def test_median_operand_not_iterable(self): """ Tests return of a TypeError if passing non-datetime.datetime objects """ x1 = 1 with self.assertRaises(TypeError): dts.median(x1)
def test_median_empty(self): """ Tests return of IndexError when passing a zero-length list """ n1 = list() with self.assertRaises(IndexError): dts.median(n1)
def test_median_invalid_operant_object(self): """ Tests return of a TypeError if passing non-datetime.datetime objects """ x1 = ['not a datetime.datetime'] with self.assertRaises(TypeError): dts.median(x1)
def test_median_tz_1(self): """ Tests that a single-length list of a datetime with timezone should return itself """ noon_zulu = dt.datetime(2014, 1, 1, 12, 0, 0, tzinfo=pytz.utc) t1 = [noon_zulu] self.assertEquals(dts.median(t1), noon_zulu)
def test_median_naive_1(self): """ Tests that a single-length list of a datetime should return itself """ naive_1 = dt.datetime(2015, 9, 10, 12, 0, 0) n1 = [naive_1] self.assertEquals(dts.median(n1), naive_1)
def test_median_same_tz_3_vals(self): """ Tests that median of 1200 UTC, 1300 UTC, and 1315 UTC is 1300 UTC. """ z_1200 = dt.datetime(2014, 1, 1, 12, 0, 0, tzinfo=pytz.utc) z_1300 = dt.datetime(2014, 1, 1, 13, 0, 0, tzinfo=pytz.utc) z_1315 = dt.datetime(2014, 1, 1, 13, 15, 0, tzinfo=pytz.utc) t3 = [z_1300, z_1315, z_1200] self.assertEquals(dts.median(t3), dt.datetime(2014, 1, 1, 13, 0, 0, tzinfo=pytz.utc))
def test_median_tz_2_same(self): """ Tests short-circuit that list of twp duplicates should just return the first element. This is faster than sorting and averaging two identical items. """ noon_zulu = dt.datetime(2014, 1, 1, 12, 0, 0, tzinfo=pytz.utc) noon_zulu_ditto = dt.datetime(2014, 1, 1, 12, 0, 0, tzinfo=pytz.utc) t2 = [noon_zulu, noon_zulu_ditto] self.assertEquals(dts.median(t2), noon_zulu)
def test_median_naive_2_same(self): """ Tests short-circuit that list of twp duplicates should just return the first element. Faster than sorting and averaging two identical items. """ naive_1 = dt.datetime(2015, 9, 10, 12, 0, 0) naive_2 = dt.datetime(2015, 9, 10, 12, 0, 0) n2 = [naive_1, naive_2] self.assertEquals(dts.median(n2), naive_1)
def test_median_diff_tz_3_vals(self):
"""
Tests median for three values across different timezones.
Note: This bypasses the PYTZ non-full unit issue as it just sorts the
odd number of timezones and returns the middle (London)
"""
nyc = dt.datetime(2014, 1, 1, 12, 0, 0, tzinfo=pytz.timezone('America/New_York'))
london = dt.datetime(2014, 1, 1, 12, 0, 0, tzinfo=pytz.timezone('Europe/London'))
singapore = dt.datetime(2014, 1, 1, 12, 0, 0, tzinfo=pytz.timezone('Asia/Singapore'))
t3 = [nyc, london, singapore]
self.assertEquals(dts.median(t3), london)
def test_median_naive_3(self): """ Tests that the median of 2015-09-10 12:00:00.00, 2015-09-30 12:00:00.00, and 2015-09-22 12:00.00.000 (all naive) is 2015-09-22 12:00.00.000. This is the middle value """ naive_1 = dt.datetime(2015, 9, 10, 12, 0, 0) naive_2 = dt.datetime(2015, 9, 30, 12, 0, 0) naive_3 = dt.datetime(2015, 9, 22, 12, 0, 0) n3 = [naive_1, naive_2, naive_3] self.assertEquals(dts.median(n3), naive_3)
def test_median_diff_tz_4_vals(self):
"""
Tests median for four values across different timezones.
Note: This sorts than computes the arithmetic mean of the middle two values.
As mentioned above, pytz does not always case to full timezone offsets
"""
chicago = dt.datetime(2014, 1, 1, 12, 0, 0, tzinfo=pytz.timezone('America/Chicago'))
nyc = dt.datetime(2014, 1, 1, 12, 0, 0, tzinfo=pytz.timezone('America/New_York'))
london = dt.datetime(2014, 1, 1, 12, 0, 0, tzinfo=pytz.timezone('Europe/London'))
singapore = dt.datetime(2014, 1, 1, 12, 0, 0, tzinfo=pytz.timezone('Asia/Singapore'))
t4 = [chicago, nyc, london, singapore]
self.assertEquals(dts.median(t4), dts.mean([nyc, london]))
def test_median_naive_4_diff(self): """ Tests that the median of 2015-09-10 12:30:00.00 and 2015-09-10 12:00:00.00 (both naive) is 2015-09-10 12:15:00.00 (arithmetic mean of middle two. Also tests length==2 condition as this is a sub-set of this test. """ naive_1 = dt.datetime(2015, 9, 10, 12, 30, 0) naive_2 = dt.datetime(2015, 9, 10, 12, 0, 0) naive_3 = dt.datetime(2014, 9, 10, 12, 30, 0) naive_4 = dt.datetime(2016, 9, 10, 12, 30, 0) n4 = [naive_1, naive_2, naive_3, naive_4] self.assertEquals(dts.median(n4), dts.mean([naive_1, naive_2]))
