Restrict refresh_remote to bounded HTTP(S) fetches

pyisolate/policy/__init__.py

@@ -4,6 +4,7 @@
 import os
 import socket
 import tempfile
+import urllib.parse
 import urllib.request
 from collections.abc import Iterable, Mapping
 from dataclasses import dataclass, field
@@ -276,6 +277,11 @@ def _is_timeout_error(exc: Exception) -> bool:
     return False
 
 
+# Policy documents are small YAML files; the bound exists because the response
+# body comes from the network and is buffered in full before validation.
+_MAX_REMOTE_POLICY_BYTES = 1 << 20  # 1 MiB
+
+
 def refresh_remote(
     url: str,
     token: str,
@@ -284,13 +290,23 @@ def refresh_remote(
     *,
     dry_run: bool = False,
 ):
-    """Fetch policy YAML from *url* and apply it."""
+    """Fetch policy YAML from *url* over HTTP(S) and apply it."""
+    scheme = urllib.parse.urlsplit(url).scheme.lower()
+    if scheme not in {"http", "https"}:
+        raise ValueError(f"policy URL scheme must be http or https, got {scheme!r}")
+
     attempts = max(1, max_retries + 1)
 
     for attempt in range(attempts):
         try:
             with urllib.request.urlopen(url, timeout=timeout) as fh:
-                text = fh.read().decode("utf-8")
+                raw = fh.read(_MAX_REMOTE_POLICY_BYTES + 1)
+            if len(raw) > _MAX_REMOTE_POLICY_BYTES:
+                raise ValueError(
+                    f"remote policy from {url} exceeds "
+                    f"{_MAX_REMOTE_POLICY_BYTES} bytes"
+                )
+            text = raw.decode("utf-8")
             break
         except Exception as exc:  # narrow to timeout conditions only
             if _is_timeout_error(exc):

tests/test_remote_policy.py

@@ -127,3 +127,44 @@ def test_refresh_remote_timeout_error(tmp_path):
     finally:
         httpd.shutdown()
         thread.join()
+
+
+def test_refresh_remote_rejects_non_http_scheme(tmp_path):
+    # urllib would happily open file:// (or ftp://) URLs; the policy fetch is
+    # documented as HTTP and must not turn into a local-file read primitive.
+    doc = tmp_path / "policy.yml"
+    doc.write_text("version: 0.1\n", encoding="utf-8")
+
+    with pytest.raises(ValueError, match="scheme"):
+        policy.refresh_remote(doc.as_uri(), token="tok")
+
+
+class OversizedHandler(BaseHTTPRequestHandler):
+    def do_GET(self):
+        body = b"#" * (policy._MAX_REMOTE_POLICY_BYTES + 1)
+        self.send_response(200)
+        self.send_header("Content-Type", "text/plain")
+        self.send_header("Content-Length", str(len(body)))
+        self.end_headers()
+        try:
+            self.wfile.write(body)
+        except BrokenPipeError:  # pragma: no cover - client may bail early
+            pass
+
+    def log_message(self, format, *args):  # pragma: no cover - quiet test output
+        return
+
+
+def test_refresh_remote_rejects_oversized_response():
+    addr = ("127.0.0.1", 0)
+    httpd = HTTPServer(addr, OversizedHandler)
+    port = httpd.server_address[1]
+
+    thread = threading.Thread(target=httpd.serve_forever, daemon=True)
+    thread.start()
+    try:
+        with pytest.raises(ValueError, match="exceeds"):
+            policy.refresh_remote(f"http://127.0.0.1:{port}", token="tok")
+    finally:
+        httpd.shutdown()
+        thread.join()