app.py

import streamlit as st
import os
import glob
import re
from urllib.parse import quote
from gradio_client import Client
import json
from datetime import datetime

# Define the markdown variables
Boxing_and_MMA_Commentary_and_Knowledge = """
# Boxing and UFC Study of 1971 - 2024 The Greatest Fights History

1. In Boxing, the most heart breaking fight in Boxing was the Boom Boom Mancini fight with Duku Kim.
2. After changes to Boxing made it more safe due to the heart break.
3. Rehydration of the brain after weight ins loss preparation for a match is life saving change.
4. Fighting went from 15 rounds to 12.

# UFC By Contrast..
1. 5 Rounds of 5 Minutes each.
2. Greatest UFC Fighters:
    - Jon Jones could be the greatest of all time (GOAT) since he never lost.
    - George St. Pierre
    - BJ Penn
    - Anderson Silva
    - Mighty Mouse MMA's heart at 125 pounds
    - Kabib retired 29 and 0
    - Fedor Milliano
    - Alex Pereira
    - James Tony
    - Randy Couture
3. You have to Judge them in their Championship Peak
4. Chris Weidman
5. Connor McGregor
6. Leg Breaking - Shin calcification and breaking baseball bats

# References:
1. Joe Rogan - Interview #2219
2. Donald J Trump
"""

Multiplayer_Custom_Hosting_Game_Servers_For_Simulated_Worlds = """
# Multiplayer Simulated Worlds

1. 7 Days To Die PC 
2. ARK: Survival Evolved PC 
3. Arma 3 PC 
4. Atlas PC 
5. Conan Exiles PC 
6. Craftopia PC 
7. DayZ PC 
8. Eco - Global Survival PC 
9. Empyrion - Galactic Survival PC 
10. Factorio PC 
11. Farming Simulator 19 PC 
12. Crossplay
13. Farming Simulator 22 
14. Last Oasis PC 
15. Last Oasis Classic PC 
16. Minecraft (Vanilla) PC 
17. Crossplay
18. Path of Titans
19. Rust PC 
20. SCP: Secret Laboratory PC 
21. SCUM PC
22. Satisfactory PC
23. Satisfactory (Experimental) PC 
24. Crossplay
25. Space Engineers 
26. Terraria (tShock & Vanilla) PC 
27. The Forest PC 
28. Crossplay
29. Valheim
"""

# Function to parse markdown text and extract terms
def extract_terms(markdown_text):
    lines = markdown_text.strip().split('\n')
    terms = []
    for line in lines:
        line = re.sub(r'^[#*\->\d\.\s]+', '', line).strip()
        if line:
            terms.append(line)
    return terms

# Function to display terms with links
def display_terms_with_links(terms):
    search_urls = {
        "🚀🌌ArXiv": lambda k: f"/?q={quote(k)}",
        "📖": 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://twitter.com/search?q={quote(k)}",
    }
    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)

# Function to perform AI lookup using Gradio client
def perform_ai_lookup(query):
    st.write("Performing AI Lookup...")
    # Initialize the Gradio client
    client = Client("awacke1/Arxiv-Paper-Search-And-QA-RAG-Pattern")
    # Perform the AI lookup using the Mixtral and Mistral models
    result1 = client.predict(
        prompt=query,
        llm_model_picked="mistralai/Mixtral-8x7B-Instruct-v0.1",
        stream_outputs=True,
        api_name="/ask_llm"
    )
    st.markdown("### Mixtral-8x7B-Instruct-v0.1 Result")
    st.markdown(result1)
    result2 = client.predict(
        prompt=query,
        llm_model_picked="mistralai/Mistral-7B-Instruct-v0.2",
        stream_outputs=True,
        api_name="/ask_llm"
    )
    st.markdown("### Mistral-7B-Instruct-v0.2 Result")
    st.markdown(result2)
    combined_result = f"{result1}\n\n{result2}"
    return combined_result

# Function to extract URLs from AI result
def extract_urls(text):
    try:
        date_pattern = re.compile(r'### (\d{2} \w{3} \d{4})')
        abs_link_pattern = re.compile(r'\[(.*?)\]\((https://arxiv\.org/abs/\d+\.\d+)\)')
        pdf_link_pattern = re.compile(r'\[⬇️\]\((https://arxiv\.org/pdf/\d+\.\d+)\)')
        title_pattern = re.compile(r'### \d{2} \w{3} \d{4} \| \[(.*?)\]')
        date_matches = date_pattern.findall(text)
        abs_link_matches = abs_link_pattern.findall(text)
        pdf_link_matches = pdf_link_pattern.findall(text)
        title_matches = title_pattern.findall(text)

        markdown_text = ""
        for i in range(len(date_matches)):
            date = date_matches[i]
            title = title_matches[i]
            abs_link = abs_link_matches[i][1]
            pdf_link = pdf_link_matches[i]
            markdown_text += f"**Date:** {date}\n\n"
            markdown_text += f"**Title:** {title}\n\n"
            markdown_text += f"**Abstract Link:** [{abs_link}]({abs_link})\n\n"
            markdown_text += f"**PDF Link:** [{pdf_link}]({pdf_link})\n\n"
            markdown_text += "---\n\n"
        return markdown_text

    except Exception as e:
        st.write(f"An error occurred in extract_urls: {e}")
        return ''

# Function to generate filename based on date and content
def generate_filename(prefix, content):
    timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
    safe_content = re.sub(r'\W+', '_', content[:50])
    filename = f"{prefix}_{timestamp}_{safe_content}.md"
    return filename

# Sidebar for file management
def file_management_sidebar():
    st.sidebar.title("📁 File Management")

    # Get list of .md files excluding README.md
    md_files = [file for file in glob.glob("*.md") if os.path.basename(file).lower() != 'readme.md']
    md_files.sort()

    if md_files:
        selected_file = st.sidebar.selectbox("Select a markdown file to view/edit", md_files)

        # Navigation buttons
        file_index = md_files.index(selected_file)
        col1, col2 = st.sidebar.columns([1,1])
        if col1.button("Previous"):
            if file_index > 0:
                selected_file = md_files[file_index -1]
                st.experimental_set_query_params(selected_file=selected_file)
                st.experimental_rerun()
        if col2.button("Next"):
            if file_index < len(md_files) -1:
                selected_file = md_files[file_index +1]
                st.experimental_set_query_params(selected_file=selected_file)
                st.experimental_rerun()

        # Load file content
        with open(selected_file, 'r', encoding='utf-8') as f:
            file_content = f.read()

        # Tabs for Markdown View and Code Editor
        tab1, tab2 = st.tabs(["Markdown View", "Code Editor"])

        with tab1:
            st.markdown(f"### {selected_file}")
            st.markdown(file_content)

        with tab2:
            edited_content = st.text_area("Edit the markdown content", file_content, height=400)
            if st.button("Save Changes"):
                with open(selected_file, 'w', encoding='utf-8') as f:
                    f.write(edited_content)
                st.success(f"Changes saved to {selected_file}")
                st.experimental_rerun()
    else:
        st.sidebar.write("No markdown files found.")

    # Option to create a new markdown file
    if st.sidebar.button("Create New Markdown File"):
        # Generate automatic filename
        timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
        new_filename = f"note_{timestamp}.md"
        with open(new_filename, 'w', encoding='utf-8') as f:
            f.write("# New Markdown File\n")
        st.sidebar.success(f"Created new file: {new_filename}")
        st.experimental_set_query_params(selected_file=new_filename)
        st.experimental_rerun()

# Main application logic
def main():
    st.title("Markdown Content with AI Lookup and File Management")

    # Display the original markdown content
    st.markdown("## Original Markdown Content")
    st.markdown(Boxing_and_MMA_Commentary_and_Knowledge)
    st.markdown(Multiplayer_Custom_Hosting_Game_Servers_For_Simulated_Worlds)

    # Parse and display terms with links
    st.markdown("## Terms with Links")
    terms1 = extract_terms(Boxing_and_MMA_Commentary_and_Knowledge)
    terms2 = extract_terms(Multiplayer_Custom_Hosting_Game_Servers_For_Simulated_Worlds)
    all_terms = terms1 + terms2
    display_terms_with_links(all_terms)

    # Process 'q' query parameter from the URL
    query_params = st.experimental_get_query_params()
    if 'q' in query_params:
        search_query = query_params['q'][0]
        st.write(f"### Search query received: {search_query}")
        # Perform AI lookup
        ai_result = perform_ai_lookup(search_query)
        # Extract URLs from AI result
        markdown_text = extract_urls(ai_result)
        st.markdown("## Extracted URLs")
        st.markdown(markdown_text)
        # Save the result as markdown file
        filename = generate_filename("AI_Result", search_query)
        with open(filename, 'w', encoding='utf-8') as f:
            f.write(markdown_text)
        st.write(f"Generated file **{filename}** with AI lookup results.")

    # File management sidebar
    file_management_sidebar()

if __name__ == "__main__":
    main()