def test_save_data(
self,
dummy_data,
tmp_path,
filepath_json,
save_version,
):
"""Test saving and reloading the data set."""
metrics_dataset = MetricsDataSet(filepath=filepath_json,
version=Version(None, save_version))
metrics_dataset.save(dummy_data)
actual_filepath = Path(metrics_dataset._filepath.as_posix())
test_filepath = tmp_path / "locally_saved.json"
test_filepath.parent.mkdir(parents=True, exist_ok=True)
with open(test_filepath, "w", encoding="utf-8") as file:
json.dump(dummy_data, file)
with open(test_filepath, encoding="utf-8") as file:
test_data = json.load(file)
with open((actual_filepath / save_version / "test.json"),
encoding="utf-8") as actual_file:
actual_data = json.load(actual_file)
assert actual_data == test_data
assert metrics_dataset._fs_open_args_load == {}
assert metrics_dataset._fs_open_args_save == {"mode": "w"}
def test_protocol_usage(self, filepath, instance_type):
data_set = MetricsDataSet(filepath=filepath)
assert isinstance(data_set._fs, instance_type)
path = filepath.split(PROTOCOL_DELIMITER, 1)[-1]
assert str(data_set._filepath) == path
assert isinstance(data_set._filepath, PurePosixPath)
def test_not_version_str_repr(self):
"""Test that version is not in string representation of the class instance."""
filepath = "test.json"
ds = MetricsDataSet(filepath=filepath)
assert filepath in str(ds)
assert "version" not in str(ds)
assert "MetricsDataSet" in str(ds)
assert "protocol" in str(ds)
# Default save_args
assert "save_args={'indent': 2}" in str(ds)
def test_version_str_repr(self):
"""Test that version is in string representation of the class instance
when applicable."""
filepath = "test.json"
ds_versioned = MetricsDataSet(filepath=filepath)
assert filepath in str(ds_versioned)
ver_str = "version=Version(load=None, save=None)"
assert ver_str in str(ds_versioned)
assert "MetricsDataSet" in str(ds_versioned)
assert "protocol" in str(ds_versioned)
# Default save_args
assert "save_args={'indent': 2}" in str(ds_versioned)
def test_version_str_repr(self, load_version, save_version):
"""Test that version is in string representation of the class instance."""
filepath = "test.json"
ds_versioned = MetricsDataSet(filepath=filepath,
version=Version(load_version,
save_version))
assert filepath in str(ds_versioned)
ver_str = f"version=Version(load={load_version}, save='{save_version}')"
assert ver_str in str(ds_versioned)
assert "MetricsDataSet" in str(ds_versioned)
assert "protocol" in str(ds_versioned)
# Default save_args
assert "save_args={'indent': 2}" in str(ds_versioned)
def explicit_versioned_metrics_dataset(filepath_json, load_version,
save_version):
return MetricsDataSet(filepath=filepath_json,
version=Version(load_version, save_version))
def test_http_filesystem_no_versioning(self):
pattern = r"HTTP\(s\) DataSet doesn't support versioning\."
with pytest.raises(DataSetError, match=pattern):
MetricsDataSet(filepath="https://example.com/file.json",
version=Version(None, None))
def metrics_dataset(filepath_json, save_args, fs_args):
return MetricsDataSet(filepath=filepath_json,
save_args=save_args,
fs_args=fs_args)
def test_catalog_release(self, mocker):
fs_mock = mocker.patch("fsspec.filesystem").return_value
filepath = "test.json"
data_set = MetricsDataSet(filepath=filepath)
data_set.release()
fs_mock.invalidate_cache.assert_called_once_with(filepath)