def test_load_csv_ignore_headers(self):
data = CSVBackend(x for x in "./timeseriesql/tests/csv/basic.csv").labels(
[
{"label": "one"},
{"label": "two"},
{"label": "three"},
{"label": "four"},
{"label": "five"},
{"label": "six"},
{"label": "seven"},
]
)[:]
# shape is expected
self.assertEqual((11, 7), data.shape)
# labels
self.assertTrue(
np.array_equal(
data.labels,
[
{"label": "one"},
{"label": "two"},
{"label": "three"},
{"label": "four"},
{"label": "five"},
{"label": "six"},
{"label": "seven"},
],
)
)
def test_load_csv(self):
data = CSVBackend(x for x in "./timeseriesql/tests/csv/basic.csv")[:]
# created new timeseries
self.assertTrue(isinstance(data, TimeSeries))
# shape is expected
self.assertEqual((11, 7), data.shape)
# labels
self.assertTrue(
np.array_equal(
data.labels,
[
{"label": "A"},
{"label": "B"},
{"label": "C"},
{"label": "D"},
{"label": "E"},
{"label": "F"},
{"label": "G"},
],
)
)
# time index is correct
self.assertTrue(np.array_equal(data.time[-2:], [1520640000.0, 1520726400.0]))
def test_skip_bottom_row(self):
data = CSVBackend(x for x in "./timeseriesql/tests/csv/basic_with_missing_data.csv")[:-1]
# created new timeseries
self.assertTrue(isinstance(data, TimeSeries))
# shape is expected
self.assertEqual((10, 7), data.shape)
def test_load_csv_filter_headers(self):
# =, !-, not in, in
# ==
data = CSVBackend(x for x in "./timeseriesql/tests/csv/basic.csv" if x.label == "A")[:]
self.assertEqual((11, 1), data.shape)
self.assertTrue(np.array_equal(data.labels, [{"label": "A"}]))
# !==
data = CSVBackend(x for x in "./timeseriesql/tests/csv/basic.csv" if x.label != "B")[:]
self.assertEqual((11, 6), data.shape)
self.assertTrue(
np.array_equal(
data.labels,
[
{"label": "A"},
{"label": "C"},
{"label": "D"},
{"label": "E"},
{"label": "F"},
{"label": "G"},
],
)
)
# in
data = CSVBackend(
x for x in "./timeseriesql/tests/csv/basic.csv" if x.label in ["B", "C", "G"]
)[:]
self.assertEqual((11, 3), data.shape)
self.assertTrue(
np.array_equal(data.labels, [{"label": "B"}, {"label": "C"}, {"label": "G"}])
)
data = CSVBackend(
x for x in "./timeseriesql/tests/csv/basic.csv" if x.label not in ["B", "C", "G"]
)[:]
self.assertEqual((11, 4), data.shape)
self.assertTrue(
np.array_equal(
data.labels, [{"label": "A"}, {"label": "D"}, {"label": "E"}, {"label": "F"}]
)
)
def test_load_csv_with_empty_columns(self):
data = CSVBackend(x for x in "./timeseriesql/tests/csv/basic_with_missing_data.csv")[:]
# created new timeseries
self.assertTrue(isinstance(data, TimeSeries))
# shape is expected
self.assertEqual((11, 7), data.shape)
# labels
self.assertTrue(
np.array_equal(
data.labels,
[
{"label": "A"},
{"label": "B"},
{"label": "C"},
{"label": "D"},
{"label": "E"},
{"label": "F"},
{"label": "G"},
],
)
)
def test_load_csv_from_url(self):
data = CSVBackend(x for x in "https://raw.githubusercontent.com/mbeale/timeseriesql/master/timeseriesql/tests/csv/basic.csv" if x.label == "A")[:]
self.assertEqual((11, 1), data.shape)
self.assertTrue(np.array_equal(data.labels, [{"label": "A"}]))
def test_load_csv_string_date_v2(self):
data = CSVBackend(x for x in "./timeseriesql/tests/csv/basic_with_string_dates_v2.csv")[:]
tstamps = [1571296500.0 + (i * 60) for i in range(11)]
self.assertTrue(np.array_equal(data.time, tstamps))