def test_matcher_match_preconditions(self):
mtcher = datetime_glob.Matcher('/some/./path/*%Y-%m-%dT%H-%M-%SZ.jpg')
# yapf: disable
table = [
('/', ValueError),
('', ValueError),
('/some/../path', ValueError),
('some/relative/path', ValueError),
('/some//path', None),
('/some/./path', None)
]
# yapf: enable
for path, expected_exception in table:
if expected_exception is not None:
with self.assertRaises(expected_exception):
_ = mtcher.match(path=path)
with self.assertRaises(expected_exception):
_ = mtcher.match(path=pathlib.Path(path))
else:
_ = mtcher.match(path=path)
_ = mtcher.match(path=pathlib.Path(path))
relative_mtcher = datetime_glob.Matcher(
'some/./path/*%Y-%m-%dT%H-%M-%SZ.jpg')
with self.assertRaises(ValueError):
relative_mtcher.match('/some/absolute/path')
def test_matcher(self):
relative_pattern = 'some/./path/*%Y-%m-%dT%H-%M-%SZ.jpg'
# yapf: disable
table = [
('some/path/ooo2016-07-03T21-22-23Z.jpg', datetime_glob.Match(2016, 7, 3, 21, 22, 23)),
('some/path/2016-07-03T21-22-23Z.jpg', datetime_glob.Match(2016, 7, 3, 21, 22, 23)),
('some/./path/2016-07-03T21-22-23Z.jpg', datetime_glob.Match(2016, 7, 3, 21, 22, 23)),
('some//path/2016-07-03T21-22-23Z.jpg', datetime_glob.Match(2016, 7, 3, 21, 22, 23)),
('some//path/2016-07-03T21-22-23Z.jpg', datetime_glob.Match(2016, 7, 3, 21, 22, 23)),
('some/other/path/2016-07-03T21-22-23Z.jpg', None),
('some//path/2016-07-03', None)
]
# yapf: enable
for relative_path, expected_match in table:
# test both absolute and relative paths
abs_matcher = datetime_glob.Matcher(
pattern='/{}'.format(relative_pattern))
abs_pth = '/{}'.format(relative_path)
self.assertTrue(
match_equal(match=abs_matcher.match(path=abs_pth),
other=expected_match))
rel_matcher = datetime_glob.Matcher(pattern=relative_pattern)
self.assertTrue(
match_equal(match=rel_matcher.match(path=relative_path),
other=expected_match))
def test_match_conversion(self) -> None:
mtcher = datetime_glob.Matcher(
pattern='/some/path/%Y-%m-%dT%H-%M-%S.%fZ.jpg')
mtch = mtcher.match(path='/some/path/2016-12-02T03-04-05.123456Z.jpg')
self.assertIsNotNone(mtch)
assert mtch is not None # needed for mypy
self.assertEqual(mtch.as_datetime(),
datetime.datetime(2016, 12, 2, 3, 4, 5, 123456))
self.assertEqual(mtch.as_maybe_datetime(),
datetime.datetime(2016, 12, 2, 3, 4, 5, 123456))
self.assertEqual(mtch.as_date(), datetime.date(2016, 12, 2))
self.assertEqual(mtch.as_maybe_date(), datetime.date(2016, 12, 2))
self.assertEqual(mtch.as_time(), datetime.time(3, 4, 5, 123456))
# match time without date
mtcher = datetime_glob.Matcher(pattern='/some/path/%H-%M-%S.jpg')
mtch = mtcher.match(path='/some/path/03-04-05.jpg')
self.assertIsNotNone(mtch)
assert mtch is not None # needed for mypy
with self.assertRaises(ValueError):
_ = mtch.as_datetime()
_ = mtch.as_date()
self.assertIsNone(mtch.as_maybe_datetime())
self.assertIsNone(mtch.as_maybe_date())
self.assertEqual(mtch.as_time(), datetime.time(3, 4, 5, 0))
def test_sort_listdir(self) -> None:
with tempfile.TemporaryDirectory() as tempdir:
pth = pathlib.Path(tempdir)
(pth / 'some-description-20.3.2016.txt').write_text('tested')
(pth / 'other-description-7.4.2016.txt').write_text('tested')
(pth / 'yet-another-description-1.1.2016.txt').write_text('tested')
matcher = datetime_glob.Matcher(pattern='*%-d.%-m.%Y.txt')
subpths_matches = [(subpth, matcher.match(subpth.name))
for subpth in pth.iterdir()]
dtimes_subpths = [
] # type: List[Tuple[datetime.datetime, pathlib.Path]]
for subpth, mtch in subpths_matches:
self.assertIsNotNone(subpth)
assert subpth is not None # needed for mypy
self.assertIsNotNone(mtch)
assert mtch is not None # needed for mypy
dtimes_subpths.append((mtch.as_datetime(), subpth))
subpths = [subpth for _, subpth in sorted(dtimes_subpths)]
# yapf: disable
expected = [
pth / 'yet-another-description-1.1.2016.txt',
pth / 'some-description-20.3.2016.txt',
pth / 'other-description-7.4.2016.txt'
]
# yapf: enable
self.assertListEqual(subpths, expected)
def test_matcher_preconditions(self):
# yapf: disable
table = [
('/', ValueError),
('', ValueError),
('/some/../path', ValueError),
('/some//path/*', None),
('/some/./path/*', None)
]
# yapf: enable
for pattern, expected_exception in table:
if expected_exception is not None:
with self.assertRaises(expected_exception):
_ = datetime_glob.Matcher(pattern=pattern)
else:
_ = datetime_glob.Matcher(pattern=pattern)
def test_match_multiple_definition(self):
mtcher = datetime_glob.Matcher(
pattern='/some/path/%Y-%m-%d/%Y-%m-%dT%H-%M-%S.%fZ.jpg')
mtch = mtcher.match(
path='/some/path/2016-12-02/2016-12-02T03-04-05.123456Z.jpg')
self.assertEqual(mtch.as_datetime(),
datetime.datetime(2016, 12, 2, 3, 4, 5, 123456))
mtch = mtcher.match(
path='/some/path/2017-04-12/2016-12-02T03-04-05.123456Z.jpg')
self.assertIsNone(mtch)
def test_sort_listdir(self):
with tempfile.TemporaryDirectory() as tempdir:
pth = pathlib.Path(tempdir)
(pth / 'some-description-20.3.2016.txt').write_text('tested')
(pth / 'other-description-7.4.2016.txt').write_text('tested')
(pth / 'yet-another-description-1.1.2016.txt').write_text('tested')
matcher = datetime_glob.Matcher(pattern='*%-d.%-m.%Y.txt')
subpths_matches = [(subpth, matcher.match(subpth.name))
for subpth in pth.iterdir()]
dtimes_subpths = [(mtch.as_datetime(), subpth)
for subpth, mtch in subpths_matches]
subpths = [subpth for _, subpth in sorted(dtimes_subpths)]
# yapf: disable
expected = [
pth / 'yet-another-description-1.1.2016.txt',
pth / 'some-description-20.3.2016.txt',
pth / 'other-description-7.4.2016.txt'
]
# yapf: enable
self.assertListEqual(subpths, expected)
def __init__(self, db):
self.date_matcher = datetime_glob.Matcher(
pattern='%Y-%m-%dT%H:%M:%S-*')
self.db = db
self.helper = ScraperUtil()
