def test_move_on_disk(test_dataset: DatasetForTests,
integration_test_data: Path,
other_dataset: DatasetForTests):
# type: (Tuple[Collection, DatasetLite, str, Path]) -> None
"""
Indexed dataset was moved over the top of another indexed dataset
"""
test_dataset.add_to_index()
other_dataset.add_to_index()
shutil.move(other_dataset.path, str(uri_to_local_path(test_dataset.uri)))
_check_sync(
collection=test_dataset.collection,
expected_paths=[
test_dataset.uri,
other_dataset.path.as_uri(),
],
expected_mismatches=[
mm.LocationMissingOnDisk(test_dataset.dataset, test_dataset.uri),
mm.LocationNotIndexed(other_dataset.dataset, test_dataset.uri),
mm.LocationMissingOnDisk(other_dataset.dataset, other_dataset.path.as_uri()),
],
expected_index_result={
test_dataset.dataset: (),
other_dataset.dataset: (test_dataset.uri,),
test_dataset.parent: (),
},
cache_path=integration_test_data,
fix_settings=dict(index_missing=True, update_locations=True)
)
def test_replace_on_disk(test_dataset: DatasetForTests,
integration_test_data: Path,
other_dataset: DatasetForTests):
# type: (Tuple[Collection, DatasetLite, str, Path]) -> None
"""
File on disk has a different id to the one in the index (ie. it was quietly reprocessed)
"""
test_dataset.add_to_index()
# move a new one over the top
shutil.move(other_dataset.path, str(uri_to_local_path(test_dataset.uri)))
_check_sync(
collection=test_dataset.collection,
expected_paths=[
test_dataset.uri
],
expected_mismatches=[
mm.LocationMissingOnDisk(test_dataset.dataset, test_dataset.uri),
mm.DatasetNotIndexed(other_dataset.dataset, test_dataset.uri),
],
expected_index_result={
test_dataset.dataset: (),
other_dataset.dataset: (test_dataset.uri,),
test_dataset.parent: (),
},
cache_path=integration_test_data,
fix_settings=dict(index_missing=True, update_locations=True)
)
def test_new_and_old_on_disk(test_dataset: DatasetForTests,
integration_test_data: Path,
other_dataset: DatasetForTests):
# type: (Tuple[Collection, DatasetLite, str, Path]) -> None
# ls8_collection, on_disk, on_disk_uri, root = syncable_environment
old_indexed = DatasetLite(uuid.UUID('5294efa6-348d-11e7-a079-185e0f80a5c0'))
# An indexed file not on disk, and disk file not in index.
missing_dataset = other_dataset
missing_dataset.add_to_index()
# Make it missing
shutil.rmtree(str(missing_dataset.copyable_path))
_check_sync(
collection=test_dataset.collection,
expected_paths=[
missing_dataset.uri,
test_dataset.uri
],
expected_mismatches=[
mm.LocationMissingOnDisk(old_indexed, missing_dataset.uri),
mm.DatasetNotIndexed(test_dataset.dataset, test_dataset.uri)
],
expected_index_result={
test_dataset.dataset: (test_dataset.uri,),
old_indexed: (),
test_dataset.parent: (),
},
cache_path=integration_test_data,
fix_settings=dict(index_missing=True, update_locations=True)
)