import streamlit as st
import nltk
import spacy
import matplotlib.pyplot as plt
from transformers import pipeline
import random
import subprocess

# Ensure spaCy model is installed
try:
    nlp = spacy.load("en_core_web_sm")
except OSError:
    subprocess.run(["python", "-m", "spacy", "download", "en_core_web_sm"])
    nlp = spacy.load("en_core_web_sm")

# Ensure NLTK resources are available
nltk.download("vader_lexicon")
from nltk.sentiment import SentimentIntensityAnalyzer
sia = SentimentIntensityAnalyzer()

# Load English emotion detection pipeline
emotion_pipeline = pipeline("text-classification", model="j-hartmann/emotion-english-distilroberta-base", return_all_scores=True)

# Load multilingual sentiment model (returns star ratings 1-5 for non-English texts)
multilingual_pipeline = pipeline("text-classification", model="nlptown/bert-base-multilingual-uncased-sentiment")

# Sample texts
sample_texts = [
    "The digital world is transforming the way we read and engage with text.",
    "Reading is an essential skill that shapes our understanding of the world.",
    "AI-driven education tools can personalize the learning experience for students."
]

# Supported languages for multilingual sentiment analysis
supported_languages = {
    "English": "en",
    "Dutch": "nl",
    "French": "fr",
    "German": "de",
    "Spanish": "es",
    "Italian": "it"
}

# Streamlit UI
st.title("📖 AI-Powered Adaptive Reading Engagement")
st.write("Analyze how users engage with digital reading using AI-powered insights.")

# Language selection
selected_language = st.selectbox("Select a language:", list(supported_languages.keys()))

# Use session state to store text input
if "user_text" not in st.session_state:
    st.session_state.user_text = ""

# Dropdown menu with a new option for "Choose Your Own Text"
text_option = st.selectbox("Choose a sample text or enter your own:", ["Choose Your Own Text"] + sample_texts)

# Text input limit
MAX_WORDS = 100  # Set the max word limit

# If user selects a sample text, use it. Otherwise, allow manual input.
if text_option == "Choose Your Own Text":
    text = st.text_area("Enter your own text:", st.session_state.user_text, height=150)
else:
    text = sample_texts[sample_texts.index(text_option)]  # Select the chosen sample text

# Show word count & limit warning only if the user is entering their own text
word_count = len(text.split())
if text_option == "Choose Your Own Text":
    st.markdown(f'<p style="font-size:12px; color:gray;">⚠️ Limit: {MAX_WORDS} words max.</p>', unsafe_allow_html=True)

    if word_count > MAX_WORDS:
        st.warning(f"⚠️ Your input has {word_count} words. Please limit it to {MAX_WORDS} words.")

# Save user input in session state only if within limit
if text_option == "Choose Your Own Text" and word_count <= MAX_WORDS:
    st.session_state.user_text = text

# Function to convert star ratings (1-5) into emotion categories
def convert_star_rating_to_emotion(stars):
    star_to_emotion = {
        "1 star": ("very negative", 0.1),
        "2 stars": ("negative", 0.3),
        "3 stars": ("neutral", 0.5),
        "4 stars": ("positive", 0.7),
        "5 stars": ("very positive", 0.9)
    }
    return star_to_emotion.get(stars, ("unknown", 0.0))

# Function to generate AI-driven feedback
def generate_feedback(sentiment_score, top_emotion):
    if sentiment_score['pos'] > 0.6 or top_emotion in ["very positive", "positive", "joy", "optimism"]:
        return "😊 Your reading is positive! This text might boost engagement and motivation."
    elif sentiment_score['neg'] > 0.6 or top_emotion in ["very negative", "negative", "anger", "disgust"]:
        return "😔 The text seems to have a strong negative tone. Consider balancing it with positive information."
    elif top_emotion in ["neutral"]:
        return "🤔 The text is neutral. You might want to explore different perspectives."
    else:
        return "⚖️ The analysis is inconclusive. Try refining your text for clearer sentiment."

# Sentiment Analysis
if st.button("Analyze Engagement"):
    if text and (text_option != "Choose Your Own Text" or word_count <= MAX_WORDS):
        # Sentiment Analysis (English)
        sentiment_score = sia.polarity_scores(text)

        # Emotion Detection
        emotion_scores = {}
        if selected_language == "English":
            emotion_results = emotion_pipeline(text)
            top_emotion = max(emotion_results[0], key=lambda x: x['score'])['label']
            for e in emotion_results[0]:
                emotion_scores[e["label"]] = e["score"]
        else:
            # Multilingual Sentiment Analysis (Converts Star Ratings to Emotion)
            multilingual_results = multilingual_pipeline(text)
            top_emotion, emotion_intensity = convert_star_rating_to_emotion(multilingual_results[0]["label"])
            emotion_scores = {top_emotion: emotion_intensity}

        # Display Sentiment
        st.subheader("📊 Sentiment Analysis")
        st.write(f"Positive: {sentiment_score['pos'] * 100:.2f}%, Negative: {sentiment_score['neg'] * 100:.2f}%, Neutral: {sentiment_score['neu'] * 100:.2f}%")

        # Display Emotion
        st.subheader("🎭 Emotion Detection")
        st.write(f"Top Emotion: **{top_emotion.capitalize()}**")

        # AI-Generated Feedback
        st.subheader("💡 AI-Generated Feedback")
        feedback = generate_feedback(sentiment_score, top_emotion)
        st.write(feedback)

        # 📊 **Visual Display Chart (Fixed & Restored)**
        if emotion_scores:
            st.subheader("📈 Emotion Intensity Chart")
            fig, ax = plt.subplots()
            ax.bar(emotion_scores.keys(), emotion_scores.values(), color=["red", "orange", "yellow", "green", "blue"])
            ax.set_ylabel("Emotion Intensity")
            ax.set_xlabel("Emotions")
            ax.set_title("Emotion Analysis")
            st.pyplot(fig)

    elif text_option == "Choose Your Own Text" and word_count > MAX_WORDS:
        st.warning("⚠️ Please reduce the text length to analyze.")
    else:
        st.warning("⚠️ Please enter a text to analyze.")