Enforce download cache quotas without returning evicted paths

This commit is contained in:
Andras Schmelczer 2026-10-27 19:59:19 +00:00
commit 421960726c

View file

@ -0,0 +1,40 @@
import io
import tarfile
from pathlib import Path
from typing import Any
from unittest.mock import MagicMock, patch
import pytest
from great_ai.large_file import LargeFileS3
def test_download_quota_keeps_the_requested_file_and_evicts_older_models(
tmp_path: Path,
) -> None:
backend = MagicMock()
backend.list_objects_v2.side_effect = lambda **kwargs: {
"Contents": [{"Key": kwargs["Prefix"] + "0", "ETag": "one"}]
}
def download(
self: LargeFileS3, remote_path: Any, path: Path, hide_progress: bool
) -> None:
data = b"x" * (2000 if self._name == "oversized" else 600)
with tarfile.open(path, "w:gz") as archive:
member = tarfile.TarInfo(self._local_name)
member.size = len(data)
archive.addfile(member, io.BytesIO(data))
with (
patch.object(LargeFileS3, "_client", backend),
patch.object(LargeFileS3, "_download", download),
patch.object(LargeFileS3, "cache_path", tmp_path / "cache"),
patch.object(LargeFileS3, "max_cache_size", "1 KB"),
):
first = LargeFileS3("first").get()
second = LargeFileS3("second").get()
assert not first.exists()
assert second.read_bytes() == b"x" * 600
with pytest.raises(ValueError, match="exceeds max_cache_size"):
LargeFileS3("oversized").get()
assert second.exists()