chat`.")

        demo = create_ui(agent)
        demo.launch(show_error=True)

    except Exception as e:
        print(f"❌ Application failed to start: {e}")
        raise