app.py

import streamlit as st
from dotenv import load_dotenv
import pinecone
import pickle
from huggingface_hub import Repository
from PyPDF2 import PdfReader
from streamlit_extras.add_vertical_space import add_vertical_space
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain.embeddings.openai import OpenAIEmbeddings
from langchain.vectorstores import FAISS
from langchain.llms import OpenAI
from langchain.chains.question_answering import load_qa_chain
from langchain.callbacks import get_openai_callback
import os

# Load all necessary environment variables at the beginning of the script
PINECONE_API_KEY = os.getenv("PINECONE_API_KEY")
OPENAI_API_KEY = os.getenv("OPENAI_API_KEY")

pinecone.init(PINECONE_API_KEY="PINECONE_API_KEY")

INDEX_NAME = "pdfbot1"
if INDEX_NAME not in pinecone.list_indexes():
    pinecone.create_index(name=INDEX_NAME, metric="cosine", shards=1)



# Step 1: Clone the Dataset Repository
repo = Repository(
    local_dir="Private_Book",
    repo_type="dataset",
    clone_from="Anne31415/Private_Book",
    token=os.environ["HUB_TOKEN"]
)
repo.git_pull()


# Step 2: Load the PDF File
pdf_file_path = "Private_Book/Glossar_HELP_DESK_combi.pdf"  # Replace with your PDF file path

with st.sidebar:
    st.title('BinDoc GmbH')
    st.markdown("Experience revolutionary interaction with BinDocs Chat App, leveraging state-of-the-art AI technology.")
    
    add_vertical_space(1)  # Adjust as per the desired spacing
    
    st.markdown("""
    Hello! I’m here to assist you with:<br><br>
    📘 **Glossary Inquiries:**<br>
    I can clarify terms like "DiGA", "AOP", or "BfArM", providing clear and concise explanations to help you understand our content better.<br><br>
    🆘 **Help Page Navigation:**<br>
    Ask me if you forgot your password or want to know more about topics related to the platform.<br><br>
    📰 **Latest Whitepapers Insights:**<br>
    Curious about our recent publications? Feel free to ask about our latest whitepapers!<br><br>
    """, unsafe_allow_html=True)
    
    add_vertical_space(1)  # Adjust as per the desired spacing

    st.write('Made with ❤️ by BinDoc GmbH')



def load_pdf(file_path):
    pdf_reader = PdfReader(file_path)
    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)

    store_name, _ = os.path.splitext(os.path.basename(file_path))
    if os.path.exists(f"{store_name}.pkl"):
        with open(f"{store_name}.pkl", "rb") as f:
            VectorStore = pickle.load(f)
    else:
        embeddings = OpenAIEmbeddings()
        VectorStore = FAISS.from_texts(chunks, embedding=embeddings)
        with open(f"{store_name}.pkl", "wb") as f:
            pickle.dump(VectorStore, f)
    vector_dict = {str(i): vector for i, vector in enumerate(VectorStore.vectors)}
    pinecone.upsert(items=vector_dict, index_name=INDEX_NAME)
    return VectorStore




def load_chatbot():
    return load_qa_chain(llm=OpenAI(), chain_type="stuff")

def main():

    hide_streamlit_style = """
            <style>
            #MainMenu {visibility: hidden;}
            footer {visibility: hidden;}
            </style>
            """
    st.markdown(hide_streamlit_style, unsafe_allow_html=True)


    # Main content
    st.title("Welcome to BinDocs ChatBot! 🤖")
    
    # Directly specifying the path to the PDF file
    pdf_path = pdf_file_path
    if not os.path.exists(pdf_path):
        st.error("File not found. Please check the file path.")
        return

    if "chat_history" not in st.session_state:
        st.session_state['chat_history'] = []

    display_chat_history(st.session_state['chat_history'])

    st.write("<!-- Start Spacer -->", unsafe_allow_html=True)
    st.write("<div style='flex: 1;'></div>", unsafe_allow_html=True)
    st.write("<!-- End Spacer -->", unsafe_allow_html=True)

    new_messages_placeholder = st.empty()

    if pdf_path is not None:
        query = st.text_input("Ask questions about your PDF file (in any preferred language):")

        if st.button("Was genau ist ein Belegarzt?"):
            query = "Was genau ist ein Belegarzt?"
        if st.button("Wofür wird die Alpha-ID verwendet?"):
            query = "Wofür wird die Alpha-ID verwendet?"
        if st.button("Was sind die Vorteile des ambulanten operierens?"):
            query = "Was sind die Vorteile des ambulanten operierens?"
        if st.button("Was kann ich mit dem Prognose-Analyse Toll machen?"):
            query = "Was kann ich mit dem Prognose-Analyse Toll machen?"
        if st.button("Was sagt mir die Farbe der Balken der Bevölkerungsentwicklung?"):
            query = "Was sagt mir die Farbe der Balken der Bevölkerungsentwicklung?"
        if st.button("Ich habe mein Meta Password vergessen, wie kann ich es zurücksetzen?"):
            query = ("Ich habe mein Meta Password vergessen, wie kann ich es zurücksetzen?")

            
        if st.button("Ask") or (not st.session_state['chat_history'] and query) or (st.session_state['chat_history'] and query != st.session_state['chat_history'][-1][1]):
            st.session_state['chat_history'].append(("User", query, "new"))

            loading_message = st.empty()
            loading_message.text('Bot is thinking...')

            VectorStore = load_pdf(pdf_path)
            chain = load_chatbot()
            docs = VectorStore.similarity_search(query=query, k=3)

            # Searching for similar documents in Pinecone
            query_vector = embeddings.embed_text(query)
            search_results = pinecone.query(queries=[query_vector], index_name=INDEX_NAME, top_k=3)
            # Extracting document ids from Pinecone's results
            doc_ids = [int(item.id) for item in search_results.results[0].matches]
            # Retrieving the actual document texts based on the ids
            docs = [texts[id] for id in doc_ids]

            with get_openai_callback() as cb:
                response = chain.run(input_documents=docs, question=query)

            st.session_state['chat_history'].append(("Bot", response, "new"))

            # Display new messages at the bottom
            new_messages = st.session_state['chat_history'][-2:]
            for chat in new_messages:
                background_color = "#FFA07A" if chat[2] == "new" else "#acf" if chat[0] == "User" else "#caf"
                new_messages_placeholder.markdown(f"<div style='background-color: {background_color}; padding: 10px; border-radius: 10px; margin: 10px;'>{chat[0]}: {chat[1]}</div>", unsafe_allow_html=True)

            # Scroll to the latest response using JavaScript
            st.write("<script>document.getElementById('response').scrollIntoView();</script>", unsafe_allow_html=True)

            loading_message.empty()

            # Clear the input field by setting the query variable to an empty string
            query = ""

        # Mark all messages as old after displaying
        st.session_state['chat_history'] = [(sender, msg, "old") for sender, msg, _ in st.session_state['chat_history']]



def display_chat_history(chat_history):
    for chat in chat_history:
        background_color = "#FFA07A" if chat[2] == "new" else "#acf" if chat[0] == "User" else "#caf"
        st.markdown(f"<div style='background-color: {background_color}; padding: 10px; border-radius: 10px; margin: 10px;'>{chat[0]}: {chat[1]}</div>", unsafe_allow_html=True)

if __name__ == "__main__":
    main()