copilot-swe-agent[bot]
8984d4da70
- Resolve merge conflicts in .jules/sentinel.md and app/utils/network.py - Refactor join_url() to use urllib.parse.urlsplit/urlunsplit and posixpath instead of sentinel-string hack, preventing corruption for any input URL - Fix test to use pytest tmp_path fixture instead of hard-coded /tmp/workdir
64 lines
2.5 KiB
Python
64 lines
2.5 KiB
Python
import ipaddress
|
|
import logging
|
|
import posixpath
|
|
import socket
|
|
from urllib.parse import urlsplit, urlunsplit
|
|
|
|
logger = logging.getLogger(__name__)
|
|
|
|
|
|
def is_private_ip(hostname: str) -> bool:
|
|
"""
|
|
Check if a hostname resolves to a private/internal IP address.
|
|
Protects against SSRF attacks by blocking access to internal networks.
|
|
"""
|
|
try:
|
|
# Try to parse as IP address directly
|
|
ip = ipaddress.ip_address(hostname)
|
|
return ip.is_private or ip.is_loopback or ip.is_link_local or ip.is_reserved
|
|
except ValueError:
|
|
# Not a direct IP, try to resolve hostname
|
|
try:
|
|
# Get all IP addresses for this hostname
|
|
addr_info = socket.getaddrinfo(hostname, None)
|
|
for info in addr_info:
|
|
ip_str = info[4][0]
|
|
ip = ipaddress.ip_address(ip_str)
|
|
# Block if ANY resolved IP is private/internal
|
|
if ip.is_private or ip.is_loopback or ip.is_link_local or ip.is_reserved:
|
|
return True
|
|
return False
|
|
except (socket.gaierror, socket.error):
|
|
# Cannot resolve.
|
|
# Fail securely: block unresolved domains to prevent DNS rebinding
|
|
# and SSRF bypasses via unresolvable addresses.
|
|
logger.warning(f"Could not resolve hostname (blocking securely): {hostname}")
|
|
return True
|
|
|
|
|
|
def join_url(base: str, *parts: str) -> str:
|
|
"""
|
|
Safely join a base URL with one or more path parts.
|
|
|
|
Uses urllib.parse to correctly handle scheme/netloc/query/fragment so that
|
|
only the path component is normalised (double slashes removed via
|
|
posixpath.join). The scheme separator ``://`` is therefore never at risk
|
|
of being collapsed.
|
|
|
|
Examples:
|
|
join_url("https://example.com/dav/", "/remote/", "file.pdf")
|
|
-> "https://example.com/dav/remote/file.pdf"
|
|
"""
|
|
parsed = urlsplit(base)
|
|
# Strip leading/trailing slashes from every part so posixpath.join
|
|
# produces a clean joined path without accidental double slashes.
|
|
stripped_parts = [p.strip("/") for p in parts if p.strip("/")]
|
|
base_path = parsed.path.rstrip("/")
|
|
if stripped_parts:
|
|
new_path = base_path + "/" + "/".join(stripped_parts)
|
|
else:
|
|
new_path = base_path
|
|
# Normalise any remaining double slashes in the path only.
|
|
new_path = posixpath.normpath(new_path) if new_path else "/"
|
|
return urlunsplit((parsed.scheme, parsed.netloc, new_path, parsed.query, parsed.fragment))
|