app.py

import streamlit as st
from PyPDF2 import PdfReader
from langchain.text_splitter import RecursiveCharacterTextSplitter
import os
from langchain_google_genai import GoogleGenerativeAIEmbeddings
import google.generativeai as genai
from langchain.vectorstores import FAISS
from langchain_google_genai import ChatGoogleGenerativeAI
from langchain.chains.question_answering import load_qa_chain
from langchain.prompts import PromptTemplate
from dotenv import load_dotenv

load_dotenv()
os.getenv("GOOGLE_API_KEY")
genai.configure(api_key=os.getenv("GOOGLE_API_KEY"))


def get_pdf_text(pdf_docs):
    """
    Extracts text from a list of PDF documents.

    Args:
        pdf_docs: A list of PDF documents.

    Returns:
        A string containing the extracted text from all PDF documents.
    """

    text = ""
    for pdf in pdf_docs:
        pdf_reader = PdfReader(pdf)
        for page in pdf_reader.pages:
            text += page.extract_text()
    return text


def get_text_chunks(text):
    """
    Splits a large text into smaller chunks for efficient processing.

    Args:
        text: A string containing the text to be split.

    Returns:
        A list of text chunks.
    """

    text_splitter = RecursiveCharacterTextSplitter(chunk_size=10000, chunk_overlap=1000)
    chunks = text_splitter.split_text(text)
    return chunks


def get_vector_store(text_chunks):
    """
    Creates a vector store from a list of text chunks.

    Args:
        text_chunks: A list of text chunks.

    Returns:
        A FAISS vector store.
    """

    embeddings = GoogleGenerativeAIEmbeddings(model="models/embedding-001")
    vector_store = FAISS.from_texts(text_chunks, embedding=embeddings)
    vector_store.save_local("faisss_index")
    return vector_store


def get_conversional_chain():
    """
    Creates a conversational chain for question answering.

    Returns:
        A LangChain question-answering chain.
    """

    prompt_template = """
    Answer the question as detailed as possible from the provided context, make sure to provide all the details. If the answer is not available in the context, just say, "answer is not available in the context", don't provide the wrong answer.

    Context:
    {context}?

    Question:
    {question}

    Answer:
    """

    model = ChatGoogleGenerativeAI(model="gemini-pro", temperature=0.3)
    prompt = PromptTemplate(template=prompt_template, input_variables=["context", "question"])
    chain = load_qa_chain(model, chain_type="stuff", prompt=prompt)
    return chain


def user_input(user_question, processed_pdf_text):
    """
    Processes user input and generates a response using the conversational chain,
    providing both the user's question and the processed PDF text as context.

    Args:
        user_question: The user's question.
        processed_pdf_text: The processed text extracted from the uploaded PDF files.

    Returns:
        The generated response from the conversational chain.
    """

    embeddings = GoogleGenerativeAIEmbeddings(model="models/embedding-001")
    new_db = FAISS.load_local("faisss_index", embeddings)
    docs = new_db.similarity_search(user_question)

    chain = get_conversional_chain()

    # Combine user question and processed PDF text as context
    context = f"{processed_pdf_text}\n\nQuestion: {user_question}"

    response = chain({"input_documents": docs, "question": user_question, "context": context}, return_only_outputs=True)

    print(response)
    st.write("Reply: ", response["output_text"])


def main():
    """
    Main function for the Streamlit app.
    """

    st.set_page_config("Chat With Multiple PDF")
    st.header("Chat with PDF's powered by Gemini 🙋‍♂️")

    user_question = st.text_input("Ask a Question from the PDF Files")

    if user_question:
        if st.session_state.get("pdf_docs"):
            processed_pdf_text = get_pdf_text(st.session_state["pdf_docs"])
            user_input(user_question, processed_pdf_text)
        else:
            st.error("Please upload PDF files first.")

    with st.sidebar:
        st.title("Menu:")
        pdf_docs = st.file_uploader("Upload Files & Click Submit to Proceed", type="pdf", accept_multiple_files=True)
        if st.button("Submit & Process"):
            with st.spinner("Processing..."):
                raw_text = ""
                text_chunks = []
                for pdf in pdf_docs:
                    pdf_reader = PdfReader(pdf)
                    for page in pdf_reader.pages:
                        raw_text += page.extract_text()
                text_chunks = get_text_chunks(raw_text)
                vector_store = get_vector_store(text_chunks)
                chain = get_conversional_chain()
                st.session_state["pdf_docs"] = pdf_docs
                st.session_state["text_chunks"] = text_chunks
                st.session_state["vector_store"] = vector_store
                st.session_state["chain"] = chain
                st.success("PDFs processed successfully!")

        if st.button("Reset"):
            st.session_state["pdf_docs"] = []
            st.session_state["text_chunks"] = []
            st.session_state["vector_store"] = None
            st.session_state["chain"] = None
            st.experimental_rerun()

        if st.session_state.get("pdf_docs"):
            st.subheader("Uploaded Files:")
            for i, pdf_doc in enumerate(st.session_state["pdf_docs"]):
                st.write(f"{i+1}. {pdf_doc.name}")


if __name__ == "__main__":
    main()