import streamlit as st
from PyPDF2 import PdfReader
import pandas as pd
from transformers import pipeline
import random

# Load the Hugging Face model for text generation and summarization (FLAN-T5 or T5-Small)
@st.cache_resource
def load_text_generator():
    return pipeline("text2text-generation", model="google/flan-t5-base")  # Efficient and professional model

text_generator = load_text_generator()

# Function to extract text from a PDF file
def extract_pdf_content(pdf_file):
    reader = PdfReader(pdf_file)
    content = ""
    for page in reader.pages:
        content += page.extract_text()
    return content

# Function to extract content from a text file
def extract_text_file(file):
    return file.read().decode("utf-8")

# Function to load a CSV file
def read_csv_file(file):
    df = pd.read_csv(file)
    return df.to_string()

# Function to search for a topic in the extracted content
def search_topic_in_content(content, topic):
    sentences = content.split(".")  # Break content into sentences
    topic_sentences = [s for s in sentences if topic.lower() in s.lower()]  # Filter sentences containing the topic
    return ". ".join(topic_sentences) if topic_sentences else None

# Function to generate structured content using Hugging Face model
def generate_professional_content(topic):
    prompt = f"Explain '{topic}' in bullet points, highlighting the key concepts, examples, and applications in a professional manner for electrical engineering students."
    response = text_generator(prompt, max_length=300, num_return_sequences=1)
    return response[0]['generated_text']

# Function to generate a quiz question
def generate_quiz(topic):
    questions = [
        f"What is the fundamental principle of {topic}?",
        f"Name a practical application of {topic}.",
        f"What are the key equations associated with {topic}?",
        f"Describe how {topic} is used in real-world scenarios.",
        f"List common problems and solutions related to {topic}.",
    ]
    return random.choice(questions)

# Streamlit App
st.title("Generative AI for Electrical Engineering Education")
st.sidebar.header("AI-Based Tutor")

# File upload section
uploaded_file = st.sidebar.file_uploader("Upload Study Material (PDF/TXT/CSV)", type=["pdf", "txt", "csv"])
topic = st.sidebar.text_input("Enter a topic (e.g., Newton's Third Law, DC Motors)")

# Process uploaded file
content = ""
if uploaded_file:
    file_type = uploaded_file.name.split(".")[-1]

    if file_type == "pdf":
        content = extract_pdf_content(uploaded_file)
    elif file_type == "txt":
        content = extract_text_file(uploaded_file)
    elif file_type == "csv":
        content = read_csv_file(uploaded_file)

    st.sidebar.success(f"{uploaded_file.name} uploaded successfully!")
    st.write("**Extracted Content from File:**")
    st.write(content[:1000] + "...")  # Display a snippet of the content

# Generate study material
if st.button("Generate Study Material"):
    if topic:
        st.header(f"Study Material: {topic}")
        # Extract relevant content from the uploaded material
        filtered_content = search_topic_in_content(content, topic) if content else ""
        if filtered_content:
            st.write("**Relevant Extracted Content from Uploaded Material:**")
            st.write(filtered_content)
        else:
            st.warning("No relevant content found in the uploaded material. Generating AI-based content instead.")
            ai_content = generate_professional_content(topic)
            st.write("**AI-Generated Content:**")
            st.write(ai_content)
    else:
        st.warning("Please enter a topic!")

# Generate quiz
if st.button("Generate Quiz"):
    if topic:
        st.header("Quiz Question")
        question = generate_quiz(topic)
        st.write(question)
    else:
        st.warning("Please enter a topic!")