fix(ingest): don't advance incremental cursor past a failed/sampled asset

A later successful asset could push last_ingest_at past an earlier asset whose
thumbnail download failed, permanently excluding it from later incremental runs
(only --full recovered it). Now the cursor never advances to/past the earliest
failed asset, never below the prior cursor, and not at all on a --subset run.
Regression tests added. (review finding #9)
This commit is contained in:
2026-06-27 18:49:24 +02:00
parent ae093881fa
commit bd8fe59b49
2 changed files with 54 additions and 5 deletions
+36
View File
@@ -102,3 +102,39 @@ def test_ingest_resilient_to_thumb_failure_and_zero_byte(tmp_path):
# The 0-byte file for "b" was replaced with real bytes.
assert os.path.getsize(os.path.join(thumbs, "b.jpg")) > 0
store.close()
def test_ingest_watermark_not_advanced_past_failed_asset(tmp_path):
# A later successful asset must NOT advance last_ingest_at past an earlier
# asset whose thumbnail failed, or that asset is stranded on the next
# incremental run (review finding #9).
store = Store(str(tmp_path / "t.db")).connect()
thumbs = str(tmp_path / "thumbs"); os.makedirs(thumbs, exist_ok=True)
class FailOld(FakeImmich):
def download_thumbnail(self, asset_id):
if asset_id == "old":
raise RuntimeError("boom")
return self._thumb
client = FailOld([_asset("old", "2008-06-01", updated="2026-01-01T00:00:00Z"),
_asset("new", "2024-03-01", updated="2026-03-01T00:00:00Z")])
run_ingest(client, store, thumbs)
assert store.get_asset("old") is None and store.get_asset("new") is not None
wm = store.get_meta("last_ingest_at")
# cursor must stay below the failed asset (here: not advanced at all)
assert wm is None or wm < "2026-01-01T00:00:00Z"
assert wm != "2026-03-01T00:00:00Z"
store.close()
def test_ingest_subset_run_does_not_advance_watermark(tmp_path):
# --subset is a sampling run; it must not move the incremental cursor or it
# would strand the un-fetched remainder.
store = Store(str(tmp_path / "t.db")).connect()
thumbs = str(tmp_path / "thumbs"); os.makedirs(thumbs, exist_ok=True)
client = FakeImmich([_asset("a", "2019-06-01", updated="2026-02-01T00:00:00Z"),
_asset("b", "2019-06-02", updated="2026-02-02T00:00:00Z")])
run_ingest(client, store, thumbs, subset=1)
assert store.get_meta("last_ingest_at") is None
store.close()