def test_dataset_3(test_data):
with pytest.raises(RuntimeError, match='is not a legal path'):
Dataset.from_http(test_data[1], 'not_exist')
tmp = Dataset.from_http(test_data[1], save_to=test_data[2])
assert tmp == str(test_data[2] / 'fetch_test.txt')
assert Path(tmp).exists()
with open(tmp, 'r') as f:
assert f.readline() == 'Test xenonpy.utils.Loader._fetch_data'
tmp = Dataset.from_http(test_data[1], save_to=test_data[2], filename=test_data[0])
assert tmp == str(test_data[2] / 'rename.txt')
assert Path(tmp).exists()
with open(tmp, 'r') as f:
assert f.readline() == 'Test xenonpy.utils.Loader._fetch_data'
def test_dataset_1(test_data):
path = Path(__file__).parents[0]
ds = Dataset()
assert ds._backend == 'pandas'
assert ds._paths == ('.', )
assert ds._prefix == ()
with pytest.warns(RuntimeWarning):
Dataset(str(path), str(path))
with pytest.raises(RuntimeError):
Dataset('no_exist_dir')
ds = Dataset(str(path), backend='pickle', prefix=('datatools', ))
assert hasattr(ds, 'datatools_test')
tmp = '%s' % ds
assert 'Dataset' in tmp
def test_dataset_2(test_data):
path = Path(__file__).parents[0]
ds = Dataset(str(path), backend='pickle')
assert hasattr(ds, 'test')
tmp = ds.test
assert isinstance(tmp, list)
assert tmp == [[1, 2], [3, 4]]
tmp = ds.csv
assert hasattr(tmp, 'test')
tmp = tmp.test
assert isinstance(tmp, pd.DataFrame)
assert np.all(np.array([[0, 1, 2], [1, 3, 4]]) == tmp.values)
tmp = ds.csv(str(path / 'test.csv'))
assert np.all(np.array([[0, 1, 2], [1, 3, 4]]) == tmp.values)
tmp = ds.pandas
assert hasattr(tmp, 'test')
tmp = tmp.test
assert isinstance(tmp, pd.DataFrame)
assert np.all(np.array([[1, 2], [3, 4]]) == tmp.values)
tmp = ds.pandas(str(path / 'test.pd.xz'))
assert np.all(np.array([[1, 2], [3, 4]]) == tmp.values)
tmp = ds.pickle
assert hasattr(tmp, 'test')
tmp = tmp.test
assert isinstance(tmp, list)
assert [[1, 2], [3, 4]] == tmp
tmp = ds.pickle(str(path / 'test.pkl.z'))
assert [[1, 2], [3, 4]] == tmp
def test_dateset_4(test_data):
file_path = test_data[2]
data = pd.DataFrame([[1, 2], [3, 4]])
file = file_path / 'test.pd'
Dataset.to(data, file)
assert file.exists()
file = file_path / 'test.str'
Dataset.to(data, str(file))
assert file.exists()
file = file_path / 'test.pkl'
Dataset.to(data.values, file)
assert file.exists()