import gradio as gr
import matplotlib.pyplot as plt
import pandas as np
import tempfile
from datetime import datetime
from langchain_core.messages import HumanMessage
from tools import tools
from agents import *
from config import *
from workflow import create_workflow

# Initialize workflow
graph = create_workflow()

# Helper Functions
def run_graph(input_message, history, user_details):
    try:
        system_prompt = (
            "You are a fitness and health assistant. "
            "Provide advice based on the user's personal details and goals. "
            "If user details like weight, height, and activity level are provided, prioritize personalized advice. "
            "Incorporate metrics such as BMI and daily caloric needs directly into your suggestions. "
            "Visualize the user's metrics and provide actionable advice tailored to their needs. "
            "If details are missing, provide general advice but encourage users to share their metrics for tailored guidance."
        )

        # Compose input for the LLM
        user_details_summary = (
            f"Name: {user_details.get('name', 'Unknown')}, "
            f"Age: {user_details.get('age', 'Unknown')}, "
            f"Gender: {user_details.get('gender', 'Unknown')}, "
            f"Weight: {user_details.get('weight', 'Unknown')} kg, "
            f"Height: {user_details.get('height', 'Unknown')} cm, "
            f"Activity Level: {user_details.get('activity_level', 'Unknown')}"
        )

        messages = [
            {"role": "system", "content": system_prompt},
            {"role": "user", "content": f"User details: {user_details_summary}"},
            {"role": "user", "content": input_message}
        ]

        # Generate the response
        response = graph.invoke({"messages": messages})
        return response["messages"][1]["content"]

    except Exception as e:
        return f"An error occurred while processing your request: {e}"


def calculate_bmi(height, weight):
    height_m = height / 100
    bmi = weight / (height_m ** 2)
    if bmi < 18.5:
        status = "underweight"
    elif 18.5 <= bmi < 24.9:
        status = "normal weight"
    elif 25 <= bmi < 29.9:
        status = "overweight"
    else:
        status = "obese"
    return bmi, status


def visualize_bmi_and_calories(bmi, calories):
    # Create a combined visualization
    categories = ["Underweight", "Normal Weight", "Overweight", "Obese"]
    bmi_values = [18.5, 24.9, 29.9, 40]
    calorie_range = [1500, 2000, 2500, 3000]

    fig, ax1 = plt.subplots(figsize=(10, 6))

    # BMI visualization
    ax1.bar(categories, bmi_values, color=['blue', 'green', 'orange', 'red'], alpha=0.6, label="BMI Ranges")
    ax1.axhline(y=bmi, color='purple', linestyle='--', linewidth=2, label=f"Your BMI: {bmi:.2f}")
    ax1.set_ylabel("BMI Value")
    ax1.set_title("BMI and Caloric Needs Visualization")
    ax1.legend(loc="upper left")

    # Calorie visualization
    ax2 = ax1.twinx()
    ax2.plot(categories, calorie_range, 'o-', color='magenta', label="Calorie Ranges")
    ax2.axhline(y=calories, color='cyan', linestyle='--', linewidth=2, label=f"Your Calorie Needs: {calories:.2f} kcal")
    ax2.set_ylabel("Calories")
    ax2.legend(loc="upper right")

    plt.tight_layout()

    # Save to a temporary file
    temp_file = tempfile.NamedTemporaryFile(suffix=".png", delete=False)
    plt.savefig(temp_file.name)
    plt.close()

    return temp_file.name


def calculate_calories(age, weight, height, activity_level, gender):
    if gender.lower() == "male":
        bmr = 10 * weight + 6.25 * height - 5 * age + 5
    else:
        bmr = 10 * weight + 6.25 * height - 5 * age - 161

    activity_multipliers = {
        "Sedentary": 1.2,
        "Lightly active": 1.375,
        "Moderately active": 1.55,
        "Very active": 1.725,
        "Extra active": 1.9,
    }

    return bmr * activity_multipliers.get(activity_level, 1.2)

# Interface Components
with gr.Blocks() as demo:
    gr.Markdown("<strong>FIT.AI - Your Fitness and Wellbeing Coach</strong>")

    # User Info
    with gr.Row():
        user_name = gr.Textbox(placeholder="Enter your name", label="Name")
        user_age = gr.Number(label="Age (years)", value=25, precision=0)
        user_gender = gr.Dropdown(choices=["Male", "Female"], label="Gender", value="Male")
        user_weight = gr.Number(label="Weight (kg)", value=70, precision=1)
        user_height = gr.Number(label="Height (cm)", value=170, precision=1)
        activity_level = gr.Dropdown(
            choices=["Sedentary", "Lightly active", "Moderately active", "Very active", "Extra active"],
            label="Activity Level",
            value="Moderately active"
        )

    # Outputs
    bmi_output = gr.Label(label="BMI Result")
    calorie_output = gr.Label(label="Calorie Needs")
    bmi_chart = gr.Image(label="BMI and Calorie Chart")

    chatbot = gr.Chatbot(label="Chat with FIT.AI")
    text_input = gr.Textbox(placeholder="Type your question here...", label="Your Question")
    submit_button = gr.Button("Submit")
    clear_button = gr.Button("Clear Chat")

    def submit_message(message, history=[]):
        user_details = {
            "name": user_name.value,
            "age": user_age.value,
            "weight": user_weight.value,
            "height": user_height.value,
            "activity_level": activity_level.value,
            "gender": user_gender.value
        }
        bmi, status = calculate_bmi(user_details['height'], user_details['weight'])
        calories = calculate_calories(
            user_details['age'], user_details['weight'], user_details['height'], user_details['activity_level'], user_details['gender']
        )
        chart_path = visualize_bmi_and_calories(bmi, calories)

        # Append metrics to response with actionable advice
        personalized_metrics = (
            f"Based on your metrics:\n"
            f"- BMI: {bmi:.2f} ({status})\n"
            f"- Daily Caloric Needs: {calories:.2f} kcal\n"
            f"\nAdvice: "
        )
        if status == "underweight":
            personalized_metrics += "Your BMI indicates you are underweight. Focus on consuming more calories from nutrient-dense foods, including lean proteins, healthy fats, and whole grains. Consider strength training exercises to build muscle mass."
        elif status == "normal weight":
            personalized_metrics += "Your BMI is in the normal range. Maintain your weight by balancing calorie intake with regular physical activity. Incorporate a mix of cardio and strength training for overall fitness."
        elif status == "overweight":
            personalized_metrics += "Your BMI indicates you are overweight. Aim for a calorie deficit by reducing portion sizes and increasing physical activity. Focus on a diet rich in vegetables, lean protein, and healthy fats."
        else:
            personalized_metrics += "Your BMI indicates you are obese. Consult a healthcare provider for a personalized weight loss plan. Start with small, sustainable changes, such as reducing sugary foods and increasing daily physical activity."

        response = personalized_metrics
        history.append((f"User", message))
        history.append(("FIT.AI", response))
        return history, chart_path

    submit_button.click(submit_message, inputs=[text_input, chatbot], outputs=[chatbot, bmi_chart])
    clear_button.click(lambda: ([], ""), inputs=None, outputs=[chatbot, bmi_chart])

    # Actions
    calculate_button = gr.Button("Calculate")

    def calculate_metrics(age, weight, height, gender, activity_level):
        bmi, status = calculate_bmi(height, weight)
        bmi_path = visualize_bmi_and_calories(bmi, calculate_calories(age, weight, height, activity_level, gender))
        calories = calculate_calories(age, weight, height, activity_level, gender)
        return f"Your BMI is {bmi:.2f}, considered {status}.", f"Daily calorie needs: {calories:.2f} kcal", bmi_path

    calculate_button.click(
        calculate_metrics,
        inputs=[user_age, user_weight, user_height, user_gender, activity_level],
        outputs=[bmi_output, calorie_output, bmi_chart]
    )

demo.launch(share=True)