R11: Validate URL host before network I/O — prevent SSRF

- Add _validate_url_host() to check allowlist before urlopen()
- Validate both original URL and redirect target
- Reject localhost, internal IPs, and non-allowlisted hosts
- Add 2 unit tests for disallowed host rejection
- Prevents SSRF attacks via malicious source URLs
This commit is contained in:
ik
2026-09-10 05:45:57 +07:00
parent 6e0077bd4d
commit 39f66481be
2 changed files with 46 additions and 3 deletions
+26
View File
@@ -31,3 +31,29 @@ def test_failed_fetch_still_reserves_site_cooldown(tmp_path: Path, monkeypatch:
monkeypatch.setattr(community_cli, "fetch_html", fail)
assert community_cli.main(["rf4db", "--state-file", str(state_file)]) == 1
assert "download.rf4db.com" in json.loads(state_file.read_text(encoding="utf-8"))
def test_validate_url_host_rejects_disallowed_hosts() -> None:
from rf4_research.community_cli import _validate_url_host
# Allowed hosts pass
assert _validate_url_host("https://download.rf4db.com/ru/catches") == "download.rf4db.com"
assert _validate_url_host("https://www.rf4-stat.ru/posts/") == "rf4-stat.ru"
assert _validate_url_host("https://rf4map.ru/point/123") == "rf4map.ru"
# Disallowed hosts raise ValueError before network I/O
with pytest.raises(ValueError, match="not in allowlist"):
_validate_url_host("http://localhost:8080/admin")
with pytest.raises(ValueError, match="not in allowlist"):
_validate_url_host("http://169.254.169.254/latest/meta-data/")
with pytest.raises(ValueError, match="not in allowlist"):
_validate_url_host("http://internal-service.corp/api")
def test_validate_url_host_rejects_missing_hostname() -> None:
from rf4_research.community_cli import _validate_url_host
with pytest.raises(ValueError, match="valid hostname"):
_validate_url_host("not-a-valid-url")
with pytest.raises(ValueError, match="valid hostname"):
_validate_url_host("")