def test_iterates_csv_without_header(self):
data = easyfile.CsvFile(self.fp.name, header=False)
data._prepare_reading()
expected = [line.split(',') for line in self.lines]
self.assertSequenceEqual(data, expected)
for x, y in zip(data, expected):
self.assertEqual(x, y)
def test_iterates_csv_with_header(self):
from collections import OrderedDict
data = easyfile.CsvFile(self.fp.name, header=True)
data._prepare_reading()
expected = [OrderedDict(zip(data._filednames, line.split(',')))
for line in self.lines[1:]]
self.assertSequenceEqual(data, expected)
for x, y in zip(data, expected):
self.assertEqual(x, y)
def __init__(self,
path: str,
encoding: str = 'utf-8',
delimiter: str = ',',
header: bool = False) -> None:
super().__init__(
easyfile.CsvFile(path=path,
encoding=encoding,
delimiter=delimiter,
header=header))
def test_dunder_init(self):
data = easyfile.CsvFile(self.fp.name)
self.assertEqual(data._path, self.fp.name)
self.assertEqual(data._encoding, 'utf-8')
self.assertFalse(data._ready)
self.assertIsNone(data._length)
self.assertIsNone(data._offsets)
self.assertIsNone(data._mm)
self.assertEqual(data._delimiter, ',')
self.assertFalse(data._header)
self.assertIsNone(data._filednames)
def easyfile_creator(path):
dataset = {}
archive_path = gdown.cached_download(url)
with tarfile.open(archive_path, 'r') as archive:
print(f'Extracting to {root}...')
archive.extractall(root)
dataset = {}
for split in ('train', 'test'):
filename = f'{key}_csv/{split}.csv'
dataset[split] = easyfile.CsvFile(os.path.join(root, filename))
with io.open(path, 'wb') as f:
pickle.dump(dataset, f)
return dataset
def test_slices_items(self):
data = easyfile.CsvFile(self.fp.name)
self.assertSequenceEqual(data[:len(self.lines)], data)
def test_loads_csv_with_header(self):
data = easyfile.CsvFile(self.fp.name, header=True)
self.assertTrue(data._header)
data._prepare_reading()
self.assertListEqual(data._filednames, self.lines[0].split(','))