test.txt

# import os
# import pickle
# import streamlit as st
# from PyPDF2 import PdfReader
# from dotenv import load_dotenv
# from streamlit_extras.add_vertical_space import add_vertical_space
# from langchain.llms import OpenAI
# from langchain.vectorstores import FAISS
# from langchain.chat_models import ChatOpenAI
# from langchain.callbacks import get_openai_callback
# from langchain.embeddings.openai import OpenAIEmbeddings
# from langchain.chains.question_answering import load_qa_chain
# from langchain.text_splitter import RecursiveCharacterTextSplitter
# from flask import Flask, request, jsonify


# app = Flask(__name__)


# @app.route('/pdf-reader', methods=['POST'])
# def pdf_reader():
#     data = request.json

#     # Extract parameters from the data (e.g., PDF file URL or content)
#     pdf_url = data.get('pdf_url')
#     query = data.get('query')

#     response_data = {
#         'result': "Your response here",
#         # Include any relevant data to send back to Munir-GPT
#     }

#     return jsonify(response_data)


# st.header("Chat with PDF 💬")
# # Sidebar contents
# with st.sidebar:
#     st.title('Munir PDF-Reader')
#     st.markdown('''
#     ## About
#     Hey there! I'm an AI-powered PDF tool ready to extract meaning from documents and have conversations about the content. Just upload a file and you can query me just like you would query a human who read the material. Give it a try!
#     ''')
#     add_vertical_space(26)
#     st.header(
#         'Extension of Munir-GPT')


# load_dotenv()


# def main():
#     # Upload the PDF File
#     pdf = st.file_uploader("Upload a PDF file", type=["pdf"])

#     # Read the PDF file
#     if pdf is not None:
#         pdf_reader = PdfReader(pdf)

#         # Extract the text from the PDF
#         text = ""
#         for page in pdf_reader.pages:
#             text += page.extract_text()

#         text_splitter = RecursiveCharacterTextSplitter(
#             chunk_size=1000,
#             chunk_overlap=200,
#             length_function=len,
#         )
#         chunks = text_splitter.split_text(text=text)

#         # Embeddings
#         store_name = pdf.name[:-4]

#         # Check if the store exists
#         if os.path.exists(f"{store_name}.pkl"):
#             with open(f"{store_name}.pkl", "rb") as f:
#                 VectorStore = pickle.load(f)
#         else:  # Create the store
#             embeddings = OpenAIEmbeddings()
#             VectorStore = FAISS.from_texts(chunks, embedding=embeddings)
#             with open(f"{store_name}.pkl", "wb") as f:
#                 pickle.dump(VectorStore, f)

#         # Ask questions
#         query = st.text_input("Ask questions about your document:")

#         if query:
#             docs = VectorStore.similarity_search(query=query, k=3)
#             # Load the Open AI Language Model
#             llm = OpenAI(model_name='gpt-3.5-turbo')
#             chain = load_qa_chain(llm=llm, chain_type="stuff")
#             with get_openai_callback() as cb:
#                 response = chain.run(input_documents=docs, question=query)
#                 print(cb)
#             st.write(response)


# if __name__ == '__main__':
#     app.run(debug=True)