GraphicAINovel

Runtime error

File size: 57,958 Bytes

import streamlit as st
import os
import json
import random

    
# Imports
import base64
import glob
import json
import math
import openai
import os
import pytz
import re
import requests
import streamlit as st
import textract
import time
import zipfile
import huggingface_hub
import dotenv
import streamlit.components.v1 as components  # Import Streamlit Components for HTML5

from audio_recorder_streamlit import audio_recorder
from bs4 import BeautifulSoup
from collections import deque
from datetime import datetime
from dotenv import load_dotenv
from huggingface_hub import InferenceClient
from io import BytesIO
from langchain.chat_models import ChatOpenAI
from langchain.chains import ConversationalRetrievalChain
from langchain.embeddings import OpenAIEmbeddings
from langchain.memory import ConversationBufferMemory
from langchain.text_splitter import CharacterTextSplitter
from langchain.vectorstores import FAISS
from openai import ChatCompletion
from PyPDF2 import PdfReader
from templates import bot_template, css, user_template
from xml.etree import ElementTree as ET
from PIL import Image
from urllib.parse import quote  # Ensure this import is included

# Set page configuration with a title and favicon
st.set_page_config(
    page_title="📖🔍GraphicNovelAI",
    page_icon="🔍📖",
    layout="wide",
    initial_sidebar_state="expanded",
    menu_items={
        'Get Help': 'https://huggingface.co/awacke1',
        'Report a bug': "https://huggingface.co/spaces/awacke1/GraphicAINovel",
        'About': "# Midjourney: https://discord.com/channels/@me/997514686608191558"
    }
)

experimentalSubProgram="""

import streamlit as st
import random
import numpy as np

"""


def SpeechSynthesis(result):
    documentHTML5='''
    <!DOCTYPE html>
    <html>
    <head>
        <title>Read It Aloud</title>
        <script type="text/javascript">
            function readAloud() {
                const text = document.getElementById("textArea").value;
                const speech = new SpeechSynthesisUtterance(text);
                window.speechSynthesis.speak(speech);
            }
        </script>
    </head>
    <body>
        <h1>🔊 Read It Aloud</h1>
        <textarea id="textArea" rows="10" cols="80">
    '''
    documentHTML5 = documentHTML5 + result
    documentHTML5 = documentHTML5 + '''
        </textarea>
        <br>
        <button onclick="readAloud()">🔊 Read Aloud</button>
    </body>
    </html>
    '''

    components.html(documentHTML5, width=1280, height=300)
    #return result

PromptPrefix = 'Create a graphic novel story with streamlit markdown outlines and tables with appropriate emojis for graphic novel rules defining the method steps of play.  Use story structure architect rules using plan, structure and top three dramatic situations matching the theme for topic of '
PromptPrefix2 = 'Create a streamlit python user app with full code listing to create a UI implementing the usable choose your own adventure graphic novel rules and story using streamlit, session_state, file_uploader, camera_input, on_change = funcction callbacks, randomness and dice rolls using emojis and st.markdown, st.expander, st.columns and other UI controls in streamlit as a game interface and create inline data tables for entities implemented as variables with python list dictionaries for the game rule entities and stats.  Design it as a fun data driven app and show full python code listing for this ruleset and thematic story plot line: '
PromptPrefix3 = 'Create a HTML5 aframe and javascript app.  Show full code listing.  Create a UI implementing storytelling, features using use three emoji appropriate text detailed plot twists and recurring interesting named characters with genius traits and file IO, randomness, ten point choice lists, math distribution tradeoffs, witty humorous dilemnas with emoji , rewards, variables, reusable functions with parameters, and data driven app with python libraries and streamlit components for Javascript and HTML5. Use appropriate emojis for labels to summarize and list parts, function, conditions for topic:'


st.markdown('''### 📖✨🔍 GraphicNovelAI ''')
with st.expander("Help / About 📚", expanded=False):
    st.markdown('''
    - 🚀 **Unlock Plots:** Elevate your vocabulary with AI. Turns plots into thrilling experiences.
    - 📚 **Features:** Creates extensive glossaries & exciting challenges.
    - 🧙‍♂️ **Experience:** Become a graphic novel plot wizard, boost your language skills.
    - 🔎 **Query Use:** Input `?q=Palindrome` or `?query=Anagram` in URL for new challenges.
    ''')


roleplaying_glossary = {
  "👨‍👩‍👧‍👦 Top Graphic Novel Plot Themes": {
    "Epic Fantasy": [
      "Ancient prophecies and mystical artifacts",
      "Epic battles between good and evil",
      "Complex world-building with diverse cultures",
      "Journey of a reluctant hero",
      "Alliance of unlikely companions",
      "Betrayal and redemption arcs",
      "Magic systems and mythical creatures",
      "Climactic confrontation with a dark lord"
    ],
    "Superhero Sagas": [
      "Origin stories of heroes and villains",
      "Struggle with personal identity and responsibility",
      "Formation of superhero teams",
      "Epic battles to save the city/world",
      "Moral dilemmas and ethical questions",
      "Interdimensional threats and cosmic wars",
      "Evolution of powers and discovery of new abilities",
      "Legacy heroes and passing of the mantle"
    ],
    "Post-Apocalyptic Survival": [
      "Survival in a world after a global catastrophe",
      "Rebuilding society from the ashes",
      "Conflict between surviving factions",
      "Quests for scarce resources",
      "Encounters with mutated creatures",
      "Moral ambiguity and survival ethics",
      "Exploration of human resilience",
      "Discovery of a safe haven or cure"
    ],
    "Science Fiction and Space Opera": [
      "Exploration of distant galaxies",
      "Conflict between alien species",
      "Advanced technology and space travel",
      "Utopian and dystopian societies",
      "Time travel and alternate realities",
      "Artificial intelligence and robotics",
      "Quests for knowledge and discovery",
      "Rebellion against oppressive regimes"
    ],
    "Horror and Supernatural": [
      "Haunted locations and ghost stories",
      "Battles against demonic forces",
      "Survival horror and psychological terror",
      "Folklore and urban legends",
      "Vampires, werewolves, and other monsters",
      "Occult practices and dark magic",
      "Apocalyptic and Lovecraftian themes",
      "Investigations into the unknown"
    ],
    "Romance and Relationship Dramas": [
      "Complex romantic entanglements",
      "Struggles with identity and societal expectations",
      "Heartbreak, healing, and growth",
      "Forbidden love and star-crossed lovers",
      "Contemporary relationship dynamics",
      "Cultural and social differences",
      "Self-discovery and personal fulfillment",
      "Romantic comedies and tragedies"
    ]
  }
}


# 9. Sidebar with UI controls to review and re-run prompts and continue responses
@st.cache_resource
def get_table_download_link(file_path):
    with open(file_path, 'r') as file:
        data = file.read()
   
    b64 = base64.b64encode(data.encode()).decode()  
    file_name = os.path.basename(file_path)
    ext = os.path.splitext(file_name)[1]  # get the file extension
    if ext == '.txt':
        mime_type = 'text/plain'
    elif ext == '.py':
        mime_type = 'text/plain'
    elif ext == '.xlsx':
        mime_type = 'text/plain'
    elif ext == '.csv':
        mime_type = 'text/plain'
    elif ext == '.htm':
        mime_type = 'text/html'
    elif ext == '.md':
        mime_type = 'text/markdown'
    elif ext == '.wav':
        mime_type = 'audio/wav'
    else:
        mime_type = 'application/octet-stream'  # general binary data type
    href = f'<a href="data:{mime_type};base64,{b64}" target="_blank" download="{file_name}">{file_name}</a>'
    return href




def FileSidebar():
    # ----------------------------------------------------- File Sidebar for Jump Gates ------------------------------------------
    # Compose a file sidebar of markdown md files:
    all_files = glob.glob("*.md")
    all_files = [file for file in all_files if len(os.path.splitext(file)[0]) >= 10]  # exclude files with short names
    all_files.sort(key=lambda x: (os.path.splitext(x)[1], x), reverse=True)  # sort by file type and file name in descending order
    if st.sidebar.button("🗑 Delete All Text"):
        for file in all_files:
            os.remove(file)
        st.experimental_rerun()
    if st.sidebar.button("⬇️ Download All"):
        zip_file = create_zip_of_files(all_files)
        st.sidebar.markdown(get_zip_download_link(zip_file), unsafe_allow_html=True)
    file_contents=''
    next_action=''
    for file in all_files:
        col1, col2, col3, col4, col5 = st.sidebar.columns([1,6,1,1,1])  # adjust the ratio as needed
        with col1:
            if st.button("🌐", key="md_"+file):  # md emoji button
                with open(file, 'r') as f:
                    file_contents = f.read()
                    next_action='md'
        with col2:
            st.markdown(get_table_download_link(file), unsafe_allow_html=True)
        with col3:
            if st.button("📂", key="open_"+file):  # open emoji button
                with open(file, 'r') as f:
                    file_contents = f.read()
                    next_action='open'
        with col4:
            if st.button("🔍", key="read_"+file):  # search emoji button
                with open(file, 'r') as f:
                    file_contents = f.read()
                    next_action='search'
        with col5:
            if st.button("🗑", key="delete_"+file):
                os.remove(file)
                st.experimental_rerun()

                
    if len(file_contents) > 0:
        if next_action=='open':
            file_content_area = st.text_area("File Contents:", file_contents, height=500)
        if next_action=='md':
            st.markdown(file_contents)

            buttonlabel = '🔍Run'
            if st.button(key='RunWithLlamaandGPT', label = buttonlabel):
                user_prompt = file_contents

                # gpt
                try:
                    #st.write('🔍Running..')
                    response2 = chat_with_model(user_prompt, file_contents, model_choice)
                    filename2 = generate_filename(file_contents, choice)
                    create_file(filename2, user_prompt, response, should_save)
                    all=response2
                except:
                    st.markdown('GPT is sleeping.  Restart ETA 30 seconds.')
    
                SpeechSynthesis(all)

            
        if next_action=='search':
            file_content_area = st.text_area("File Contents:", file_contents, height=500)
            #st.write('🔍Running..')

            user_prompt = file_contents

            # gpt
            try:
                #st.write('🔍Running with GPT.')
                response2 = chat_with_model(user_prompt, file_contents, model_choice)
                filename2 = generate_filename(file_contents, choice)
                create_file(filename2, user_prompt, response, should_save)
                all=response2
            except:
                st.markdown('GPT is sleeping.  Restart ETA 30 seconds.')

            SpeechSynthesis(all)
    # ----------------------------------------------------- File Sidebar for Jump Gates ------------------------------------------


FileSidebar()




    
# ---- Art Card Sidebar with Random Selection of image:
def get_image_as_base64(url):
    response = requests.get(url)
    if response.status_code == 200:
        # Convert the image to base64
        return base64.b64encode(response.content).decode("utf-8")
    else:
        return None

def create_download_link(filename, base64_str):
    href = f'<a href="data:file/png;base64,{base64_str}" download="{filename}">Download Image</a>'
    return href

# List of image URLs
image_urls = [
    "https://cdn-uploads.huggingface.co/production/uploads/620630b603825909dcbeba35/KGvIFUpU1N-X2tX8hMRva.png",
    "https://cdn-uploads.huggingface.co/production/uploads/620630b603825909dcbeba35/KvmVfcpAbowZIxcViGMmd.png",
    "https://cdn-uploads.huggingface.co/production/uploads/620630b603825909dcbeba35/_ANl0q3ZGDa9CxQqpUmYP.png",
    "https://cdn-uploads.huggingface.co/production/uploads/620630b603825909dcbeba35/SIGnnyVv7eLu8NnrtxdQP.png",
    "https://cdn-uploads.huggingface.co/production/uploads/620630b603825909dcbeba35/5u06M9ue8FXK6Dsi2SF-I.png",
    "https://cdn-uploads.huggingface.co/production/uploads/620630b603825909dcbeba35/gTmhCUdpPaIJcmNXceDi4.png"
]

# Select a random URL from the list
selected_image_url = random.choice(image_urls)

# Get the base64 encoded string of the selected image
selected_image_base64 = get_image_as_base64(selected_image_url)

if selected_image_base64 is not None:
    with st.sidebar:
        st.markdown("""### Graphic Novel AI""")
        # Display the image
        st.markdown(f"![image](data:image/png;base64,{selected_image_base64})")
        
        # Create and display the download link
        download_link = create_download_link("downloaded_image.png", selected_image_base64)
        st.markdown(download_link, unsafe_allow_html=True)
else:
    st.sidebar.write("Failed to load the image.")
# ---- Art Card Sidebar with random selection of image.


    
    

# Ensure the directory for storing scores exists
score_dir = "scores"
os.makedirs(score_dir, exist_ok=True)

# Function to generate a unique key for each button, including an emoji
def generate_key(label, header, idx):
    return f"{header}_{label}_{idx}_key"

# Function to increment and save score
def update_score(key, increment=1):
    score_file = os.path.join(score_dir, f"{key}.json")
    if os.path.exists(score_file):
        with open(score_file, "r") as file:
            score_data = json.load(file)
    else:
        score_data = {"clicks": 0, "score": 0}
    
    score_data["clicks"] += 1
    score_data["score"] += increment
    
    with open(score_file, "w") as file:
        json.dump(score_data, file)
    
    return score_data["score"]

# Function to load score
def load_score(key):
    score_file = os.path.join(score_dir, f"{key}.json")
    if os.path.exists(score_file):
        with open(score_file, "r") as file:
            score_data = json.load(file)
        return score_data["score"]
    return 0


def search_glossary(query):
    for category, terms in roleplaying_glossary.items():
        if query.lower() in (term.lower() for term in terms):
            st.markdown(f"#### {category}")
            st.write(f"- {query}")

    all=""

    query2 = PromptPrefix + query # Add prompt preface for method step task behavior
    # st.write('## ' + query2)
    #st.write('## 🔍 Running with GPT.')  #  -------------------------------------------------------------------------------------------------
    response = chat_with_model(query2)    
    filename = generate_filename(query2 + ' --- ' + response, "md")
    create_file(filename, query, response, should_save)
    
    query3 = PromptPrefix2 + query + ' for story outline of method steps: ' + response # Add prompt preface for coding task behavior
    # st.write('## ' + query3)
    st.write('## 🔍 Coding with GPT.')  #  -------------------------------------------------------------------------------------------------
    response2 = chat_with_model(query3)
    filename_txt = generate_filename(query + ' --- ' + response2, "py")
    create_file(filename_txt, query, response2, should_save)
    
    all = '# Query: ' + query + '# Response: ' + response + '# Response2: ' + response2
    filename_txt2 = generate_filename(query + ' --- ' + all, "md")
    create_file(filename_txt2, query, all, should_save)
    SpeechSynthesis(all)
    return all
    
# Function to display the glossary in a structured format
def display_glossary(glossary, area):
    if area in glossary:
        st.subheader(f"📘 Glossary for {area}")
        for game, terms in glossary[area].items():
            st.markdown(f"### {game}")
            for idx, term in enumerate(terms, start=1):
                st.write(f"{idx}. {term}")


# Function to display the entire glossary in a grid format with links
def display_glossary_grid(roleplaying_glossary):
    search_urls = {
        "📖": lambda k: f"https://en.wikipedia.org/wiki/{quote(k)}",
        "🔍": lambda k: f"https://www.google.com/search?q={quote(k)}",
        "▶️": lambda k: f"https://www.youtube.com/results?search_query={quote(k)}",
        "🔎": lambda k: f"https://www.bing.com/search?q={quote(k)}",
        "🎲": lambda k: f"https://huggingface.co/spaces/awacke1/GraphicAINovel?q={quote(k)}",  # this url plus query!
        "🃏": lambda k: f"https://huggingface.co/spaces/awacke1/GraphicAINovel?q={quote(PromptPrefix)}{quote(k)}",  # this url plus query!
        "📚": lambda k: f"https://huggingface.co/spaces/awacke1/GraphicAINovel?q={quote(PromptPrefix2)}{quote(k)}",  # this url plus query!
        "📚": lambda k: f"https://huggingface.co/spaces/awacke1/GraphicAINovel?q={quote(PromptPrefix3)}{quote(k)}",  # this url plus query!
        
    }

    for category, details in roleplaying_glossary.items():
        st.write(f"### {category}")
        cols = st.columns(len(details))  # Create dynamic columns based on the number of games
        for idx, (game, terms) in enumerate(details.items()):
            with cols[idx]:
                st.markdown(f"#### {game}")
                for term in terms:
                    links_md = ' '.join([f"[{emoji}]({url(term)})" for emoji, url in search_urls.items()])
                    st.markdown(f"{term} {links_md}", unsafe_allow_html=True)




game_emojis = {
    "Dungeons and Dragons": "🐉",
    "Call of Cthulhu": "🐙",
    "GURPS": "🎲",
    "Pathfinder": "🗺️",
    "Kindred of the East": "🌅",
    "Changeling": "🍃",
}

topic_emojis = {
    "Core Rulebooks": "📚",
    "Maps & Settings": "🗺️",
    "Game Mechanics & Tools": "⚙️",
    "Monsters & Adversaries": "👹",
    "Campaigns & Adventures": "📜",
    "Creatives & Assets": "🎨",
    "Game Master Resources": "🛠️",
    "Lore & Background": "📖",
    "Character Development": "🧍",
    "Homebrew Content": "🔧",
    "General Topics": "🌍",
}

# Adjusted display_buttons_with_scores function
def display_buttons_with_scores():
    for category, games in roleplaying_glossary.items():
        category_emoji = topic_emojis.get(category, "🔍")  # Default to search icon if no match
        st.markdown(f"## {category_emoji} {category}")
        for game, terms in games.items():
            game_emoji = game_emojis.get(game, "🎮")  # Default to generic game controller if no match
            for term in terms:
                key = f"{category}_{game}_{term}".replace(' ', '_').lower()
                score = load_score(key)
                if st.button(f"{game_emoji} {term} {score}", key=key):
                    update_score(key)
                    # Create a dynamic query incorporating emojis and formatting for clarity
                    query_prefix = f"{category_emoji} {game_emoji} **{game} - {category}:**"
                    # ----------------------------------------------------------------------------------------------
                    #query_body = f"Create a detailed outline for **{term}** with subpoints highlighting key aspects, using emojis for visual engagement. Include step-by-step rules and boldface important entities and ruleset elements."
                    query_body = f"Create a streamlit python app.py that produces a detailed markdown outline and emoji laden user interface with labels with the entity name and emojis in all labels with a set of streamlit UI components with drop down lists and dataframes and buttons with expander and sidebar for the app to run the data as default values mostly in text boxes. Feature a 3 point outline sith 3 subpoints each where each line has about six words describing this and also contain appropriate emoji for creating sumamry of all aspeccts of this topic. an outline for **{term}** with subpoints highlighting key aspects, using emojis for visual engagement. Include step-by-step rules and boldface important entities and ruleset elements."
                    response = search_glossary(query_prefix + query_body)


def fetch_wikipedia_summary(keyword):
    # Placeholder function for fetching Wikipedia summaries
    # In a real app, you might use requests to fetch from the Wikipedia API
    return f"Summary for {keyword}. For more information, visit Wikipedia."

def create_search_url_youtube(keyword):
    base_url = "https://www.youtube.com/results?search_query="
    return base_url + keyword.replace(' ', '+')

def create_search_url_bing(keyword):
    base_url = "https://www.bing.com/search?q="
    return base_url + keyword.replace(' ', '+')

def create_search_url_wikipedia(keyword):
    base_url = "https://www.wikipedia.org/search-redirect.php?family=wikipedia&language=en&search="
    return base_url + keyword.replace(' ', '+')

def create_search_url_google(keyword):
    base_url = "https://www.google.com/search?q="
    return base_url + keyword.replace(' ', '+')

def create_search_url_ai(keyword):
    base_url = "https://huggingface.co/spaces/awacke1/GraphicAINovel?q="
    return base_url + keyword.replace(' ', '+')

def display_videos_and_links():
    video_files = [f for f in os.listdir('.') if f.endswith('.mp4')]
    if not video_files:
        st.write("No MP4 videos found in the current directory.")
        return

    # Sort video_files based on the length of the keyword to create a visually consistent grid
    video_files_sorted = sorted(video_files, key=lambda x: len(x.split('.')[0]))
    # Calculate the grid size based on the sorted keywords
    grid_sizes = [len(f.split('.')[0]) for f in video_files_sorted]
    # Dynamically adjust column size based on keyword length
    col_sizes = ['small' if size <= 4 else 'medium' if size <= 8 else 'large' for size in grid_sizes]

    # Create a map for number of columns to use for each size
    num_columns_map = {"small": 4, "medium": 3, "large": 2}
    current_grid_size = 0

    for video_file, col_size in zip(video_files_sorted, col_sizes):
        if current_grid_size != num_columns_map[col_size]:
            cols = st.columns(num_columns_map[col_size])
            current_grid_size = num_columns_map[col_size]
            col_index = 0

        with cols[col_index % current_grid_size]:
            st.video(video_file, format='video/mp4', start_time=0)

            # Display Wikipedia and Google search links
            keyword = video_file.split('.')[0]  # Assumes keyword is the file name without extension
            wikipedia_url = create_search_url_wikipedia(keyword)
            google_url = create_search_url_google(keyword)
            youtube_url = create_search_url_youtube(keyword)
            bing_url = create_search_url_bing(keyword)
            ai_url = create_search_url_ai(keyword)
            ai_url2 = create_search_url_ai(keyword + ' ' + PromptPrefix)
            ai_url3 = create_search_url_ai(keyword + ' ' + PromptPrefix2)
            links_md = f"""
            [Wikipedia]({wikipedia_url}) | 
            [Google]({google_url}) | 
            [YouTube]({youtube_url}) | 
            [Bing]({bing_url}) | 
            [AI]({ai_url})|
            [AI Novel]({ai_url2} )|
            [AI Novel App]({ai_url3})
            """
            st.markdown(links_md)
            col_index += 1
 

def display_images_and_wikipedia_summaries():
    image_files = [f for f in os.listdir('.') if f.endswith('.png')]
    if not image_files:
        st.write("No PNG images found in the current directory.")
        return

    # Sort image_files based on the length of the keyword to create a visually consistent grid
    image_files_sorted = sorted(image_files, key=lambda x: len(x.split('.')[0]))
    # Calculate the grid size based on the sorted keywords
    grid_sizes = [len(f.split('.')[0]) for f in image_files_sorted]
    # Dynamically adjust column size based on keyword length
    col_sizes = ['small' if size <= 4 else 'medium' if size <= 8 else 'large' for size in grid_sizes]

    # Create a map for number of columns to use for each size
    num_columns_map = {"small": 4, "medium": 3, "large": 2}
    current_grid_size = 0

    for image_file, col_size in zip(image_files_sorted, col_sizes):
        if current_grid_size != num_columns_map[col_size]:
            cols = st.columns(num_columns_map[col_size])
            current_grid_size = num_columns_map[col_size]
            col_index = 0

        with cols[col_index % current_grid_size]:
            image = Image.open(image_file)
            st.image(image, caption=image_file, use_column_width=True)

            # Display Wikipedia and Google search links
            keyword = image_file.split('.')[0]  # Assumes keyword is the file name without extension
            wikipedia_url = create_search_url_wikipedia(keyword)
            google_url = create_search_url_google(keyword)
            youtube_url = create_search_url_youtube(keyword)
            bing_url = create_search_url_bing(keyword)
            ai_url = create_search_url_ai(keyword)
            ai_url2 = create_search_url_ai(keyword + ' ' + PromptPrefix)
            ai_url3 = create_search_url_ai(keyword + ' ' + PromptPrefix2)
            links_md = f"""
            [Wikipedia]({wikipedia_url}) | 
            [Google]({google_url}) | 
            [YouTube]({youtube_url}) | 
            [Bing]({bing_url}) | 
            [AI]({ai_url})|
            [AI Novel]({ai_url2} )|
            [AI Novel App]({ai_url3})
            """
            st.markdown(links_md)
            col_index += 1


def get_all_query_params(key):
    return st.query_params().get(key, [])

def clear_query_params():
    st.query_params()  
                

# Function to display content or image based on a query
def display_content_or_image(query):
    # Check if the query matches any glossary term
    for category, terms in transhuman_glossary.items():
        for term in terms:
            if query.lower() in term.lower():
                st.subheader(f"Found in {category}:")
                st.write(term)
                return True  # Return after finding and displaying the first match
    
    # Check for an image match in a predefined directory (adjust path as needed)
    image_dir = "images"  # Example directory where images are stored
    image_path = f"{image_dir}/{query}.png"  # Construct image path with query
    if os.path.exists(image_path):
        st.image(image_path, caption=f"Image for {query}")
        return True
    
    # If no content or image is found
    st.warning("No matching content or image found.")
    return False





# ------------------------------------

def add_Med_Licensing_Exam_Dataset():
    import streamlit as st
    from datasets import load_dataset
    dataset = load_dataset("augtoma/usmle_step_1")['test']  # Using 'test' split
    st.title("USMLE Step 1 Dataset Viewer")
    if len(dataset) == 0:
        st.write("😢 The dataset is empty.")
    else:
        st.write("""
        🔍 Use the search box to filter questions or use the grid to scroll through the dataset.
        """)
    
        # 👩‍🔬 Search Box
        search_term = st.text_input("Search for a specific question:", "")
        
        # 🎛 Pagination
        records_per_page = 100
        num_records = len(dataset)
        num_pages = max(int(num_records / records_per_page), 1)
        
        # Skip generating the slider if num_pages is 1 (i.e., all records fit in one page)
        if num_pages > 1:
            page_number = st.select_slider("Select page:", options=list(range(1, num_pages + 1)))
        else:
            page_number = 1  # Only one page
        
        # 📊 Display Data
        start_idx = (page_number - 1) * records_per_page
        end_idx = start_idx + records_per_page
    
        # 🧪 Apply the Search Filter
        filtered_data = []
        for record in dataset[start_idx:end_idx]:
            if isinstance(record, dict) and 'text' in record and 'id' in record:
                if search_term:
                    if search_term.lower() in record['text'].lower():
                        st.markdown(record)
                        filtered_data.append(record)
                else:
                    filtered_data.append(record)
    
        # 🌐 Render the Grid
        for record in filtered_data:
            st.write(f"## Question ID: {record['id']}")
            st.write(f"### Question:")
            st.write(f"{record['text']}")
            st.write(f"### Answer:")
            st.write(f"{record['answer']}")
            st.write("---")
    
        st.write(f"😊 Total Records: {num_records} | 📄 Displaying {start_idx+1} to {min(end_idx, num_records)}")

# 1. Constants and Top Level UI Variables

# My Inference API Copy
API_URL = 'https://qe55p8afio98s0u3.us-east-1.aws.endpoints.huggingface.cloud'  # Dr Llama
# Meta's Original - Chat HF Free Version:
#API_URL = "https://api-inference.huggingface.co/models/meta-llama/Llama-2-7b-chat-hf"
API_KEY = os.getenv('API_KEY')
MODEL1="meta-llama/Llama-2-7b-chat-hf"
MODEL1URL="https://huggingface.co/meta-llama/Llama-2-7b-chat-hf"
HF_KEY = os.getenv('HF_KEY')
headers = {
    "Authorization": f"Bearer {HF_KEY}",
    "Content-Type": "application/json"
}
key = os.getenv('OPENAI_API_KEY')
prompt = f"Write instructions to teach discharge planning along with guidelines and patient education. List entities, features and relationships to CCDA and FHIR objects in boldface."
should_save = st.sidebar.checkbox("💾 Save", value=True, help="Save your session data.")

# 2. Prompt label button demo for LLM
def add_witty_humor_buttons():
    with st.expander("Wit and Humor 🤣", expanded=True):
        # Tip about the Dromedary family
        st.markdown("🔬 **Fun Fact**: Dromedaries, part of the camel family, have a single hump and are adapted to arid environments. Their 'superpowers' include the ability to survive without water for up to 7 days, thanks to their specialized blood cells and water storage in their hump.")
        
        # Define button descriptions
        descriptions = {
            "Generate Limericks 😂": "Write ten random adult limericks based on quotes that are tweet length and make you laugh 🎭",
            "Wise Quotes 🧙": "Generate ten wise quotes that are tweet length 🦉",
            "Funny Rhymes 🎤": "Create ten funny rhymes that are tweet length 🎶",
            "Medical Jokes 💉": "Create ten medical jokes that are tweet length 🏥",
            "Minnesota Humor ❄️": "Create ten jokes about Minnesota that are tweet length 🌨️",
            "Top Funny Stories 📖": "Create ten funny stories that are tweet length 📚",
            "More Funny Rhymes 🎙️": "Create ten more funny rhymes that are tweet length 🎵"
        }
        
        # Create columns
        col1, col2, col3 = st.columns([1, 1, 1], gap="small")
        
        # Add buttons to columns
        if col1.button("Wise Limericks 😂"):
            StreamLLMChatResponse(descriptions["Generate Limericks 😂"])
        
        if col2.button("Wise Quotes 🧙"):
            StreamLLMChatResponse(descriptions["Wise Quotes 🧙"])
        
        #if col3.button("Funny Rhymes 🎤"):
        #    StreamLLMChatResponse(descriptions["Funny Rhymes 🎤"])
        
        col4, col5, col6 = st.columns([1, 1, 1], gap="small")
        
        if col4.button("Top Ten Funniest Clean Jokes 💉"):
            StreamLLMChatResponse(descriptions["Top Ten Funniest Clean Jokes  💉"])
        
        if col5.button("Minnesota Humor ❄️"):
            StreamLLMChatResponse(descriptions["Minnesota Humor ❄️"])
        
        if col6.button("Origins of Medical Science True Stories"):
            StreamLLMChatResponse(descriptions["Origins of Medical Science True Stories"])
        
        col7 = st.columns(1, gap="small")
        
        if col7[0].button("Top Ten Best Write a streamlit python program prompts to build AI programs. 🎙️"):
            StreamLLMChatResponse(descriptions["Top Ten Best Write a streamlit python program prompts to build AI programs. 🎙️"])



# 3. Stream Llama Response
# @st.cache_resource
def StreamLLMChatResponse(prompt):
    try:
        endpoint_url = API_URL
        hf_token = API_KEY
        #st.write('Running client ' + endpoint_url)
        client = InferenceClient(endpoint_url, token=hf_token)
        gen_kwargs = dict(
            max_new_tokens=512,
            top_k=30,
            top_p=0.9,
            temperature=0.2,
            repetition_penalty=1.02,
            stop_sequences=["\nUser:", "<|endoftext|>", "</s>"],
        )
        stream = client.text_generation(prompt, stream=True, details=True, **gen_kwargs)
        report=[]
        res_box = st.empty()
        collected_chunks=[]
        collected_messages=[]
        allresults=''
        for r in stream:
            if r.token.special:
                continue
            if r.token.text in gen_kwargs["stop_sequences"]:
                break
            collected_chunks.append(r.token.text)
            chunk_message = r.token.text
            collected_messages.append(chunk_message)
            try:
                report.append(r.token.text)
                if len(r.token.text) > 0:
                    result="".join(report).strip()
                    res_box.markdown(f'*{result}*')
                    
            except:
                st.write('Stream llm issue')
        SpeechSynthesis(result)
        return result
    except:
        st.write('Llama model is asleep. Starting up now on A10 - please give 5 minutes then retry as KEDA scales up from zero to activate running container(s).')

# 4. Run query with payload
def query(payload):
    response = requests.post(API_URL, headers=headers, json=payload)
    st.markdown(response.json())
    return response.json()
def get_output(prompt):
    return query({"inputs": prompt})

# 5. Auto name generated output files from time and content
def generate_filename(prompt, file_type):
    central = pytz.timezone('US/Central')
    safe_date_time = datetime.now(central).strftime("%m%d_%H%M")
    replaced_prompt = prompt.replace(" ", "_").replace("\n", "_")
    safe_prompt = "".join(x for x in replaced_prompt if x.isalnum() or x == "_")[:255]  # 255 is linux max, 260 is windows max
    #safe_prompt = "".join(x for x in replaced_prompt if x.isalnum() or x == "_")[:45]
    return f"{safe_date_time}_{safe_prompt}.{file_type}"

# 6. Speech transcription via OpenAI service
def transcribe_audio(openai_key, file_path, model):
    openai.api_key = openai_key
    OPENAI_API_URL = "https://api.openai.com/v1/audio/transcriptions"
    headers = {
        "Authorization": f"Bearer {openai_key}",
    }
    with open(file_path, 'rb') as f:
        data = {'file': f}
        st.write('STT transcript ' + OPENAI_API_URL)
        response = requests.post(OPENAI_API_URL, headers=headers, files=data, data={'model': model})
    if response.status_code == 200:
        st.write(response.json())
        chatResponse = chat_with_model(response.json().get('text'), '') # *************************************
        transcript = response.json().get('text')
        filename = generate_filename(transcript, 'txt')
        response = chatResponse
        user_prompt = transcript
        create_file(filename, user_prompt, response, should_save)
        return transcript
    else:
        st.write(response.json())
        st.error("Error in API call.")
        return None

# 7. Auto stop on silence audio control for recording WAV files
def save_and_play_audio(audio_recorder):
    audio_bytes = audio_recorder(key='audio_recorder')
    if audio_bytes:
        filename = generate_filename("Recording", "wav")
        with open(filename, 'wb') as f:
            f.write(audio_bytes)
        st.audio(audio_bytes, format="audio/wav")
        return filename
    return None

# 8. File creator that interprets type and creates output file for text, markdown and code
def create_file(filename, prompt, response, should_save=True):
    if not should_save:
        return
    base_filename, ext = os.path.splitext(filename)
    if ext in ['.txt', '.htm', '.md']:
        with open(f"{base_filename}.md", 'w') as file:
            try:
                content = prompt.strip() + '\r\n' + response
                file.write(content)
            except:
                st.write('.')

    #has_python_code = re.search(r"```python([\s\S]*?)```", prompt.strip() + '\r\n' + response)
    #has_python_code = bool(re.search(r"```python([\s\S]*?)```", prompt.strip() + '\r\n' + response))
        #if has_python_code:
        #    python_code = re.findall(r"```python([\s\S]*?)```", response)[0].strip()
        #    with open(f"{base_filename}-Code.py", 'w') as file:
        #        file.write(python_code)
        #    with open(f"{base_filename}.md", 'w') as file:
        #        content = prompt.strip() + '\r\n' + response
        #        file.write(content)
            
def truncate_document(document, length):
    return document[:length]
def divide_document(document, max_length):
    return [document[i:i+max_length] for i in range(0, len(document), max_length)]

def CompressXML(xml_text):
    root = ET.fromstring(xml_text)
    for elem in list(root.iter()):
        if isinstance(elem.tag, str) and 'Comment' in elem.tag:
            elem.parent.remove(elem)
    return ET.tostring(root, encoding='unicode', method="xml")

# 10. Read in and provide UI for past files
@st.cache_resource
def read_file_content(file,max_length):
    if file.type == "application/json":
        content = json.load(file)
        return str(content)
    elif file.type == "text/html" or file.type == "text/htm":
        content = BeautifulSoup(file, "html.parser")
        return content.text
    elif file.type == "application/xml" or file.type == "text/xml":
        tree = ET.parse(file)
        root = tree.getroot()
        xml = CompressXML(ET.tostring(root, encoding='unicode'))
        return xml
    elif file.type == "text/markdown" or file.type == "text/md":
        md = mistune.create_markdown()
        content = md(file.read().decode())
        return content
    elif file.type == "text/plain":
        return file.getvalue().decode()
    else:
        return ""

# 11. Chat with GPT - Caution on quota - now favoring fastest AI pipeline STT Whisper->LLM Llama->TTS
@st.cache_resource
def chat_with_model(prompt, document_section='', model_choice='gpt-3.5-turbo'):
    model = model_choice
    conversation = [{'role': 'system', 'content': 'You are a helpful assistant.'}]
    conversation.append({'role': 'user', 'content': prompt})
    if len(document_section)>0:
        conversation.append({'role': 'assistant', 'content': document_section})
    start_time = time.time()
    report = []
    res_box = st.empty()
    collected_chunks = []
    collected_messages = []

    st.write('LLM stream ' + 'gpt-3.5-turbo')
    for chunk in openai.ChatCompletion.create(model='gpt-3.5-turbo', messages=conversation, temperature=0.5, stream=True):
        collected_chunks.append(chunk)  
        chunk_message = chunk['choices'][0]['delta']  
        collected_messages.append(chunk_message) 
        content=chunk["choices"][0].get("delta",{}).get("content")
        try:
            report.append(content)
            if len(content) > 0:
                result = "".join(report).strip()
                res_box.markdown(f'*{result}*') 
        except:
            st.write(' ')
    full_reply_content = ''.join([m.get('content', '') for m in collected_messages])
    st.write("Elapsed time:")
    st.write(time.time() - start_time)
    return full_reply_content

# 12. Embedding VectorDB for LLM query of documents to text to compress inputs and prompt together as Chat memory using Langchain
@st.cache_resource
def chat_with_file_contents(prompt, file_content, model_choice='gpt-3.5-turbo'):
    conversation = [{'role': 'system', 'content': 'You are a helpful assistant.'}]
    conversation.append({'role': 'user', 'content': prompt})
    if len(file_content)>0:
        conversation.append({'role': 'assistant', 'content': file_content})
    response = openai.ChatCompletion.create(model=model_choice, messages=conversation)
    return response['choices'][0]['message']['content']

def extract_mime_type(file):
    if isinstance(file, str):
        pattern = r"type='(.*?)'"
        match = re.search(pattern, file)
        if match:
            return match.group(1)
        else:
            raise ValueError(f"Unable to extract MIME type from {file}")
    elif isinstance(file, streamlit.UploadedFile):
        return file.type
    else:
        raise TypeError("Input should be a string or a streamlit.UploadedFile object")

def extract_file_extension(file):
    # get the file name directly from the UploadedFile object
    file_name = file.name
    pattern = r".*?\.(.*?)$"
    match = re.search(pattern, file_name)
    if match:
        return match.group(1)
    else:
        raise ValueError(f"Unable to extract file extension from {file_name}")

# Normalize input as text from PDF and other formats
@st.cache_resource
def pdf2txt(docs):
    text = ""
    for file in docs:
        file_extension = extract_file_extension(file)
        st.write(f"File type extension: {file_extension}")
        if file_extension.lower() in ['py', 'txt', 'html', 'htm', 'xml', 'json']:
            text += file.getvalue().decode('utf-8')
        elif file_extension.lower() == 'pdf':
            from PyPDF2 import PdfReader
            pdf = PdfReader(BytesIO(file.getvalue()))
            for page in range(len(pdf.pages)):
                text += pdf.pages[page].extract_text() # new PyPDF2 syntax
    return text

def txt2chunks(text):
    text_splitter = CharacterTextSplitter(separator="\n", chunk_size=1000, chunk_overlap=200, length_function=len)
    return text_splitter.split_text(text)

# Vector Store using FAISS
@st.cache_resource
def vector_store(text_chunks):
    embeddings = OpenAIEmbeddings(openai_api_key=key)
    return FAISS.from_texts(texts=text_chunks, embedding=embeddings)

# Memory and Retrieval chains
@st.cache_resource
def get_chain(vectorstore):
    llm = ChatOpenAI()
    memory = ConversationBufferMemory(memory_key='chat_history', return_messages=True)
    return ConversationalRetrievalChain.from_llm(llm=llm, retriever=vectorstore.as_retriever(), memory=memory)

def process_user_input(user_question):
    response = st.session_state.conversation({'question': user_question})
    st.session_state.chat_history = response['chat_history']
    for i, message in enumerate(st.session_state.chat_history):
        template = user_template if i % 2 == 0 else bot_template
        st.write(template.replace("{{MSG}}", message.content), unsafe_allow_html=True)
        filename = generate_filename(user_question, 'txt')
        response = message.content
        user_prompt = user_question
        create_file(filename, user_prompt, response, should_save)       

def divide_prompt(prompt, max_length):
    words = prompt.split()
    chunks = []
    current_chunk = []
    current_length = 0
    for word in words:
        if len(word) + current_length <= max_length:
            current_length += len(word) + 1 
            current_chunk.append(word)
        else:
            chunks.append(' '.join(current_chunk))
            current_chunk = [word]
            current_length = len(word)
    chunks.append(' '.join(current_chunk))
    return chunks

    
# 13. Provide way of saving all and deleting all to give way of reviewing output and saving locally before clearing it
    
@st.cache_resource
def create_zip_of_files(files):
    zip_name = "all_files.zip"
    with zipfile.ZipFile(zip_name, 'w') as zipf:
        for file in files:
            zipf.write(file)
    return zip_name
    
@st.cache_resource
def get_zip_download_link(zip_file):
    with open(zip_file, 'rb') as f:
        data = f.read()
    b64 = base64.b64encode(data).decode()
    href = f'<a href="data:application/zip;base64,{b64}" download="{zip_file}">Download All</a>'
    return href

# 14. Inference Endpoints for Whisper (best fastest STT) on NVIDIA T4 and Llama (best fastest AGI LLM) on NVIDIA A10
# My Inference Endpoint
API_URL_IE = f'https://tonpixzfvq3791u9.us-east-1.aws.endpoints.huggingface.cloud'
# Original
API_URL_IE = "https://api-inference.huggingface.co/models/openai/whisper-small.en"
MODEL2 = "openai/whisper-small.en"
MODEL2_URL = "https://huggingface.co/openai/whisper-small.en"
#headers = {
#	"Authorization": "Bearer XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX",
#	"Content-Type": "audio/wav"
#}
# HF_KEY = os.getenv('HF_KEY')
HF_KEY = st.secrets['HF_KEY']
headers = {
    "Authorization": f"Bearer {HF_KEY}",
    "Content-Type": "audio/wav"
}

#@st.cache_resource
def query(filename):
    with open(filename, "rb") as f:
        data = f.read()
    response = requests.post(API_URL_IE, headers=headers, data=data)
    return response.json()

def generate_filename(prompt, file_type):
    central = pytz.timezone('US/Central')
    safe_date_time = datetime.now(central).strftime("%m%d_%H%M")
    replaced_prompt = prompt.replace(" ", "_").replace("\n", "_")
    safe_prompt = "".join(x for x in replaced_prompt if x.isalnum() or x == "_")[:90]
    return f"{safe_date_time}_{safe_prompt}.{file_type}"

# 15. Audio recorder to Wav file 
def save_and_play_audio(audio_recorder):
    audio_bytes = audio_recorder()
    if audio_bytes:
        filename = generate_filename("Recording", "wav")
        with open(filename, 'wb') as f:
            f.write(audio_bytes)
        st.audio(audio_bytes, format="audio/wav")
        return filename

# 16. Speech transcription to file output
def transcribe_audio(filename):
    output = query(filename)
    return output

def whisper_main():
    #st.title("Speech to Text")
    #st.write("Record your speech and get the text.")

    # Audio, transcribe, GPT:
    filename = save_and_play_audio(audio_recorder)
    if filename is not None:
        transcription = transcribe_audio(filename)
        try:
            transcript = transcription['text']
            st.write(transcript)

        except:
            transcript=''
            st.write(transcript)

        
        # Whisper to GPT: New!! ---------------------------------------------------------------------
        st.write('Reasoning with your inputs with GPT..')
        response = chat_with_model(transcript)
        st.write('Response:')
        st.write(response)

        filename = generate_filename(response, "txt")
        create_file(filename, transcript, response, should_save)
        # Whisper to GPT: New!! ---------------------------------------------------------------------
        
        
        # Whisper to Llama:
        response = StreamLLMChatResponse(transcript)
        filename_txt = generate_filename(transcript, "md")
        create_file(filename_txt, transcript, response, should_save)

        filename_wav = filename_txt.replace('.txt', '.wav')
        import shutil
        try: 
            if os.path.exists(filename):
                shutil.copyfile(filename, filename_wav)
        except:
            st.write('.')

        if os.path.exists(filename):
            os.remove(filename)

        #st.experimental_rerun()
        #except:
        #    st.write('Starting Whisper Model on GPU.  Please retry in 30 seconds.')



# Sample function to demonstrate a response, replace with your own logic
def StreamMedChatResponse(topic):
    st.write(f"Showing resources or questions related to: {topic}")


                

# 17. Main
def main():
    prompt = f"Write ten funny jokes that are tweet length stories that make you laugh.  Show as markdown outline with emojis for each."
    # Add Wit and Humor buttons
    # add_witty_humor_buttons()
    # add_medical_exam_buttons()

    with st.expander("Prompts 📚", expanded=False):
        example_input = st.text_input("Enter your prompt text for Llama:", value=prompt, help="Enter text to get a response from DromeLlama.")
        if st.button("Run Prompt With Llama model", help="Click to run the prompt."):
            try:
                response=StreamLLMChatResponse(example_input)
                create_file(filename, example_input, response, should_save)
            except:
                st.write('Llama model is asleep. Starting now on A10 GPU.  Please wait one minute then retry.  KEDA triggered.')

        openai.api_key = os.getenv('OPENAI_API_KEY')
        if openai.api_key == None: openai.api_key = st.secrets['OPENAI_API_KEY']
        
        menu = ["txt", "htm", "xlsx", "csv", "md", "py"]
        choice = st.sidebar.selectbox("Output File Type:", menu)
        
        model_choice = st.sidebar.radio("Select Model:", ('gpt-3.5-turbo', 'gpt-3.5-turbo-0301'))        
        
        user_prompt = st.text_area("Enter prompts, instructions & questions:", '', height=100)
        collength, colupload = st.columns([2,3])  # adjust the ratio as needed
        with collength:
            max_length = st.slider("File section length for large files", min_value=1000, max_value=128000, value=12000, step=1000)
        with colupload:
            uploaded_file = st.file_uploader("Add a file for context:", type=["pdf", "xml", "json", "xlsx", "csv", "html", "htm", "md", "txt"])
        document_sections = deque()
        document_responses = {}
        if uploaded_file is not None:
            file_content = read_file_content(uploaded_file, max_length)
            document_sections.extend(divide_document(file_content, max_length))
        if len(document_sections) > 0:
            if st.button("👁️ View Upload"):
                st.markdown("**Sections of the uploaded file:**")
                for i, section in enumerate(list(document_sections)):
                    st.markdown(f"**Section {i+1}**\n{section}")
            st.markdown("**Chat with the model:**")
            for i, section in enumerate(list(document_sections)):
                if i in document_responses:
                    st.markdown(f"**Section {i+1}**\n{document_responses[i]}")
                else:
                    if st.button(f"Chat about Section {i+1}"):
                        st.write('Reasoning with your inputs...')
                        #response = chat_with_model(user_prompt, section, model_choice)
                        st.write('Response:')
                        st.write(response)
                        document_responses[i] = response
                        filename = generate_filename(f"{user_prompt}_section_{i+1}", choice)
                        create_file(filename, user_prompt, response, should_save)
                        st.sidebar.markdown(get_table_download_link(filename), unsafe_allow_html=True)

                        
        if st.button('💬 Chat'):
            st.write('Reasoning with your inputs...')
            user_prompt_sections = divide_prompt(user_prompt, max_length)
            full_response = ''
            for prompt_section in user_prompt_sections:
                response = chat_with_model(prompt_section, ''.join(list(document_sections)), model_choice)
                full_response += response + '\n'  # Combine the responses
            response = full_response
            st.write('Response:')
            st.write(response)
            filename = generate_filename(user_prompt, choice)
            create_file(filename, user_prompt, response, should_save)
            

    # Function to encode file to base64
    def get_base64_encoded_file(file_path):
        with open(file_path, "rb") as file:
            return base64.b64encode(file.read()).decode()

    # Function to create a download link
    def get_audio_download_link(file_path):
        base64_file = get_base64_encoded_file(file_path)
        return f'<a href="data:file/wav;base64,{base64_file}" download="{os.path.basename(file_path)}">⬇️ Download Audio</a>'

    # Compose a file sidebar of past encounters
    all_files = glob.glob("*.wav")
    all_files = [file for file in all_files if len(os.path.splitext(file)[0]) >= 10]  # exclude files with short names
    all_files.sort(key=lambda x: (os.path.splitext(x)[1], x), reverse=True)  # sort by file type and file name in descending order

    filekey = 'delall'
    if st.sidebar.button("🗑 Delete All Audio", key=filekey):
        for file in all_files:
            os.remove(file)
        st.experimental_rerun()

    for file in all_files:
        col1, col2 = st.sidebar.columns([6, 1])  # adjust the ratio as needed
        with col1:
            st.markdown(file)
            if st.button("🎵", key="play_" + file):  # play emoji button
                audio_file = open(file, 'rb')
                audio_bytes = audio_file.read()
                st.audio(audio_bytes, format='audio/wav')
                #st.markdown(get_audio_download_link(file), unsafe_allow_html=True)
                #st.text_input(label="", value=file)
        with col2:
            if st.button("🗑", key="delete_" + file):
                os.remove(file)
                st.experimental_rerun()



    # Feedback
    # Step: Give User a Way to Upvote or Downvote
    GiveFeedback=False
    if GiveFeedback:
        with st.expander("Give your feedback 👍", expanded=False):
    
            feedback = st.radio("Step 8: Give your feedback", ("👍 Upvote", "👎 Downvote"))
            if feedback == "👍 Upvote":
                st.write("You upvoted 👍. Thank you for your feedback!")
            else:
                st.write("You downvoted 👎. Thank you for your feedback!")
                
            load_dotenv()
            st.write(css, unsafe_allow_html=True)
            st.header("Chat with documents :books:")
            user_question = st.text_input("Ask a question about your documents:")
            if user_question:
                process_user_input(user_question)
            with st.sidebar:
                st.subheader("Your documents")
                docs = st.file_uploader("import documents", accept_multiple_files=True)
                with st.spinner("Processing"):
                    raw = pdf2txt(docs)
                    if len(raw) > 0:
                        length = str(len(raw))
                        text_chunks = txt2chunks(raw)
                        vectorstore = vector_store(text_chunks)
                        st.session_state.conversation = get_chain(vectorstore)
                        st.markdown('# AI Search Index of Length:' + length + ' Created.')  # add timing
                        filename = generate_filename(raw, 'txt')
                        create_file(filename, raw, '', should_save)
    
    # Relocated!  Hope you like your new space - enjoy!
    # Display instructions and handle query parameters
    #st.markdown("## Glossary Lookup\nEnter a term in the URL query, like `?q=Nanotechnology` or `?query=Martian Syndicate`.")
    
    try:
        query_params = st.query_params
        #query = (query_params.get('q') or query_params.get('query') or [''])[0]
        query = (query_params.get('q') or query_params.get('query') or [''])
        #st.markdown('# Running query: ' + query)
        if query: search_glossary(query)
    except:
        st.markdown(' ')

    # Display the glossary grid
    st.markdown("### 🎲🗺️ Graphic Novel Gallery")
    display_videos_and_links()
    display_images_and_wikipedia_summaries()
    display_glossary_grid(roleplaying_glossary)
    st.markdown("## Explore the vast universe of Graphic Novels including card and board games with fascinating readable content and semantic rules where language is fun!.🌠")
    display_buttons_with_scores()
    
    # Assuming the transhuman_glossary and other setup code remains the same
    #st.write("Current Query Parameters:", st.query_params)
    #st.markdown("### Query Parameters - These Deep Link Map to Remixable Methods, Navigate or Trigger Functionalities")
    
    # Example: Using query parameters to navigate or trigger functionalities
    if 'action' in st.query_params:
        action = st.query_params()['action'][0]  # Get the first (or only) 'action' parameter
        if action == 'show_message':
            st.success("Showing a message because 'action=show_message' was found in the URL.")
        elif action == 'clear':
            clear_query_params()
            st.experimental_rerun()
    
    # Handling repeated keys
    if 'multi' in st.query_params:
        multi_values = get_all_query_params('multi')
        st.write("Values for 'multi':", multi_values)
    
    # Manual entry for demonstration
    st.write("Enter query parameters in the URL like this: ?action=show_message&multi=1&multi=2")
    
    if 'query' in st.query_params:
        query = st.query_params['query'][0]  # Get the query parameter
        # Display content or image based on the query
        display_content_or_image(query)
    
    # Add a clear query parameters button for convenience
    if st.button("Clear Query Parameters", key='ClearQueryParams'):
        # This will clear the browser URL's query parameters
        st.experimental_set_query_params
        st.experimental_rerun()
                    
# 18. Run AI Pipeline
if __name__ == "__main__":
    whisper_main()
    main()