Spaces:
Running
Running
File size: 9,089 Bytes
5e3183d ab2a9d9 5e3183d ab2a9d9 5e3183d ab2a9d9 5e3183d ab2a9d9 5e3183d ab2a9d9 5e3183d ab2a9d9 5e3183d ab2a9d9 5e3183d ab2a9d9 5e3183d ab2a9d9 5e3183d ab2a9d9 5e3183d ab2a9d9 5e3183d ab2a9d9 5e3183d ab2a9d9 5e3183d ab2a9d9 5e3183d |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 |
import gradio as gr
import requests
from bs4 import BeautifulSoup
import re
from urllib.parse import urljoin, urlparse
import markdown
from concurrent.futures import ThreadPoolExecutor
import asyncio
from collections import defaultdict
import time
import logging
# Set up logging
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)
class WebsiteCrawler:
def __init__(self, max_depth=3, max_pages=50, timeout=30):
self.max_depth = max_depth
self.max_pages = max_pages
self.timeout = timeout
self.visited_urls = set()
self.url_content = {}
self.url_metadata = defaultdict(dict)
self.headers = {
'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36'
}
def is_valid_url(self, url, base_domain):
"""Check if URL is valid and belongs to the same domain"""
try:
parsed = urlparse(url)
base_parsed = urlparse(base_domain)
return (parsed.netloc == base_parsed.netloc and
parsed.scheme in ['http', 'https'] and
not url.endswith(('.pdf', '.jpg', '.png', '.gif', '.zip')))
except:
return False
def extract_content(self, soup):
"""Extract meaningful content from HTML"""
# Remove script and style elements
for element in soup(['script', 'style', 'nav', 'footer', 'header']):
element.decompose()
# Get main content
main_content = soup.find('main') or soup.find('article') or soup.find('div', {'class': re.compile(r'content|main', re.I)})
if main_content:
return main_content.get_text(strip=True)
return soup.get_text(strip=True)
def get_page_metadata(self, soup, url):
"""Extract metadata from the page"""
metadata = {
'title': None,
'description': None,
'importance': 0,
'category': 'Optional'
}
# Title extraction
metadata['title'] = (
soup.find('meta', property='og:title')['content'] if soup.find('meta', property='og:title') else
soup.find('title').text if soup.find('title') else
soup.find('h1').text if soup.find('h1') else
url.split('/')[-1]
)
# Description extraction
metadata['description'] = (
soup.find('meta', {'name': 'description'})['content'] if soup.find('meta', {'name': 'description'}) else
soup.find('meta', property='og:description')['content'] if soup.find('meta', property='og:description') else
""
)
# Calculate importance based on various factors
importance = 0
if 'docs' in url.lower() or 'documentation' in url.lower():
importance += 5
metadata['category'] = 'Docs'
if 'api' in url.lower():
importance += 4
metadata['category'] = 'API'
if 'guide' in url.lower() or 'tutorial' in url.lower():
importance += 3
metadata['category'] = 'Guides'
if 'example' in url.lower():
importance += 2
metadata['category'] = 'Examples'
if 'blog' in url.lower():
importance += 1
metadata['category'] = 'Blog'
metadata['importance'] = importance
return metadata
async def crawl_page(self, url, depth, base_domain):
"""Crawl a single page and extract information"""
if depth > self.max_depth or url in self.visited_urls or len(self.visited_urls) >= self.max_pages:
return []
try:
response = requests.get(url, headers=self.headers, timeout=self.timeout)
response.raise_for_status()
self.visited_urls.add(url)
soup = BeautifulSoup(response.text, 'html.parser')
content = self.extract_content(soup)
metadata = self.get_page_metadata(soup, url)
self.url_content[url] = content
self.url_metadata[url] = metadata
# Find all links
links = []
for a in soup.find_all('a', href=True):
next_url = urljoin(url, a['href'])
if self.is_valid_url(next_url, base_domain):
links.append(next_url)
return links
except Exception as e:
logger.error(f"Error crawling {url}: {str(e)}")
return []
async def crawl_website(self, start_url):
"""Crawl website starting from the given URL"""
base_domain = start_url
queue = [(start_url, 0)]
seen = {start_url}
while queue and len(self.visited_urls) < self.max_pages:
current_url, depth = queue.pop(0)
if depth > self.max_depth:
continue
links = await self.crawl_page(current_url, depth, base_domain)
for link in links:
if link not in seen:
seen.add(link)
queue.append((link, depth + 1))
def generate_llms_txt(self):
"""Generate llms.txt content from crawled data"""
# Sort URLs by importance
sorted_urls = sorted(
self.url_metadata.items(),
key=lambda x: (x[1]['importance'], x[0]),
reverse=True
)
# Group URLs by category
categorized_urls = defaultdict(list)
for url, metadata in sorted_urls:
categorized_urls[metadata['category']].append((url, metadata))
# Generate content
content = []
# Add main title and description
if sorted_urls:
main_metadata = sorted_urls[0][1]
content.append(f"# {main_metadata['title']}\n")
content.append(f"> {main_metadata['description']}\n")
# Add categorized sections
priority_order = ['Docs', 'API', 'Guides', 'Examples', 'Blog', 'Optional']
for category in priority_order:
if category in categorized_urls:
content.append(f"\n## {category}\n")
for url, metadata in categorized_urls[category]:
title = metadata['title']
desc = metadata['description']
if desc:
content.append(f"- [{title}]({url}): {desc[:100]}...\n")
else:
content.append(f"- [{title}]({url})\n")
return "\n".join(content)
def save_llms_txt(content, save_path="llms.txt"):
"""Save the generated content to a file"""
try:
with open(save_path, 'w', encoding='utf-8') as f:
f.write(content)
return f"Successfully saved to {save_path}"
except Exception as e:
return f"Error saving file: {str(e)}"
async def process_url(url, max_depth, max_pages, save_to_file=False):
"""Process URL and generate llms.txt"""
try:
crawler = WebsiteCrawler(max_depth=int(max_depth), max_pages=int(max_pages))
await crawler.crawl_website(url)
content = crawler.generate_llms_txt()
if save_to_file:
save_message = save_llms_txt(content)
return content, f"Crawled {len(crawler.visited_urls)} pages. {save_message}"
return content, f"Crawled {len(crawler.visited_urls)} pages. File not saved (checkbox not selected)"
except Exception as e:
return "", f"Error: {str(e)}"
# Create the Gradio interface with custom CSS for Open Sans font
css = """
@import url('https://fonts.googleapis.com/css2?family=Open+Sans:wght@400;600&display=swap');
body, .gradio-container {
font-family: 'Open Sans', sans-serif !important;
}
.gr-box {
border-radius: 8px !important;
border: 1px solid #e5e7eb !important;
}
.gr-button {
font-family: 'Open Sans', sans-serif !important;
font-weight: 600 !important;
}
"""
# Create the Gradio interface
iface = gr.Interface(
fn=lambda url, max_depth, max_pages, save: asyncio.run(process_url(url, max_depth, max_pages, save)),
inputs=[
gr.Textbox(label="Website URL", placeholder="Enter the website URL..."),
gr.Slider(minimum=1, maximum=5, value=3, step=1, label="Maximum Crawl Depth"),
gr.Slider(minimum=10, maximum=100, value=50, step=10, label="Maximum Pages to Crawl"),
gr.Checkbox(label="Save to file", value=False)
],
outputs=[
gr.Textbox(label="Generated llms.txt Content", lines=20),
gr.Textbox(label="Status")
],
title="llms.txt Generator",
description="Generate an llms.txt file from a website following the specification. The tool crawls the website and creates a structured markdown file suitable for LLMs.",
examples=[
["https://example.com", 3, 50, False],
["https://docs.python.org", 3, 50, True]
],
theme=gr.themes.Soft(),
css=css
)
# Launch the app
if __name__ == "__main__":
iface.launch() |