app.py

import streamlit as st
from sentence_transformers import SentenceTransformer
import numpy as np
from PIL import Image
import pinecone
import pinecone_text
import random
import time
from dotenv import load_dotenv
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
import uuid
import json
import pandas as pd
import pydeck as pdk
from urllib.error import URLError

# Initialize session state variables
if 'chat_history_page1' not in st.session_state:
    st.session_state['chat_history_page1'] = []

if 'chat_history_page2' not in st.session_state:
    st.session_state['chat_history_page2'] = []

if 'chat_history_page3' not in st.session_state:
    st.session_state['chat_history_page3'] = []

# This session ID will be unique per user session and consistent across all pages.
if 'session_id' not in st.session_state:
    st.session_state['session_id'] = str(uuid.uuid4())

from pinecone import Pinecone

# Initialize Pinecone client
api_key = os.environ["api_key"]

# configure client
pc = Pinecone(api_key=api_key)

index_name = "canopy--document-uploader"  # Replace with your chosen index name

index = pc.Index(name=index_name)


# Step 1: Clone the Dataset Repository
repo = Repository(
    local_dir="Private_Book",  # Local directory to clone the repository
    repo_type="dataset",  # Specify that this is a dataset repository
    clone_from="Anne31415/Private_Book",  # Replace with your repository URL
    token=os.environ["HUB_TOKEN"]  # Use the secret token to authenticate
)
repo.git_pull()  # Pull the latest changes (if any)


# Step 1: Clone the ChatSet Repository - save all the chats anonymously
repo2 = Repository(
    local_dir="Chat_Store",  # Local directory to clone the repository
    repo_type="dataset",  # Specify that this is a dataset repository
    clone_from="Anne31415/Chat_Store",  # Replace with your repository URL
    token=os.environ["HUB_TOKEN"]  # Use the secret token to authenticate
)
repo.git_pull()  # Pull the latest changes (if any)


# Step 2: Load the PDF File
pdf_path = "Private_Book/KH_Reform230124.pdf"  # Replace with your PDF file path

pdf_path2 = "Private_Book/Buch_23012024.pdf"  

pdf_path3 = "Private_Book/Kosten_Strukturdaten_RAG_vorbereited.pdf"  

api_key = os.getenv("OPENAI_API_KEY")
# Retrieve the API key from st.secrets



@st.cache_resource
def load_vector_store(file_path, store_name, force_reload=False):
    local_repo_path = "Private_Book"
    vector_store_path = os.path.join(local_repo_path, f"{store_name}.pkl")

    # Check if vector store already exists and force_reload is False
    if not force_reload and os.path.exists(vector_store_path):
        with open(vector_store_path, "rb") as f:
            VectorStore = pickle.load(f)
        #st.text(f"Loaded existing vector store from {vector_store_path}")
    else:
        # Load and process the PDF, then create the vector store
        text_splitter = RecursiveCharacterTextSplitter(chunk_size=800, chunk_overlap=200, length_function=len)
        text = load_pdf_text(file_path)
        chunks = text_splitter.split_text(text=text)
        embeddings = OpenAIEmbeddings()
        VectorStore = FAISS.from_texts(chunks, embedding=embeddings)

        # Serialize the vector store
        with open(vector_store_path, "wb") as f:
            pickle.dump(VectorStore, f)
        #st.text(f"Created and saved vector store at {vector_store_path}")

        # Change working directory for Git operations
        original_dir = os.getcwd()
        os.chdir(local_repo_path)
        
        try:
            # Check current working directory and list files for debugging
            #st.text(f"Current working directory: {os.getcwd()}")
            #st.text(f"Files in current directory: {os.listdir()}")
        
            # Adjusted file path for Git command
            repo.git_add(f"{store_name}.pkl")  # Use just the file name
            repo.git_commit(f"Update vector store: {store_name}")
            repo.git_push()
        except Exception as e:
            st.error(f"Error during Git operations: {e}")
        finally:
            # Change back to the original directory
            os.chdir(original_dir)

    return VectorStore


# Utility function to load text from a PDF
def load_pdf_text(file_path):
    pdf_reader = PdfReader(file_path)
    text = ""
    for page in pdf_reader.pages:
        text += page.extract_text() or ""  # Add fallback for pages where text extraction fails
    return text

def load_chatbot():
    #return load_qa_chain(llm=OpenAI(), chain_type="stuff")
    return load_qa_chain(llm=OpenAI(model_name="gpt-3.5-turbo-instruct"), chain_type="stuff")


def display_chat_history(chat_history):
    for chat in chat_history:
        background_color = "#ffeecf" if chat[2] == "new" else "#ffeecf" if chat[0] == "User" else "#ffeecf"
        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)


def handle_no_answer(response):
    no_answer_phrases = [
        "ich weiß es nicht",
        "ich weiß nicht",
        "ich bin mir nicht sicher",
        "es wird nicht erwähnt",
        "Leider kann ich diese Frage nicht beantworten",
        "kann ich diese Frage nicht beantworten",
        "ich kann diese Frage nicht beantworten",
        "ich kann diese Frage leider nicht beantworten",
        "keine information",
        "das ist unklar",
        "da habe ich keine antwort",
        "das kann ich nicht beantworten",
        "i don't know",
        "i am not sure",
        "it is not mentioned",
        "no information",
        "that is unclear",
        "i have no answer",
        "i cannot answer that",
        "unable to provide an answer",
        "not enough context",
        "Sorry, I do not have enough information",
        "I do not have enough information",
        "I don't have enough information",
        "Sorry, I don't have enough context to answer that question.",
        "I don't have enough context to answer that question.",
        "to answer that question.",
        "Sorry",
        "I'm sorry",
        "I don't understand the question",
        "I don't understand"

    ]

    alternative_responses = [
        "Hmm, das ist eine knifflige Frage. Lass uns das gemeinsam erkunden. Kannst du mehr Details geben?",
        "Interessante Frage! Ich bin mir nicht sicher, aber wir können es herausfinden. Hast du weitere Informationen?",
        "Das ist eine gute Frage. Ich habe momentan keine Antwort darauf, aber vielleicht kannst du sie anders formulieren?",
        "Da bin ich überfragt. Kannst du die Frage anders stellen oder mir mehr Kontext geben?",
        "Ich stehe hier etwas auf dem Schlauch. Gibt es noch andere Aspekte der Frage, die wir betrachten könnten?",
        # Add more alternative responses as needed
    ]

    # Check if response matches any phrase in no_answer_phrases
    if any(phrase in response.lower() for phrase in no_answer_phrases):
        return random.choice(alternative_responses)  # Randomly select a response
    return response

def ask_bot(query):
    # Definiere den standardmäßigen Prompt
    standard_prompt = "Antworte immer in der Sprache in der der User schreibt. Formuliere immer ganze freundliche ganze Sätze und biete wenn möglich auch mehr Informationen (aber nicht mehr als 1 Satz mehr). Wenn der User sehr vage schreibt frage nach. Wenn du zu einer bestimmten Frage Daten aus mehreren Jahren hast, nenne das aktuellste und ein weiters. "
    # Kombiniere den standardmäßigen Prompt mit der Benutzeranfrage
    full_query = standard_prompt + query
    return full_query

def save_conversation(chat_histories, session_id):
    base_path = "Chat_Store/conversation_logs"
    if not os.path.exists(base_path):
        os.makedirs(base_path)

    filename = f"{base_path}/{session_id}.json"

    # Check if the log file already exists
    existing_data = {"page1": [], "page2": [], "page3": []}
    if os.path.exists(filename):
        with open(filename, 'r', encoding='utf-8') as file:
            existing_data = json.load(file)

    # Append the new chat history to the existing data for each page
    for page_number, chat_history in enumerate(chat_histories, start=1):
        existing_data[f"page{page_number}"] += chat_history

    with open(filename, 'w', encoding='utf-8') as file:
        json.dump(existing_data, file, indent=4, ensure_ascii=False)

    # Git operations
    try:
        # Change directory to Chat_Store for Git operations
        original_dir = os.getcwd()
        os.chdir('Chat_Store')
    
        # Correct file path relative to the Git repository's root
        git_file_path = f"conversation_logs/{session_id}.json"
    
        repo2.git_add(git_file_path)
        repo2.git_commit(f"Add/update conversation log for session {session_id}")
        repo2.git_push()
    
        # Change back to the original directory
        os.chdir(original_dir)
    except Exception as e:
        st.error(f"Error during Git operations: {e}")

        
def display_session_id():
    session_id = st.session_state['session_id']
    st.sidebar.markdown(f"**Ihre Session ID:** `{session_id}`")
    st.sidebar.markdown("Verwenden Sie diese ID als Referenz bei Mitteilungen oder Rückmeldungen.")

    
def query_pinecone(vector, index, top_k=5):
    # Query Pinecone index for similar vectors
    query_results = index.query(vector=vector, top_k=top_k)
    return query_results["matches"]

from sentence_transformers import SentenceTransformer

# Initialize the Sentence Transformer model
model = SentenceTransformer('all-MiniLM-L6-v2')

def text_to_vector(text):
    # Convert input text to vector
    embedding = model.encode(text)
    return embedding  # No need to convert to list, depending on how you use it later



def page1():
    try:
        hide_streamlit_style = """
                <style>
                #MainMenu {visibility: hidden;}
                footer {visibility: hidden;}
                </style>
                """
        st.markdown(hide_streamlit_style, unsafe_allow_html=True)
    
         # Create columns for layout
        col1, col2 = st.columns([3, 1])  # Adjust the ratio to your liking

        with col1:
            st.title("Alles zur aktuellen Krankenhausreform!")

        with col2:
            # Load and display the image in the right column, which will be the top-right corner of the page
            image = Image.open('BinDoc Logo (Quadratisch).png')
            st.image(image, use_column_width='always')

 
        if not os.path.exists(pdf_path):
            st.error("File not found. Please check the file path.")
            return

        VectorStore = load_vector_store(pdf_path, "KH_Reform_2301", force_reload=False)

        display_chat_history(st.session_state['chat_history_page1'])

        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()

        query = st.text_input("Geben Sie hier Ihre Frage ein / Enter your question here:")

        add_vertical_space(2)  # Adjust as per the desired spacing
        
        # Create two columns for the buttons
        col1, col2 = st.columns(2)
        
        with col1:
            if st.button("Wie viele Ärzte benötigt eine Klinik in der Leistungsgruppe Stammzell-transplantation?"):
                query = "Wie viele Ärzte benötigt eine Klinik in der Leistungsgruppe Stammzell-transplantation?"
            if st.button("Wie viele Leistungsgruppen soll es durch die neue KH Reform geben?"):
                query = ("Wie viele Leistungsgruppen soll es durch die neue KH Reform geben?")
            if st.button("Was sind die hauptsächlichen Änderungsvorhaben der Krankenhausreform?"):
                query = "Was sind die hauptsächlichen Änderungsvorhaben der Krankenhausreform?"

        
        with col2:
            if st.button("Welche technischen Gerätevorgaben und Personalvorgaben muss die LG Allgemeine Chirugie erfüllen?"):
                query = "Welche technischen Gerätevorgaben und Personalvorgaben muss die LG Allgemeine Chirugie erfüllen?"
            if st.button("Was soll die Reform der Notfallversorgung beinhalten?"):
                query = "Was soll die Reform der Notfallversorgung beinhalten?"
            if st.button("Was bedeutet die Vorhaltefinanzierung?"):
                query = "Was bedeutet die Vorhaltefinanzierung?"


    
        if query:
            full_query = ask_bot(query)
            st.session_state['chat_history_page1'].append(("User", query, "new"))
        
            # Start timing
            start_time = time.time()
        
            # Create a placeholder for the response time
            response_time_placeholder = st.empty()
        
            # Include the spinner around all processing and display operations
            with st.spinner('Eve denkt über Ihre Frage nach...'):
                chain = load_chatbot()
                docs = VectorStore.similarity_search(query=query, k=5)
                with get_openai_callback() as cb:
                    response = chain.run(input_documents=docs, question=full_query)
                    response = handle_no_answer(response)
        
                # Stop timing
                end_time = time.time()
        
                # Calculate duration
                duration = end_time - start_time
        
                st.session_state['chat_history_page1'].append(("Eve", response, "new"))
        
                # Combine chat histories from all pages
                all_chat_histories = [
                    st.session_state['chat_history_page1'],
                    st.session_state['chat_history_page2'],
                    st.session_state['chat_history_page3']
                ]
        
                # Save the combined chat histories
                save_conversation(all_chat_histories, st.session_state['session_id'])
        
                # Display new messages at the bottom
                new_messages = st.session_state['chat_history_page1'][-2:]
                for chat in new_messages:
                    background_color = "#ffeecf" if chat[2] == "new" else "#ffeecf" if chat[0] == "User" else "#ffeecf"
                    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)
        
                # Update the response time placeholder after the messages are displayed
                response_time_placeholder.text(f"Response time: {duration:.2f} seconds")
        
            # Clear the input field after the query is made
            query = ""       

        # Mark all messages as old after displaying
        st.session_state['chat_history_page1'] = [(sender, msg, "old") for sender, msg, _ in st.session_state['chat_history_page1']]

    except Exception as e:
        st.error(f"Upsi, an unexpected error occurred: {e}")
        # Optionally log the exception details to a file or error tracking service



def page2():
    try:
        hide_streamlit_style = """
                <style>
                #MainMenu {visibility: hidden;}
                footer {visibility: hidden;}
                </style>
                """
        st.markdown(hide_streamlit_style, unsafe_allow_html=True)
    
         # Create columns for layout
        col1, col2 = st.columns([3, 1])  # Adjust the ratio to your liking

        with col1:
            st.title("Die wichtigsten 100 Kennzahlen und KPIs!")

        with col2:
            # Load and display the image in the right column, which will be the top-right corner of the page
            image = Image.open('BinDoc Logo (Quadratisch).png')
            st.image(image, use_column_width='always')

            
        if not os.path.exists(pdf_path2):
            st.error("File not found. Please check the file path.")
            return

        VectorStore = load_vector_store(pdf_path2, "Buch_2301", force_reload=False)



        display_chat_history(st.session_state['chat_history_page2'])

        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()

        query = st.text_input("Geben Sie hier Ihre Frage ein / Enter your question here:")

        add_vertical_space(2)  # Adjust as per the desired spacing
        
        # Create two columns for the buttons
        col1, col2 = st.columns(2)
        
        with col1:
            if st.button("Erstelle mir eine Liste mit 3 wichtigen Personalkennzahlen im Krankenhaus."):
                query = "Erstelle mir eine Liste mit 3 wichtigen Personalkennzahlen im Krankenhaus."
            if st.button("Wie ist die durchschnittliche Bettenauslastung eines Krankenhauses im Jahr 2020?"):
                query = ("Wie ist die durchschnittliche Bettenauslastung eines Krankenhauses im Jahr 2020?")
            if st.button("Welches sind die Top 1-5 DRGs, die von den Krankenhäusern 2020 abgerechnet wurden?"):
                query = "Welches sind die Top 1-5 DRGs, die von den Krankenhäusern 2020 abgerechnet wurden? "

        
        with col2:
            if st.button("Wie viel Casemixpunkte werden im Median von einer ärztlichen VK ärztlicher Dienst 2020 erbracht?"):
                query = "Wie viel Casemixpunkte werden im Median von einer ärztlichen VK ärztlicher Dienst 2020 erbracht?"
            if st.button("Bitte erstelle mir einer Übersicht des BBFW, Planbetten und CM-relevanten Erlöse eines KH der Grund- und Regelversorgung."):
                query = "Bitte erstelle mir einer Übersicht des BBFW, Planbetten und CM-relevanten Erlöse eines KH der Grund- und Regelversorgung."
            if st.button("Wie viele Patienten eines Grund- und Regelversorgers kommen aus einem 10, 20, 30, 40 Minuten Radius?"):
                query = "Wie viele Patienten eines Grund- und Regelversorgers kommen aus einem 10, 20, 30, 40 Minuten Radius?"

    

        if query:
            full_query = ask_bot(query)
            st.session_state['chat_history_page2'].append(("User", query, "new"))

            # Start timing
            start_time = time.time()

            # Create a placeholder for the response time
            response_time_placeholder = st.empty()
            
            with st.spinner('Eve denkt über Ihre Frage nach...'):
                chain = load_chatbot()
                docs = VectorStore.similarity_search(query=query, k=5)
                with get_openai_callback() as cb:
                    response = chain.run(input_documents=docs, question=full_query)
                    response = handle_no_answer(response)  # Process the response through the new function


                    
            # Stop timing
            end_time = time.time()
            
            # Calculate duration
            duration = end_time - start_time
            
            st.session_state['chat_history_page2'].append(("Eve", response, "new"))


            # Combine chat histories from all pages
            all_chat_histories = [
                st.session_state['chat_history_page1'],
                st.session_state['chat_history_page2'],
                st.session_state['chat_history_page3']
            ]

            # Save the combined chat histories
            save_conversation(all_chat_histories, st.session_state['session_id'])

            # Display new messages at the bottom
            new_messages = st.session_state['chat_history_page2'][-2:]
            for chat in new_messages:
                background_color = "#ffeecf" if chat[2] == "new" else "#ffeecf" if chat[0] == "User" else "#ffeecf"
                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)
    
                # Update the response time placeholder after the messages are displayed
                response_time_placeholder.text(f"Response time: {duration:.2f} seconds")
                
            # Clear the input field after the query is made
            query = ""

        # Mark all messages as old after displaying
        st.session_state['chat_history_page2'] = [(sender, msg, "old") for sender, msg, _ in st.session_state['chat_history_page2']]

    except Exception as e:
        st.error(f"Upsi, an unexpected error occurred: {e}")
        # Optionally log the exception details to a file or error tracking service



def page3():
    try:
        st.markdown("""
            <style>
            #MainMenu {visibility: hidden;}
            footer {visibility: hidden;}
            </style>
            """, unsafe_allow_html=True)
    
        col1, col2 = st.columns([3, 1])
        with col1:
            st.title("Kosten- und Strukturdaten der Krankenhäuser")
        with col2:
            image_path = 'BinDoc Logo (Quadratisch).png'
            if os.path.exists(image_path):
                image = Image.open(image_path)
                st.image(image, use_column_width='always')

        display_chat_history(st.session_state.get('chat_history_page3', []))

        query = st.text_input("Geben Sie hier Ihre Frage ein / Enter your question here:")

        if query:
            query_vector = text_to_vector(query)
            # Ensure the vector is in the correct format for Pinecone
            query_vector = query_vector.tolist() if isinstance(query_vector, np.ndarray) else query_vector
            
            # Query the Pinecone index
            results = pc.query(index=index_name, vector=query_vector, top_k=5)

            # Process and display results
            for result in results['matches']:
                matched_id = result['id']
                score = result['score']
                # Assuming you have metadata to display, adjust as necessary
                st.write(f"Matched ID: {matched_id}, Score: {score}")

            # Update chat history
            st.session_state['chat_history_page3'].append(("User", query))
            for result in results['matches']:
                st.session_state['chat_history_page3'].append(("Result", f"Matched ID: {result['id']}, Score: {result['score']}"))

            # Save the updated chat history
            save_conversation(st.session_state['chat_history_page3'], st.session_state['session_id'])

    except Exception as e:
        st.error(f"An unexpected error occurred: {e}")


def page4():
    try: 
        st.header(":mailbox: Kontakt & Feedback!")
        st.markdown("Ihre Session-ID finden Sie auf der linken Seite!")

        contact_form = """
        <form action="https://formsubmit.co/[email protected]" method="POST">
             <input type="hidden" name="_captcha" value="false">
             <input type="text" name="Session-ID" placeholder="Your Session-ID goes here" required>
             <input type="email" name="email" placeholder="Your email" required>
             <textarea name="message" placeholder="Your message here"></textarea>
             <form action="https://formsubmit.co/your-random-string" method="POST" />
             <button type="submit">Send</button>
        </form>
        """
        
        st.markdown(contact_form, unsafe_allow_html=True)
        
        # Use Local CSS File
        def local_css(file_name):
            with open(file_name) as f:
                st.markdown(f"<style>{f.read()}</style>", unsafe_allow_html=True)
        
        
        local_css("style.css")
        
    except Exception as e:
        st.error(f"Upsi, an unexpected error occurred: {e}")
        # Optionally log the exception details to a file or error tracking service


def display_session_id():
    session_id = st.session_state['session_id']
    st.sidebar.markdown(f"**Your Session ID:** `{session_id}`")
    st.sidebar.markdown("Verwenden Sie diese ID als Referenz bei Mitteilungen oder Rückmeldungen.")

# Main function
def main():
    # Sidebar content
    with st.sidebar:
        st.title('BinDoc GmbH')
        st.markdown("Tauchen Sie ein in eine revolutionäre Erfahrung mit BinDocs Chat-App - angetrieben von fortschrittlichster KI-Technologie.")
        add_vertical_space(1)
        page = st.sidebar.selectbox("Wählen Sie eine Seite aus:", ["Krankenhausreform!", "Kennzahlen und KPIs!", "Kosten- und Strukturdaten", "Kontakt & Feedback!"])
        add_vertical_space(4)
        display_session_id()  # Display the session ID in the sidebar
        st.write('Made with ❤️ by BinDoc GmbH')

        
    # Main area content based on page selection
    if page == "Krankenhausreform!":
        page1()
    elif page == "Kennzahlen und KPIs!":
        page2()
    elif page == "Kosten- und Strukturdaten":
        page3()
    elif page == "Kontakt & Feedback!":
        page4()


if __name__ == "__main__":
    main()