mirror of
https://github.com/calibrain/shelfmark.git
synced 2026-11-14 20:38:30 +00:00
splitsec2
973cf472a2
`test_cache_hit_flushes_the_index_once_the_window_has_passed` (from #1544) failed on the first attempt of two of my PRs' Python Tests runs today (#1576, #1579) and passed on rerun, with nothing in either PR touching the image cache. The test sets the clock to exactly `last_saved + ACCESS_FLUSH_SECONDS` and expects a flush. `(now + 30) - now` isn't always 30.0 in floating point. For example `105.65207077218265 + 30 - 105.65207077218265` is `29.999999999999986`, so the `>=` check is false and the flush never happens. That only shows up when the monotonic value is small, which fits a freshly booted runner (on this box the uptime is in the millions of seconds and it never fails). I haven't been able to see the runner's actual uptime in the CI logs, so that part is an inference, but pinning `_last_saved` to that value reproduces the exact failure (`assert [] == [1]`) locally. The test now puts the clock one second past the window. The production check is fine, real elapsed time doesn't have this edge, so only the test changes.
126 lines
4.4 KiB
Python
126 lines
4.4 KiB
Python
"""Tests for targeted image cache safety and fetch fallbacks."""
|
|
|
|
import requests
|
|
|
|
from shelfmark.core.image_cache import ImageCacheService
|
|
|
|
|
|
def test_fetch_and_cache_rejects_backslash_authority_bypass_before_request(
|
|
tmp_path, monkeypatch
|
|
) -> None:
|
|
cache = ImageCacheService(tmp_path)
|
|
calls = []
|
|
|
|
def fake_get(url, **_kwargs):
|
|
calls.append(url)
|
|
raise AssertionError("unsafe URL should not be requested")
|
|
|
|
monkeypatch.setattr("shelfmark.core.image_cache.requests.get", fake_get)
|
|
|
|
assert cache.fetch_and_cache("cover-ssrf", "http://127.0.0.1:6666\\@1.1.1.1") is None
|
|
assert calls == []
|
|
assert "cover-ssrf" not in cache._index
|
|
|
|
|
|
def test_is_safe_url_rejects_encoded_separator_in_authority() -> None:
|
|
assert ImageCacheService._is_safe_url("http://127.0.0.1:6666%5c@1.1.1.1") is False
|
|
assert ImageCacheService._is_safe_url("http://127.0.0.1:6666%2f@1.1.1.1") is False
|
|
|
|
|
|
def test_is_safe_url_rejects_invalid_ipv6_url() -> None:
|
|
assert ImageCacheService._is_safe_url("http://[") is False
|
|
|
|
|
|
def test_fetch_and_cache_blocks_unsafe_redirect(tmp_path, monkeypatch) -> None:
|
|
cache = ImageCacheService(tmp_path)
|
|
|
|
def fake_getaddrinfo(hostname, *_args, **_kwargs):
|
|
addresses = {
|
|
"example.com": "93.184.216.34",
|
|
"127.0.0.1": "127.0.0.1",
|
|
}
|
|
return [(None, None, None, None, (addresses[hostname], 0))]
|
|
|
|
class RedirectResponse:
|
|
is_redirect = True
|
|
headers = {"location": "http://127.0.0.1/cover.jpg"}
|
|
|
|
def close(self):
|
|
return None
|
|
|
|
calls = []
|
|
|
|
def fake_get(url, **_kwargs):
|
|
calls.append(url)
|
|
return RedirectResponse()
|
|
|
|
monkeypatch.setattr("shelfmark.download.network.socket.getaddrinfo", fake_getaddrinfo)
|
|
monkeypatch.setattr("shelfmark.core.image_cache.requests.get", fake_get)
|
|
|
|
assert cache.fetch_and_cache("cover-redirect", "https://example.com/cover.jpg") is None
|
|
assert calls == ["https://example.com/cover.jpg"]
|
|
assert "cover-redirect" not in cache._index
|
|
|
|
|
|
def test_fetch_and_cache_returns_none_on_request_exception(tmp_path, monkeypatch) -> None:
|
|
cache = ImageCacheService(tmp_path)
|
|
|
|
def fake_get(*_args, **_kwargs):
|
|
raise requests.exceptions.TooManyRedirects("too many redirects")
|
|
|
|
monkeypatch.setattr("shelfmark.core.image_cache.requests.get", fake_get)
|
|
|
|
assert cache.fetch_and_cache("cover-1", "https://example.com/cover.jpg") is None
|
|
assert "cover-1" not in cache._index
|
|
|
|
|
|
def test_cover_cache_key_depends_on_the_url_only() -> None:
|
|
from shelfmark.core.image_cache import cover_cache_key
|
|
|
|
key = cover_cache_key("https://example.com/cover.jpg")
|
|
|
|
assert key == cover_cache_key(" https://example.com/cover.jpg \n")
|
|
assert key != cover_cache_key("https://example.com/other.jpg")
|
|
assert key != cover_cache_key("https://example.com/cover.jpg?size=large")
|
|
assert key.isalnum() and key == key.lower()
|
|
assert "/" not in key and "." not in key
|
|
|
|
|
|
def test_cache_hits_do_not_rewrite_the_index_each_time(tmp_path, monkeypatch) -> None:
|
|
"""A hit only moves the last-access time, so the index is flushed on a timer."""
|
|
cache = ImageCacheService(tmp_path)
|
|
assert cache.put("cover-a", b"\xff\xd8\xff" + b"0" * 64, "image/jpeg")
|
|
saves: list[int] = []
|
|
real_save = cache._save_index
|
|
monkeypatch.setattr(cache, "_save_index", lambda: (saves.append(1), real_save())[1])
|
|
|
|
before = cache._index["cover-a"]["accessed_at"]
|
|
for _ in range(5):
|
|
assert cache.get("cover-a") is not None
|
|
|
|
assert saves == []
|
|
assert cache._index["cover-a"]["accessed_at"] >= before
|
|
|
|
|
|
def test_cache_hit_flushes_the_index_once_the_window_has_passed(tmp_path, monkeypatch) -> None:
|
|
from shelfmark.core import image_cache
|
|
|
|
cache = ImageCacheService(tmp_path)
|
|
assert cache.put("cover-a", b"\xff\xd8\xff" + b"0" * 64, "image/jpeg")
|
|
saves: list[int] = []
|
|
real_save = cache._save_index
|
|
monkeypatch.setattr(cache, "_save_index", lambda: (saves.append(1), real_save())[1])
|
|
|
|
now = cache._last_saved
|
|
# One second past the window. Exactly on it, (now + 30) - now can come out a hair under 30
|
|
# for the small monotonic values of a freshly booted runner.
|
|
monkeypatch.setattr(
|
|
image_cache.time, "monotonic", lambda: now + image_cache.ACCESS_FLUSH_SECONDS + 1
|
|
)
|
|
assert cache.get("cover-a") is not None
|
|
assert saves == [1]
|
|
|
|
# The flush restarts the window.
|
|
assert cache.get("cover-a") is not None
|
|
assert saves == [1]
|