import streamlit as st
from transformers import pipeline, AutoModelForSequenceClassification, AutoTokenizer
import torch
import numpy as np

def main():
    # Load the spam detection pipeline
    spam_pipeline = pipeline("text-classification", model="cybersectony/phishing-email-detection-distilbert_v2.4.1")

    # Load the sentiment model and tokenizer
    sentiment_model = AutoModelForSequenceClassification.from_pretrained("ISOM5240GP4/email_sentiment", num_labels=2)
    tokenizer = AutoTokenizer.from_pretrained("distilbert-base-uncased")

    st.title("Email Analysis Tool")
    st.write("Enter an email body below or select a sample to analyze its spam status and sentiment.")

    if "email_body" not in st.session_state:
        st.session_state.email_body = ""

    email_body = st.text_area("Email Body", value=st.session_state.email_body, height=200, key="email_input")

    # Sample emails
    sample_spam = """[Spam email content]"""
    sample_not_spam_positive = """[Positive email content]"""
    sample_not_spam_negative = """[Negative email content]"""

    col1, col2, col3 = st.columns(3)
    with col1:
        if st.button("Spam Email"):
            st.session_state.email_body = sample_spam
            st.rerun()
    with col2:
        if st.button("Not Spam, Positive"):
            st.session_state.email_body = sample_not_spam_positive
            st.rerun()
    with col3:
        if st.button("Not Spam, Negative"):
            st.session_state.email_body = sample_not_spam_negative
            st.rerun()

    if st.button("Analyze Email"):
        if email_body:
            spam_result = spam_pipeline(email_body)
            spam_label = spam_result[0]["label"]
            spam_confidence = spam_result[0]["score"]

            if spam_label == "LABEL_1":
                st.write(f"This is a spam email (Confidence: {spam_confidence:.2f}). No follow-up needed.")
            else:
                inputs = tokenizer(email_body, padding=True, truncation=True, return_tensors='pt')
                outputs = sentiment_model(**inputs)
                predictions = torch.nn.functional.softmax(outputs.logits, dim=-1)
                predictions = predictions.cpu().detach().numpy()
                sentiment_index = np.argmax(predictions)
                sentiment_confidence = predictions[0][sentiment_index]

                sentiment = "Positive" if sentiment_index == 1 else "Negative"
                if sentiment == "Positive":
                    st.write(f"This email is not spam (Confidence: {spam_confidence:.2f}).")
                    st.write(f"Sentiment: {sentiment} (Confidence: {sentiment_confidence:.2f}). No follow-up needed.")
                else:
                    st.write(f"This email is not spam (Confidence: {spam_confidence:.2f}).")
                    st.write(f"Sentiment: {sentiment} (Confidence: {sentiment_confidence:.2f}).")
                    st.write("**This email needs follow-up as it is not spam and has negative sentiment.**")
        else:
            st.write("Please enter an email body or select a sample to analyze.")

if __name__ == "__main__":
    main()