app.py

import gradio as gr
import yaml
import json
import os
import traceback
import matplotlib.pyplot as plt
import numpy as np

from engine.loader import load_persona
from engine.drift import apply_stimuli
from engine.responder import generate_response
from engine.utils import safe_log
from engine.logger import log_transcript

# Paths
persona_dir = "./personas"
stimuli_path = "./stimuli/events.json"
error_log_path = "./driftline_errors.log"



# Load available personas
def get_persona_choices():
    return [f for f in os.listdir(persona_dir) if f.endswith(".yml")]

# Load available stimuli
def get_stimuli_choices():
    try:
        with open(stimuli_path, "r") as f:
            events = json.load(f)
        return [e["event"] for e in events]
    except Exception as e:
        safe_log("Stimuli load error", str(e))
        return []

# Generate radar chart for traits
def plot_traits(state):
    traits = ["innovation", "openness", "risk_tolerance", "peer_influence"]
    values = [state.get(t, 0.0) for t in traits]
    angles = np.linspace(0, 2 * np.pi, len(traits), endpoint=False).tolist()
    values += values[:1]
    angles += angles[:1]

    fig, ax = plt.subplots(figsize=(4, 4), subplot_kw=dict(polar=True))
    ax.plot(angles, values, color="blue", linewidth=2)
    ax.fill(angles, values, color="blue", alpha=0.25)
    ax.set_xticks(angles[:-1])
    ax.set_xticklabels(traits)
    ax.set_yticklabels([])
    ax.set_title("Dynamic Trait Profile", fontsize=12)
    fig.tight_layout()

    chart_path = "./trait_chart.png"
    fig.savefig(chart_path)
    plt.close(fig)
    return chart_path

# Main simulation function
def simulate(prompt, selected_event, selected_persona_file):
    try:
        persona_path = os.path.join(persona_dir, selected_persona_file)
        persona = load_persona(persona_path)

        with open(stimuli_path, "r") as f:
            events = json.load(f)
        event = next((e for e in events if e["event"] == selected_event), None)

        if event:
            persona = apply_stimuli(persona, event)

        response = generate_response(prompt, persona)
        state_yaml = yaml.dump(persona["dynamic_state"], sort_keys=False)
        chart_path = plot_traits(persona["dynamic_state"])

        # ? Log transcript here, after response is generated
        from engine.logger import log_transcript
        transcript_path = log_transcript(persona, prompt, selected_event, response)

        return response, state_yaml, chart_path

    except Exception as e:
        error_msg = traceback.format_exc()
        safe_log("Simulation error", error_msg)
        return "[ERROR] Simulation failed. Check logs.", "", None

# Gradio UI
with gr.Blocks(title="Driftline HCP Simulator") as ui:
    gr.Markdown("## 🧠 Driftline: Adaptive HCP Simulation")
    gr.Markdown("Simulate how healthcare personas evolve in response to market stimuli.")

    with gr.Row():
        persona_files = get_persona_choices()
        default_persona = persona_files[0] if persona_files else None

        persona_selector = gr.Dropdown(
        label="Choose Persona",
        choices=persona_files,
        value=default_persona,
        allow_custom_value=False
)
        event_selector = gr.Dropdown(label="Market Stimulus", choices=get_stimuli_choices(), value="FDA_approval")

    prompt = gr.Textbox(label="Interviewer Prompt", lines=2, placeholder="Ask about a new therapy, trial data, or prescribing behavior...")

    with gr.Row():
        simulate_btn = gr.Button("Run Simulation")
        clear_btn = gr.Button("Clear")

    response_output = gr.Textbox(label="Simulated HCP Response", lines=6)
    state_output = gr.Textbox(label="Updated Persona State", lines=10)
    trait_chart = gr.Image(label="Trait Radar Chart")

    simulate_btn.click(
        fn=simulate,
        inputs=[prompt, event_selector, persona_selector],
        outputs=[response_output, state_output, trait_chart]
    )

    clear_btn.click(
        fn=lambda: ("", "", None),
        inputs=[],
        outputs=[response_output, state_output, trait_chart]
    )

ui.launch()