Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -2,8 +2,10 @@ import streamlit as st
|
|
| 2 |
import os
|
| 3 |
import glob
|
| 4 |
import re
|
| 5 |
-
import datetime
|
| 6 |
from urllib.parse import quote
|
|
|
|
|
|
|
|
|
|
| 7 |
|
| 8 |
# Define the markdown variables
|
| 9 |
Boxing_and_MMA_Commentary_and_Knowledge = """
|
|
@@ -38,7 +40,6 @@ Boxing_and_MMA_Commentary_and_Knowledge = """
|
|
| 38 |
"""
|
| 39 |
|
| 40 |
Multiplayer_Custom_Hosting_Game_Servers_For_Simulated_Worlds = """
|
| 41 |
-
|
| 42 |
# Multiplayer Simulated Worlds
|
| 43 |
|
| 44 |
1. 7 Days To Die PC
|
|
@@ -70,43 +71,89 @@ Multiplayer_Custom_Hosting_Game_Servers_For_Simulated_Worlds = """
|
|
| 70 |
27. The Forest PC
|
| 71 |
28. Crossplay
|
| 72 |
29. Valheim
|
| 73 |
-
|
| 74 |
"""
|
| 75 |
|
| 76 |
-
# Function to parse markdown text and create external links for terms
|
| 77 |
-
def display_external_links(term):
|
| 78 |
-
search_urls = {
|
| 79 |
-
"ArXiv": lambda k: f"https://arxiv.org/search/?query={quote(k)}",
|
| 80 |
-
"Wikipedia": lambda k: f"https://en.wikipedia.org/wiki/{quote(k)}",
|
| 81 |
-
"Google": lambda k: f"https://www.google.com/search?q={quote(k)}",
|
| 82 |
-
"YouTube": lambda k: f"https://www.youtube.com/results?search_query={quote(k)}",
|
| 83 |
-
}
|
| 84 |
-
links_md = ' | '.join([f"[{name}]({url(term)})" for name, url in search_urls.items()])
|
| 85 |
-
st.markdown(f"- **{term}** - {links_md}")
|
| 86 |
-
|
| 87 |
# Function to parse markdown text and extract terms
|
| 88 |
def extract_terms(markdown_text):
|
| 89 |
-
# Split text into lines
|
| 90 |
lines = markdown_text.strip().split('\n')
|
| 91 |
terms = []
|
| 92 |
for line in lines:
|
| 93 |
-
# Remove markdown special characters
|
| 94 |
line = re.sub(r'^[#*\->\d\.\s]+', '', line).strip()
|
| 95 |
if line:
|
| 96 |
terms.append(line)
|
| 97 |
return terms
|
| 98 |
|
| 99 |
-
# Function to
|
| 100 |
-
def
|
| 101 |
-
|
| 102 |
-
|
| 103 |
-
|
| 104 |
-
|
| 105 |
-
|
| 106 |
-
|
| 107 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 108 |
def generate_filename(prefix, content):
|
| 109 |
-
timestamp = datetime.
|
| 110 |
safe_content = re.sub(r'\W+', '_', content[:50])
|
| 111 |
filename = f"{prefix}_{timestamp}_{safe_content}.md"
|
| 112 |
return filename
|
|
@@ -160,7 +207,7 @@ def file_management_sidebar():
|
|
| 160 |
# Option to create a new markdown file
|
| 161 |
if st.sidebar.button("Create New Markdown File"):
|
| 162 |
# Generate automatic filename
|
| 163 |
-
timestamp = datetime.
|
| 164 |
new_filename = f"note_{timestamp}.md"
|
| 165 |
with open(new_filename, 'w', encoding='utf-8') as f:
|
| 166 |
f.write("# New Markdown File\n")
|
|
@@ -170,46 +217,36 @@ def file_management_sidebar():
|
|
| 170 |
|
| 171 |
# Main application logic
|
| 172 |
def main():
|
| 173 |
-
st.title("Markdown Content with
|
| 174 |
|
| 175 |
# Display the original markdown content
|
| 176 |
st.markdown("## Original Markdown Content")
|
| 177 |
st.markdown(Boxing_and_MMA_Commentary_and_Knowledge)
|
| 178 |
st.markdown(Multiplayer_Custom_Hosting_Game_Servers_For_Simulated_Worlds)
|
| 179 |
|
| 180 |
-
# Parse and display
|
| 181 |
-
st.markdown("##
|
| 182 |
-
|
| 183 |
-
|
| 184 |
-
|
| 185 |
-
|
| 186 |
-
terms = extract_terms(Multiplayer_Custom_Hosting_Game_Servers_For_Simulated_Worlds)
|
| 187 |
-
for term in terms:
|
| 188 |
-
display_external_links(term)
|
| 189 |
-
|
| 190 |
-
# Parse and display internal links
|
| 191 |
-
st.markdown("## Internal Links Generated from Markdown Content")
|
| 192 |
-
terms = extract_terms(Boxing_and_MMA_Commentary_and_Knowledge)
|
| 193 |
-
for term in terms:
|
| 194 |
-
display_internal_links(term)
|
| 195 |
-
|
| 196 |
-
terms = extract_terms(Multiplayer_Custom_Hosting_Game_Servers_For_Simulated_Worlds)
|
| 197 |
-
for term in terms:
|
| 198 |
-
display_internal_links(term)
|
| 199 |
|
| 200 |
# Process 'q' query parameter from the URL
|
| 201 |
query_params = st.experimental_get_query_params()
|
| 202 |
if 'q' in query_params:
|
| 203 |
search_query = query_params['q'][0]
|
| 204 |
st.write(f"### Search query received: {search_query}")
|
| 205 |
-
#
|
| 206 |
-
|
| 207 |
-
|
| 208 |
-
|
| 209 |
-
|
|
|
|
|
|
|
|
|
|
| 210 |
with open(filename, 'w', encoding='utf-8') as f:
|
| 211 |
-
f.write(
|
| 212 |
-
st.write(f"Generated file **{filename}** with
|
| 213 |
|
| 214 |
# File management sidebar
|
| 215 |
file_management_sidebar()
|
|
|
|
| 2 |
import os
|
| 3 |
import glob
|
| 4 |
import re
|
|
|
|
| 5 |
from urllib.parse import quote
|
| 6 |
+
from gradio_client import Client
|
| 7 |
+
import json
|
| 8 |
+
from datetime import datetime
|
| 9 |
|
| 10 |
# Define the markdown variables
|
| 11 |
Boxing_and_MMA_Commentary_and_Knowledge = """
|
|
|
|
| 40 |
"""
|
| 41 |
|
| 42 |
Multiplayer_Custom_Hosting_Game_Servers_For_Simulated_Worlds = """
|
|
|
|
| 43 |
# Multiplayer Simulated Worlds
|
| 44 |
|
| 45 |
1. 7 Days To Die PC
|
|
|
|
| 71 |
27. The Forest PC
|
| 72 |
28. Crossplay
|
| 73 |
29. Valheim
|
|
|
|
| 74 |
"""
|
| 75 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 76 |
# Function to parse markdown text and extract terms
|
| 77 |
def extract_terms(markdown_text):
|
|
|
|
| 78 |
lines = markdown_text.strip().split('\n')
|
| 79 |
terms = []
|
| 80 |
for line in lines:
|
|
|
|
| 81 |
line = re.sub(r'^[#*\->\d\.\s]+', '', line).strip()
|
| 82 |
if line:
|
| 83 |
terms.append(line)
|
| 84 |
return terms
|
| 85 |
|
| 86 |
+
# Function to display terms with links
|
| 87 |
+
def display_terms_with_links(terms):
|
| 88 |
+
search_urls = {
|
| 89 |
+
"🚀🌌ArXiv": lambda k: f"/?q={quote(k)}",
|
| 90 |
+
"📖": lambda k: f"https://en.wikipedia.org/wiki/{quote(k)}",
|
| 91 |
+
"🔍": lambda k: f"https://www.google.com/search?q={quote(k)}",
|
| 92 |
+
"▶️": lambda k: f"https://www.youtube.com/results?search_query={quote(k)}",
|
| 93 |
+
"🔎": lambda k: f"https://www.bing.com/search?q={quote(k)}",
|
| 94 |
+
"🐦": lambda k: f"https://twitter.com/search?q={quote(k)}",
|
| 95 |
+
}
|
| 96 |
+
for term in terms:
|
| 97 |
+
links_md = ' '.join([f"[{emoji}]({url(term)})" for emoji, url in search_urls.items()])
|
| 98 |
+
st.markdown(f"**{term}** {links_md}", unsafe_allow_html=True)
|
| 99 |
+
|
| 100 |
+
# Function to perform AI lookup using Gradio client
|
| 101 |
+
def perform_ai_lookup(query):
|
| 102 |
+
st.write("Performing AI Lookup...")
|
| 103 |
+
# Initialize the Gradio client
|
| 104 |
+
client = Client("awacke1/Arxiv-Paper-Search-And-QA-RAG-Pattern")
|
| 105 |
+
# Perform the AI lookup using the Mixtral and Mistral models
|
| 106 |
+
result1 = client.predict(
|
| 107 |
+
prompt=query,
|
| 108 |
+
llm_model_picked="mistralai/Mixtral-8x7B-Instruct-v0.1",
|
| 109 |
+
stream_outputs=True,
|
| 110 |
+
api_name="/ask_llm"
|
| 111 |
+
)
|
| 112 |
+
st.markdown("### Mixtral-8x7B-Instruct-v0.1 Result")
|
| 113 |
+
st.markdown(result1)
|
| 114 |
+
result2 = client.predict(
|
| 115 |
+
prompt=query,
|
| 116 |
+
llm_model_picked="mistralai/Mistral-7B-Instruct-v0.2",
|
| 117 |
+
stream_outputs=True,
|
| 118 |
+
api_name="/ask_llm"
|
| 119 |
+
)
|
| 120 |
+
st.markdown("### Mistral-7B-Instruct-v0.2 Result")
|
| 121 |
+
st.markdown(result2)
|
| 122 |
+
combined_result = f"{result1}\n\n{result2}"
|
| 123 |
+
return combined_result
|
| 124 |
+
|
| 125 |
+
# Function to extract URLs from AI result
|
| 126 |
+
def extract_urls(text):
|
| 127 |
+
try:
|
| 128 |
+
date_pattern = re.compile(r'### (\d{2} \w{3} \d{4})')
|
| 129 |
+
abs_link_pattern = re.compile(r'\[(.*?)\]\((https://arxiv\.org/abs/\d+\.\d+)\)')
|
| 130 |
+
pdf_link_pattern = re.compile(r'\[⬇️\]\((https://arxiv\.org/pdf/\d+\.\d+)\)')
|
| 131 |
+
title_pattern = re.compile(r'### \d{2} \w{3} \d{4} \| \[(.*?)\]')
|
| 132 |
+
date_matches = date_pattern.findall(text)
|
| 133 |
+
abs_link_matches = abs_link_pattern.findall(text)
|
| 134 |
+
pdf_link_matches = pdf_link_pattern.findall(text)
|
| 135 |
+
title_matches = title_pattern.findall(text)
|
| 136 |
+
|
| 137 |
+
markdown_text = ""
|
| 138 |
+
for i in range(len(date_matches)):
|
| 139 |
+
date = date_matches[i]
|
| 140 |
+
title = title_matches[i]
|
| 141 |
+
abs_link = abs_link_matches[i][1]
|
| 142 |
+
pdf_link = pdf_link_matches[i]
|
| 143 |
+
markdown_text += f"**Date:** {date}\n\n"
|
| 144 |
+
markdown_text += f"**Title:** {title}\n\n"
|
| 145 |
+
markdown_text += f"**Abstract Link:** [{abs_link}]({abs_link})\n\n"
|
| 146 |
+
markdown_text += f"**PDF Link:** [{pdf_link}]({pdf_link})\n\n"
|
| 147 |
+
markdown_text += "---\n\n"
|
| 148 |
+
return markdown_text
|
| 149 |
+
|
| 150 |
+
except Exception as e:
|
| 151 |
+
st.write(f"An error occurred in extract_urls: {e}")
|
| 152 |
+
return ''
|
| 153 |
+
|
| 154 |
+
# Function to generate filename based on date and content
|
| 155 |
def generate_filename(prefix, content):
|
| 156 |
+
timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
|
| 157 |
safe_content = re.sub(r'\W+', '_', content[:50])
|
| 158 |
filename = f"{prefix}_{timestamp}_{safe_content}.md"
|
| 159 |
return filename
|
|
|
|
| 207 |
# Option to create a new markdown file
|
| 208 |
if st.sidebar.button("Create New Markdown File"):
|
| 209 |
# Generate automatic filename
|
| 210 |
+
timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
|
| 211 |
new_filename = f"note_{timestamp}.md"
|
| 212 |
with open(new_filename, 'w', encoding='utf-8') as f:
|
| 213 |
f.write("# New Markdown File\n")
|
|
|
|
| 217 |
|
| 218 |
# Main application logic
|
| 219 |
def main():
|
| 220 |
+
st.title("Markdown Content with AI Lookup and File Management")
|
| 221 |
|
| 222 |
# Display the original markdown content
|
| 223 |
st.markdown("## Original Markdown Content")
|
| 224 |
st.markdown(Boxing_and_MMA_Commentary_and_Knowledge)
|
| 225 |
st.markdown(Multiplayer_Custom_Hosting_Game_Servers_For_Simulated_Worlds)
|
| 226 |
|
| 227 |
+
# Parse and display terms with links
|
| 228 |
+
st.markdown("## Terms with Links")
|
| 229 |
+
terms1 = extract_terms(Boxing_and_MMA_Commentary_and_Knowledge)
|
| 230 |
+
terms2 = extract_terms(Multiplayer_Custom_Hosting_Game_Servers_For_Simulated_Worlds)
|
| 231 |
+
all_terms = terms1 + terms2
|
| 232 |
+
display_terms_with_links(all_terms)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 233 |
|
| 234 |
# Process 'q' query parameter from the URL
|
| 235 |
query_params = st.experimental_get_query_params()
|
| 236 |
if 'q' in query_params:
|
| 237 |
search_query = query_params['q'][0]
|
| 238 |
st.write(f"### Search query received: {search_query}")
|
| 239 |
+
# Perform AI lookup
|
| 240 |
+
ai_result = perform_ai_lookup(search_query)
|
| 241 |
+
# Extract URLs from AI result
|
| 242 |
+
markdown_text = extract_urls(ai_result)
|
| 243 |
+
st.markdown("## Extracted URLs")
|
| 244 |
+
st.markdown(markdown_text)
|
| 245 |
+
# Save the result as markdown file
|
| 246 |
+
filename = generate_filename("AI_Result", search_query)
|
| 247 |
with open(filename, 'w', encoding='utf-8') as f:
|
| 248 |
+
f.write(markdown_text)
|
| 249 |
+
st.write(f"Generated file **{filename}** with AI lookup results.")
|
| 250 |
|
| 251 |
# File management sidebar
|
| 252 |
file_management_sidebar()
|