Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion pyiceberg/table/update/snapshot.py
Original file line number Diff line number Diff line change
Expand Up @@ -652,7 +652,7 @@ def _get_entries(manifest: ManifestFile) -> list[ManifestEntry]:
return [
ManifestEntry.from_args(
status=ManifestEntryStatus.DELETED,
snapshot_id=entry.snapshot_id,
snapshot_id=self._snapshot_id,
sequence_number=entry.sequence_number,
file_sequence_number=entry.file_sequence_number,
data_file=entry.data_file,
Expand Down
49 changes: 49 additions & 0 deletions tests/integration/test_deletes.py
Original file line number Diff line number Diff line change
Expand Up @@ -975,3 +975,52 @@ def assert_manifest_entry(expected_status: ManifestEntryStatus, expected_snapsho
assert after_delete_snapshot is not None

assert_manifest_entry(ManifestEntryStatus.DELETED, after_delete_snapshot.snapshot_id)


@pytest.mark.integration
def test_manifest_entry_snapshot_id_after_partial_deletes(session_catalog: RestCatalog) -> None:
identifier = "default.test_manifest_entry_snapshot_id_after_partial_deletes"
try:
session_catalog.drop_table(identifier)
except NoSuchTableError:
pass

schema = pa.schema(
[
("id", pa.int32()),
("name", pa.string()),
]
)

table = session_catalog.create_table(identifier, schema)
data = pa.Table.from_pylist(
[
{"id": 1, "name": "keep"},
{"id": 2, "name": "keep"},
{"id": 3, "name": "delete_me"},
{"id": 4, "name": "delete_me"},
],
schema=schema,
)
table.append(data)

# Partial delete: only some rows match, triggering CoW overwrite via _OverwriteFiles
table.delete(EqualTo("name", "delete_me"))
after_delete_snapshot = table.refresh().current_snapshot()
assert after_delete_snapshot is not None

manifests = after_delete_snapshot.manifests(table.io)
deleted_entries = [
entry
for manifest in manifests
for entry in manifest.fetch_manifest_entry(table.io, discard_deleted=False)
if entry.status == ManifestEntryStatus.DELETED
]

assert len(deleted_entries) > 0, "Expected at least one DELETED manifest entry from the CoW overwrite"

for entry in deleted_entries:
assert entry.snapshot_id == after_delete_snapshot.snapshot_id, (
f"DELETED entry snapshot_id should be {after_delete_snapshot.snapshot_id} "
f"(the deleting snapshot), but was {entry.snapshot_id}"
)
Loading