import gradio as gr
import pickle
import re
import nltk
from nltk.corpus import stopwords

# Download stopwords if not already present
nltk.download('stopwords', quiet=True)
stop_words = set(stopwords.words('english'))

def clean_text(text):
    """Clean and preprocess text similar to training preprocessing"""
    text = text.lower()  # Convert to lowercase
    text = re.sub(r'<.*?>', '', text)  # Remove HTML tags
    text = re.sub(r'[^a-z\s]', '', text)  # Remove non-alphabetic characters
    words = text.split()  # Split into words
    words = [w for w in words if w not in stop_words]  # Remove stopwords
    return ' '.join(words)  # Join words back

# Load model and vectorizer
with open("model.pkl", "rb") as f:
    model = pickle.load(f)
with open("vectorizer.pkl", "rb") as f:
    vectorizer = pickle.load(f)

def analyze_review(review_text):
    if not review_text.strip():
        return "⚠️ Please enter a review."
    try:
        cleaned_text = clean_text(review_text)
        text_vectorized = vectorizer.transform([cleaned_text])
        prediction = model.predict(text_vectorized)[0]
        if prediction == 1:
            return "✅ Positive Review"
        else:
            return "❌ Negative Review"
    except Exception as e:
        return f"Error during prediction: {e}"

# Create Gradio interface
iface = gr.Interface(
    fn=analyze_review,
    inputs=gr.Textbox(lines=6, placeholder="Enter your movie review here..."),
    outputs="text",
    title="🎬 IMDB Sentiment Analysis",
    description="Enter a movie review and get a Positive/Negative sentiment prediction."
)

if __name__ == "__main__":
    iface.launch()