import gradio as gr
from huggingface_hub import InferenceClient

# Emissions data
EMISSIONS_FACTORS = {
    "transportation": {
        "car": 2.3,
        "bus": 0.1,
        "train": 0.04,
        "plane": 0.25,
    },
    "food": {
        "meat": 6.0,
        "vegetarian": 1.5,
        "vegan": 1.0,
    }
}

# Carbon footprint calculator
def calculate_footprint(car_km, bus_km, train_km, air_km, meat_meals, vegetarian_meals, vegan_meals):
    transport_emissions = (
        car_km * EMISSIONS_FACTORS["transportation"]["car"]
        + bus_km * EMISSIONS_FACTORS["transportation"]["bus"]
        + train_km * EMISSIONS_FACTORS["transportation"]["train"]
        + air_km * EMISSIONS_FACTORS["transportation"]["plane"]
    )

    food_emissions = (
        meat_meals * EMISSIONS_FACTORS["food"]["meat"]
        + vegetarian_meals * EMISSIONS_FACTORS["food"]["vegetarian"]
        + vegan_meals * EMISSIONS_FACTORS["food"]["vegan"]
    )

    total_emissions = transport_emissions + food_emissions

    stats = {
        "trees": round(total_emissions / 21),
        "flights": round(total_emissions / 500),
        "driving100km": round(total_emissions / 230),
    }

    return total_emissions, stats

# Response generator
def respond(
    message,
    history: list[dict[str, str]],
    system_message,
    car_km,
    bus_km,
    train_km,
    air_km,
    meat_meals,
    vegetarian_meals,
    vegan_meals,
    max_tokens,
    temperature,
    top_p,
    hf_token_textbox,
):
    client = InferenceClient(token=hf_token_textbox, model="openai/gpt-oss-20b")

    footprint, stats = calculate_footprint(
        car_km, bus_km, train_km, air_km, meat_meals, vegetarian_meals, vegan_meals
    )

    custom_prompt = f"""
The user's estimated weekly footprint is **{footprint:.1f} kg CO2**.
That's equivalent to planting about {stats['trees']} trees 🌳 or taking {stats['flights']} short flights ✈️.
Their breakdown includes both transportation and food habits.
Your job is to give them personalized, practical, and encouraging suggestions to reduce this footprint.

{system_message}
    """

    messages = [{"role": "system", "content": custom_prompt}]
    messages.extend(history)
    messages.append({"role": "user", "content": message})

    response = ""
    for message in client.chat_completion(
        messages,
        max_tokens=max_tokens,
        stream=True,
        temperature=temperature,
        top_p=top_p,
    ):
        choices = message.choices
        token = ""
        if len(choices) and choices[0].delta.content:
            token = choices[0].delta.content
        response += token
        yield response

# Chat UI
chatbot = gr.ChatInterface(
    fn=respond,
    chatbot=gr.Chatbot(),
    type="messages",
    additional_inputs=[
        gr.Textbox(value="You are Sustainable.ai, a friendly and practical climate coach.", label="System message"),
        gr.Number(value=0, label="🚘 Car Travel (km/week)"),
        gr.Number(value=0, label="🚌 Bus Travel (km/week)"),
        gr.Number(value=0, label="🚆 Train Travel (km/week)"),
        gr.Number(value=0, label="✈️ Air Travel (km/month)"),
        gr.Number(value=0, label="🥩 Meat Meals (per week)"),
        gr.Number(value=0, label="🥗 Vegetarian Meals (per week)"),
        gr.Number(value=0, label="🌱 Vegan Meals (per week)"),
        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
        gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p (nucleus sampling)"),
        gr.Textbox(value="", label="🔐 Hugging Face Token (paste here)", type="password"),
    ],
)

# Launch with public link
if __name__ == "__main__":
    chatbot.launch(share=True)