import re import logging from flask import Flask, request, Response import requests from urllib.parse import urlparse, unquote # --- Basic Configuration --- logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s') app = Flask(__name__) # --- Whitelisted URL Patterns for GitHub --- # This list defines all URL patterns that are permitted to be proxied. # It's a security measure to prevent the proxy from being used to access unintended domains. ALLOWED_PATTERNS = [ # Repositories: releases, archives, blobs, raw content re.compile(r'^https://github\.com/[^/]+/[^/]+/(?:releases|archive)/.*$', re.IGNORECASE), re.compile(r'^https://github\.com/[^/]+/[^/]+/(?:blob|raw)/.*$', re.IGNORECASE), # Git operations (clone, pull, push) re.compile(r'^https://github\.com/[^/]+/[^/]+/(?:info|git-).*$', re.IGNORECASE), # Raw content from various GitHub domains re.compile(r'^https://raw\.(?:githubusercontent|github)\.com/[^/]+/[^/]+/.*/.*$', re.IGNORECASE), re.compile(r'^https://gist\.(?:githubusercontent|github)\.com/[^/]+/[^/]+/.*/.*$', re.IGNORECASE), # Repository tags and assets re.compile(r'^https://github\.com/[^/]+/[^/]+/tags.*$', re.IGNORECASE), re.compile(r'^https://avatars\.githubusercontent\.com/.*$', re.IGNORECASE), re.compile(r'^https://github\.githubassets\.com/.*$', re.IGNORECASE), # Main repository/user pages re.compile(r'^https://github\.com/[^/]+/?$', re.IGNORECASE), re.compile(r'^https://github\.com/[^/]+/[^/]+/?$', re.IGNORECASE), ] # --- Custom Index Page --- INDEX_PAGE_HTML = """ GitHub Proxy

GitHub Reverse Proxy

Warning: You are accessing GitHub content through a reverse proxy. All content served from this page is provided directly by GitHub.

How to Use

To access GitHub content, simply append the GitHub URL to this proxy's address.

For example, to clone a repository:

git clone {YOUR_PROXY_URL}/https://github.com/owner/repo.git

Or to view a repository page:

{YOUR_PROXY_URL}/https://github.com/owner/repo
""" def is_url_allowed(url): """Check if the given URL matches any pattern in the whitelist.""" for pattern in ALLOWED_PATTERNS: if pattern.match(url): return True return False # --- Core Proxy Logic --- @app.route('/', defaults={'path': ''}, methods=['GET', 'POST', 'PUT', 'DELETE']) @app.route('/', methods=['GET', 'POST', 'PUT', 'DELETE']) def proxy(path): """ Proxies requests to GitHub after validating them against a whitelist. It streams the response back to the client. """ # The full path might be URL-encoded, so we decode it. target_path = unquote(request.full_path) if target_path.startswith('/'): target_path = target_path[1:] # For the root path, display the custom warning page. if not target_path: return INDEX_PAGE_HTML, 200 # Prepend 'https://' if the scheme is missing. if not target_path.startswith(('http://', 'https://')): target_url = 'https://' + target_path else: target_url = target_path # Security check: Ensure the URL is in the whitelist. if not is_url_allowed(target_url): logging.warning(f"URL Denied! No pattern matched: {target_url}") return "

403 Forbidden

Request blocked by proxy security policy.

", 403 try: target_host = urlparse(target_url).hostname if not target_host: raise ValueError("Hostname could not be parsed from the target URL.") except Exception as e: logging.error(f"Invalid target URL provided: {target_url} | Error: {e}") return f"Invalid target URL in path: {e}", 400 # Forward headers, but set the 'Host' header to the target's hostname. headers = {key: value for (key, value) in request.headers if key.lower() != 'host'} headers['Host'] = target_host try: # Stream the request to handle large files efficiently. resp = requests.request( method=request.method, url=target_url, headers=headers, data=request.get_data(), cookies=request.cookies, allow_redirects=False, # Redirects are handled by the client. stream=True, timeout=30 # Timeout for the connection. ) # Exclude headers that can interfere with streaming. excluded_headers = ['content-encoding', 'content-length', 'transfer-encoding', 'connection'] response_headers = [(name, value) for (name, value) in resp.raw.headers.items() if name.lower() not in excluded_headers] # Stream the response back to the original client. return Response(resp.iter_content(chunk_size=8192), resp.status_code, response_headers) except requests.exceptions.RequestException as e: logging.error(f"Error while proxying to {target_url}: {e}") return "An error occurred while proxying the request.", 502 if __name__ == '__main__': app.run(host='0.0.0.0', port=7860)