import os
import gradio as gr
import nltk
import numpy as np
import tflearn
import random
import json
import pickle
from nltk.tokenize import word_tokenize
from nltk.stem.lancaster import LancasterStemmer
from transformers import AutoTokenizer, AutoModelForSequenceClassification, pipeline
import pandas as pd
import torch

# Disable TensorFlow GPU warnings (safe since we are using CPU)
os.environ["CUDA_VISIBLE_DEVICES"] = "-1"

# Download necessary NLTK resources
nltk.download("punkt")

# Initialize Lancaster Stemmer for text preprocessing
stemmer = LancasterStemmer()

# Load intents.json for the chatbot
with open("intents.json") as file:
    intents_data = json.load(file)

# Load tokenized training data for chatbot
with open("data.pickle", "rb") as f:
    words, labels, training, output = pickle.load(f)

# Build TFlearn Chatbot Model
def build_chatbot_model():
    net = tflearn.input_data(shape=[None, len(training[0])])
    net = tflearn.fully_connected(net, 8)
    net = tflearn.fully_connected(net, 8)
    net = tflearn.fully_connected(net, len(output[0]), activation="softmax")
    net = tflearn.regression(net)
    model = tflearn.DNN(net)
    model.load("MentalHealthChatBotmodel.tflearn")
    return model

chatbot_model = build_chatbot_model()

# Function: Bag of Words
def bag_of_words(s, words):
    bag = [0 for _ in range(len(words))]
    s_words = word_tokenize(s)
    s_words = [stemmer.stem(word.lower()) for word in s_words if word.isalnum()]
    for se in s_words:
        for i, w in enumerate(words):
            if w == se:
                bag[i] = 1
    return np.array(bag)

# Chatbot Response Function
def chatbot_response(message, history):
    """Generates a chatbot response."""
    history = history or []
    try:
        result = chatbot_model.predict([bag_of_words(message, words)])
        idx = np.argmax(result)
        tag = labels[idx]
        response = "I didn't understand that. 🤔"
        for intent in intents_data["intents"]:
            if intent["tag"] == tag:
                response = random.choice(intent["responses"])
                break
    except Exception as e:
        response = f"Error generating response: {str(e)} 💥"

    history.append({"role": "user", "content": message})
    history.append({"role": "assistant", "content": response})
    return history, response

# Emotion Detection Function
emotion_tokenizer = AutoTokenizer.from_pretrained("j-hartmann/emotion-english-distilroberta-base")
emotion_model = AutoModelForSequenceClassification.from_pretrained("j-hartmann/emotion-english-distilroberta-base")

def detect_emotion(user_input):
    pipe = pipeline("text-classification", model=emotion_model, tokenizer=emotion_tokenizer)
    try:
        result = pipe(user_input)
        emotion = result[0]["label"]
        emotion_map = {
            "joy": "😊 Joy",
            "anger": "😠 Anger",
            "sadness": "😢 Sadness",
            "fear": "😨 Fear",
            "surprise": "😲 Surprise",
            "neutral": "😐 Neutral",
        }
        return emotion_map.get(emotion, "Unknown Emotion 🤔")
    except Exception as e:
        return f"Error detecting emotion: {str(e)} 💥"

# Sentiment Analysis Function
sentiment_tokenizer = AutoTokenizer.from_pretrained("cardiffnlp/twitter-roberta-base-sentiment")
sentiment_model = AutoModelForSequenceClassification.from_pretrained("cardiffnlp/twitter-roberta-base-sentiment")

def analyze_sentiment(user_input):
    """Analyze sentiment based on input."""
    inputs = sentiment_tokenizer(user_input, return_tensors="pt")
    try:
        with torch.no_grad():
            outputs = sentiment_model(**inputs)
        sentiment = torch.argmax(outputs.logits, dim=1).item()
        sentiment_map = ["Negative 😔", "Neutral 😐", "Positive 😊"]
        return sentiment_map[sentiment]
    except Exception as e:
        return f"Error in sentiment analysis: {str(e)} 💥"

# Suggestions Based on Emotion
def generate_suggestions(emotion):
    suggestions_map = {
        "😊 Joy": [
            {"Title": "Mindful Meditation 🧘", "Link": "https://www.helpguide.org/meditation"},
            {"Title": "Learn a New Skill ✨", "Link": "https://www.skillshare.com/"},
        ],
        "😢 Sadness": [
            {"Title": "Talk to a Professional 💬", "Link": "https://www.betterhelp.com/"},
            {"Title": "Mental Health Toolkit 🛠️", "Link": "https://www.psychologytoday.com/"},
        ],
        "😠 Anger": [
            {"Title": "Anger Management Tips 🔥", "Link": "https://www.mentalhealth.org.uk"},
            {"Title": "Stress Relieving Exercises 🌿", "Link": "https://www.calm.com/"},
        ],
    }
    return suggestions_map.get(emotion, [{"Title": "General Wellness Resources 🌈", "Link": "https://www.helpguide.org/wellness"}])

# Dummy Nearby Professionals Function
def search_nearby_professionals(location, query):
    """Simulates the search for nearby professionals."""
    if location and query:
        return [
            {"Name": "Wellness Center", "Address": "123 Wellness Way"},
            {"Name": "Mental Health Clinic", "Address": "456 Recovery Road"},
            {"Name": "Therapy Hub", "Address": "789 Peace Avenue"},
        ]
    return []

# Main App Logic
def well_being_app(user_input, location, query, history):
    """Handles chatbot interaction, emotion detection, sentiment analysis, and professional search results."""
    # Chatbot Response
    history, _ = chatbot_response(user_input, history)

    # Emotion Detection
    emotion = detect_emotion(user_input)

    # Sentiment Analysis
    sentiment = analyze_sentiment(user_input)

    # Emotion-based Suggestions
    emotion_name = emotion.split(": ")[-1]
    suggestions = generate_suggestions(emotion_name)
    suggestions_df = pd.DataFrame(suggestions)

    # Nearby Professionals Lookup
    professionals = search_nearby_professionals(location, query)

    return history, sentiment, emotion, suggestions_df, professionals

# Gradio Interface
with gr.Blocks() as interface:
    gr.Markdown("## 🌱 Well-being Companion")
    gr.Markdown("> Empowering Your Health! 💚")

    with gr.Row():
        user_input = gr.Textbox(label="Your Message", placeholder="How are you feeling today? (e.g. I feel happy)")
        location_input = gr.Textbox(label="Location", placeholder="Enter your city (e.g., New York)")
        query_input = gr.Textbox(label="Search Query", placeholder="What are you searching for? (e.g., therapists)")
        submit_button = gr.Button("Submit", variant="primary")

    # Chatbot Section
    chatbot_output = gr.Chatbot(label="Chatbot Interaction", type="messages", value=[])

    # Sentiment and Emotion Outputs
    sentiment_output = gr.Textbox(label="Sentiment Analysis")
    emotion_output = gr.Textbox(label="Emotion Detected")

    # Suggestions Table
    suggestions_output = gr.DataFrame(label="Suggestions", value=[], headers=["Title", "Link"])

    # Professionals Table
    nearby_professionals_output = gr.DataFrame(label="Nearby Professionals", value=[], headers=["Name", "Address"])

    # Connect Inputs to Outputs
    submit_button.click(
        well_being_app,
        inputs=[user_input, location_input, query_input, chatbot_output],
        outputs=[
            chatbot_output,
            sentiment_output,
            emotion_output,
            suggestions_output,
            nearby_professionals_output,
        ],
    )

# Run Gradio Application
interface.launch()