import tempfile
import os
import streamlit as st
from langchain.document_loaders import PyPDFLoader
from langchain.vectorstores import FAISS
from langchain.embeddings import Embedding
from langchain_community.embeddings.groq import GroqEmbedding

# Function to process PDF
def process_pdf(file):
    # Save the uploaded file into a temporary file
    with tempfile.NamedTemporaryFile(delete=False, suffix=".pdf") as tmpfile:
        tmpfile.write(file.read())  # Write the uploaded file's content
        tmpfile_path = tmpfile.name  # Get the file path
    return tmpfile_path

# Main function to run the app
def main():
    st.title("PDF Embedding and Query System")
    
    uploaded_file = st.file_uploader("Upload a PDF file", type=["pdf"])
    
    if uploaded_file is not None:
        # Process the uploaded PDF file
        tmp_file_path = process_pdf(uploaded_file)
        
        # Load the PDF content
        loader = PyPDFLoader(tmp_file_path)
        documents = loader.load()
        
        # Use Groq embeddings (assuming Groq API key is set correctly)
        embeddings = GroqEmbedding(api_key="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")
        
        # Create a vector database
        vector_db = FAISS.from_documents(documents, embeddings)
        
        # Perform search or other actions
        query = st.text_input("Enter a query to search:")
        if query:
            results = vector_db.similarity_search(query, k=5)
            for result in results:
                st.write(result["text"])

# Run the app
if __name__ == "__main__":
    main()