Skip to content

Commit c68180e

Browse files
jrebackjreback
committed
more tests
1 parent 94d340f commit c68180e

File tree

1 file changed

+94
-64
lines changed

1 file changed

+94
-64
lines changed

‎pandas/tests/test_algos.py

+94-64
Original file line numberDiff line numberDiff line change
@@ -6,7 +6,8 @@
66
from numpy import nan
77
from datetime import datetime
88
from itertools import permutations
9-
from pandas import Series, Categorical, CategoricalIndex, Index
9+
from pandas import (Series, Categorical, CategoricalIndex, Index,
10+
Timestamp, DatetimeIndex)
1011
import pandas as pd
1112

1213
from pandas import compat
@@ -34,7 +35,7 @@ def test_ints(self):
3435
expected = Series(np.array([0, 2, 1, 1, 0, 2, np.nan, 0]))
3536
tm.assert_series_equal(result, expected)
3637

37-
s = pd.Series(np.arange(5), dtype=np.float32)
38+
s = Series(np.arange(5), dtype=np.float32)
3839
result = algos.match(s, [2, 4])
3940
expected = np.array([-1, -1, 0, -1, 1], dtype=np.int64)
4041
self.assert_numpy_array_equal(result, expected)
@@ -204,20 +205,20 @@ def test_mixed(self):
204205
def test_datelike(self):
205206

206207
# M8
207-
v1 = pd.Timestamp('20130101 09:00:00.00004')
208-
v2 = pd.Timestamp('20130101')
208+
v1 = Timestamp('20130101 09:00:00.00004')
209+
v2 = Timestamp('20130101')
209210
x = Series([v1, v1, v1, v2, v2, v1])
210211
labels, uniques = algos.factorize(x)
211212

212213
exp = np.array([0, 0, 0, 1, 1, 0], dtype=np.intp)
213214
self.assert_numpy_array_equal(labels, exp)
214-
exp = pd.DatetimeIndex([v1, v2])
215+
exp = DatetimeIndex([v1, v2])
215216
self.assert_index_equal(uniques, exp)
216217

217218
labels, uniques = algos.factorize(x, sort=True)
218219
exp = np.array([1, 1, 1, 0, 0, 1], dtype=np.intp)
219220
self.assert_numpy_array_equal(labels, exp)
220-
exp = pd.DatetimeIndex([v2, v1])
221+
exp = DatetimeIndex([v2, v1])
221222
self.assert_index_equal(uniques, exp)
222223

223224
# period
@@ -350,7 +351,7 @@ def test_datetime64_dtype_array_returned(self):
350351
tm.assert_numpy_array_equal(result, expected)
351352
self.assertEqual(result.dtype, expected.dtype)
352353

353-
s = pd.Series(dt_index)
354+
s = Series(dt_index)
354355
result = algos.unique(s)
355356
tm.assert_numpy_array_equal(result, expected)
356357
self.assertEqual(result.dtype, expected.dtype)
@@ -369,7 +370,7 @@ def test_timedelta64_dtype_array_returned(self):
369370
tm.assert_numpy_array_equal(result, expected)
370371
self.assertEqual(result.dtype, expected.dtype)
371372

372-
s = pd.Series(td_index)
373+
s = Series(td_index)
373374
result = algos.unique(s)
374375
tm.assert_numpy_array_equal(result, expected)
375376
self.assertEqual(result.dtype, expected.dtype)
@@ -380,11 +381,12 @@ def test_timedelta64_dtype_array_returned(self):
380381
self.assertEqual(result.dtype, expected.dtype)
381382

382383
def test_uint64_overflow(self):
383-
s = pd.Series([1, 2, 2**63, 2**63], dtype=np.uint64)
384+
s = Series([1, 2, 2**63, 2**63], dtype=np.uint64)
384385
exp = np.array([1, 2, 2**63], dtype=np.uint64)
385386
tm.assert_numpy_array_equal(algos.unique(s), exp)
386387

387388
def test_categorical(self):
389+
# GH 15939
388390
c = pd.Categorical(list('aabc'))
389391
result = c.unique()
390392
expected = pd.Categorical(list('abc'))
@@ -397,28 +399,56 @@ def test_categorical(self):
397399
expected = Series(expected, name='foo')
398400
tm.assert_series_equal(result, expected)
399401

402+
def test_datetime64tz_aware(self):
403+
# GH 15939
404+
result = Series(
405+
pd.Index([Timestamp('20160101', tz='US/Eastern'),
406+
Timestamp('20160101', tz='US/Eastern')])).unique()
407+
expected = np.array([Timestamp('2016-01-01 00:00:00-0500',
408+
tz='US/Eastern')], dtype=object)
409+
tm.assert_numpy_array_equal(result, expected)
410+
411+
result = pd.Index([Timestamp('20160101', tz='US/Eastern'),
412+
Timestamp('20160101', tz='US/Eastern')]).unique()
413+
expected = DatetimeIndex(['2016-01-01 00:00:00'],
414+
dtype='datetime64[ns, US/Eastern]', freq=None)
415+
tm.assert_index_equal(result, expected)
416+
417+
result = pd.unique(
418+
Series(pd.Index([Timestamp('20160101', tz='US/Eastern'),
419+
Timestamp('20160101', tz='US/Eastern')])))
420+
expected = DatetimeIndex(['2016-01-01 00:00:00'],
421+
dtype='datetime64[ns, US/Eastern]', freq=None)
422+
tm.assert_index_equal(result, expected)
423+
424+
result = pd.unique(pd.Index([Timestamp('20160101', tz='US/Eastern'),
425+
Timestamp('20160101', tz='US/Eastern')]))
426+
expected = DatetimeIndex(['2016-01-01 00:00:00'],
427+
dtype='datetime64[ns, US/Eastern]', freq=None)
428+
tm.assert_index_equal(result, expected)
429+
400430
def test_order_of_appearance(self):
401431
# 9346
402432
# light testing of guarantee of order of appearance
403433
# these also are the doc-examples
404-
result = pd.unique(pd.Series([2, 1, 3, 3]))
434+
result = pd.unique(Series([2, 1, 3, 3]))
405435
tm.assert_numpy_array_equal(result, np.array([2, 1, 3]))
406436

407-
result = pd.unique(pd.Series([2] + [1] * 5))
437+
result = pd.unique(Series([2] + [1] * 5))
408438
tm.assert_numpy_array_equal(result, np.array([2, 1]))
409439

410-
result = pd.unique(Series([pd.Timestamp('20160101'),
411-
pd.Timestamp('20160101')]))
440+
result = pd.unique(Series([Timestamp('20160101'),
441+
Timestamp('20160101')]))
412442
expected = np.array(['2016-01-01T00:00:00.000000000'],
413443
dtype='datetime64[ns]')
414444
tm.assert_numpy_array_equal(result, expected)
415445

416446
result = pd.unique(pd.Index(
417-
[pd.Timestamp('20160101', tz='US/Eastern'),
418-
pd.Timestamp('20160101', tz='US/Eastern')]))
419-
expected = pd.DatetimeIndex(['2016-01-01 00:00:00'],
420-
dtype='datetime64[ns, US/Eastern]',
421-
freq=None)
447+
[Timestamp('20160101', tz='US/Eastern'),
448+
Timestamp('20160101', tz='US/Eastern')]))
449+
expected = DatetimeIndex(['2016-01-01 00:00:00'],
450+
dtype='datetime64[ns, US/Eastern]',
451+
freq=None)
422452
tm.assert_index_equal(result, expected)
423453

424454
result = pd.unique(list('aabc'))
@@ -448,27 +478,27 @@ def test_basic(self):
448478
expected = np.array([True, False])
449479
tm.assert_numpy_array_equal(result, expected)
450480

451-
result = algos.isin(pd.Series([1, 2]), [1])
481+
result = algos.isin(Series([1, 2]), [1])
452482
expected = np.array([True, False])
453483
tm.assert_numpy_array_equal(result, expected)
454484

455-
result = algos.isin(pd.Series([1, 2]), pd.Series([1]))
485+
result = algos.isin(Series([1, 2]), Series([1]))
456486
expected = np.array([True, False])
457487
tm.assert_numpy_array_equal(result, expected)
458488

459-
result = algos.isin(pd.Series([1, 2]), set([1]))
489+
result = algos.isin(Series([1, 2]), set([1]))
460490
expected = np.array([True, False])
461491
tm.assert_numpy_array_equal(result, expected)
462492

463493
result = algos.isin(['a', 'b'], ['a'])
464494
expected = np.array([True, False])
465495
tm.assert_numpy_array_equal(result, expected)
466496

467-
result = algos.isin(pd.Series(['a', 'b']), pd.Series(['a']))
497+
result = algos.isin(Series(['a', 'b']), Series(['a']))
468498
expected = np.array([True, False])
469499
tm.assert_numpy_array_equal(result, expected)
470500

471-
result = algos.isin(pd.Series(['a', 'b']), set(['a']))
501+
result = algos.isin(Series(['a', 'b']), set(['a']))
472502
expected = np.array([True, False])
473503
tm.assert_numpy_array_equal(result, expected)
474504

@@ -565,33 +595,33 @@ def test_value_counts_nat(self):
565595
self.assertEqual(len(vc), 1)
566596
self.assertEqual(len(vc_with_na), 2)
567597

568-
exp_dt = pd.Series({pd.Timestamp('2014-01-01 00:00:00'): 1})
598+
exp_dt = Series({Timestamp('2014-01-01 00:00:00'): 1})
569599
tm.assert_series_equal(algos.value_counts(dt), exp_dt)
570600
# TODO same for (timedelta)
571601

572602
def test_value_counts_datetime_outofbounds(self):
573603
# GH 13663
574-
s = pd.Series([datetime(3000, 1, 1), datetime(5000, 1, 1),
575-
datetime(5000, 1, 1), datetime(6000, 1, 1),
576-
datetime(3000, 1, 1), datetime(3000, 1, 1)])
604+
s = Series([datetime(3000, 1, 1), datetime(5000, 1, 1),
605+
datetime(5000, 1, 1), datetime(6000, 1, 1),
606+
datetime(3000, 1, 1), datetime(3000, 1, 1)])
577607
res = s.value_counts()
578608

579609
exp_index = pd.Index([datetime(3000, 1, 1), datetime(5000, 1, 1),
580610
datetime(6000, 1, 1)], dtype=object)
581-
exp = pd.Series([3, 2, 1], index=exp_index)
611+
exp = Series([3, 2, 1], index=exp_index)
582612
tm.assert_series_equal(res, exp)
583613

584614
# GH 12424
585-
res = pd.to_datetime(pd.Series(['2362-01-01', np.nan]),
615+
res = pd.to_datetime(Series(['2362-01-01', np.nan]),
586616
errors='ignore')
587-
exp = pd.Series(['2362-01-01', np.nan], dtype=object)
617+
exp = Series(['2362-01-01', np.nan], dtype=object)
588618
tm.assert_series_equal(res, exp)
589619

590620
def test_categorical(self):
591621
s = Series(pd.Categorical(list('aaabbc')))
592622
result = s.value_counts()
593-
expected = pd.Series([3, 2, 1],
594-
index=pd.CategoricalIndex(['a', 'b', 'c']))
623+
expected = Series([3, 2, 1],
624+
index=pd.CategoricalIndex(['a', 'b', 'c']))
595625
tm.assert_series_equal(result, expected, check_index_type=True)
596626

597627
# preserve order?
@@ -604,11 +634,11 @@ def test_categorical_nans(self):
604634
s = Series(pd.Categorical(list('aaaaabbbcc'))) # 4,3,2,1 (nan)
605635
s.iloc[1] = np.nan
606636
result = s.value_counts()
607-
expected = pd.Series([4, 3, 2], index=pd.CategoricalIndex(
637+
expected = Series([4, 3, 2], index=pd.CategoricalIndex(
608638
['a', 'b', 'c'], categories=['a', 'b', 'c']))
609639
tm.assert_series_equal(result, expected, check_index_type=True)
610640
result = s.value_counts(dropna=False)
611-
expected = pd.Series([
641+
expected = Series([
612642
4, 3, 2, 1
613643
], index=pd.CategoricalIndex(['a', 'b', 'c', np.nan]))
614644
tm.assert_series_equal(result, expected, check_index_type=True)
@@ -618,12 +648,12 @@ def test_categorical_nans(self):
618648
list('aaaaabbbcc'), ordered=True, categories=['b', 'a', 'c']))
619649
s.iloc[1] = np.nan
620650
result = s.value_counts()
621-
expected = pd.Series([4, 3, 2], index=pd.CategoricalIndex(
651+
expected = Series([4, 3, 2], index=pd.CategoricalIndex(
622652
['a', 'b', 'c'], categories=['b', 'a', 'c'], ordered=True))
623653
tm.assert_series_equal(result, expected, check_index_type=True)
624654

625655
result = s.value_counts(dropna=False)
626-
expected = pd.Series([4, 3, 2, 1], index=pd.CategoricalIndex(
656+
expected = Series([4, 3, 2, 1], index=pd.CategoricalIndex(
627657
['a', 'b', 'c', np.nan], categories=['b', 'a', 'c'], ordered=True))
628658
tm.assert_series_equal(result, expected, check_index_type=True)
629659

@@ -640,33 +670,33 @@ def test_dropna(self):
640670
# https://github.com/pandas-dev/pandas/issues/9443#issuecomment-73719328
641671

642672
tm.assert_series_equal(
643-
pd.Series([True, True, False]).value_counts(dropna=True),
644-
pd.Series([2, 1], index=[True, False]))
673+
Series([True, True, False]).value_counts(dropna=True),
674+
Series([2, 1], index=[True, False]))
645675
tm.assert_series_equal(
646-
pd.Series([True, True, False]).value_counts(dropna=False),
647-
pd.Series([2, 1], index=[True, False]))
676+
Series([True, True, False]).value_counts(dropna=False),
677+
Series([2, 1], index=[True, False]))
648678

649679
tm.assert_series_equal(
650-
pd.Series([True, True, False, None]).value_counts(dropna=True),
651-
pd.Series([2, 1], index=[True, False]))
680+
Series([True, True, False, None]).value_counts(dropna=True),
681+
Series([2, 1], index=[True, False]))
652682
tm.assert_series_equal(
653-
pd.Series([True, True, False, None]).value_counts(dropna=False),
654-
pd.Series([2, 1, 1], index=[True, False, np.nan]))
683+
Series([True, True, False, None]).value_counts(dropna=False),
684+
Series([2, 1, 1], index=[True, False, np.nan]))
655685
tm.assert_series_equal(
656-
pd.Series([10.3, 5., 5.]).value_counts(dropna=True),
657-
pd.Series([2, 1], index=[5., 10.3]))
686+
Series([10.3, 5., 5.]).value_counts(dropna=True),
687+
Series([2, 1], index=[5., 10.3]))
658688
tm.assert_series_equal(
659-
pd.Series([10.3, 5., 5.]).value_counts(dropna=False),
660-
pd.Series([2, 1], index=[5., 10.3]))
689+
Series([10.3, 5., 5.]).value_counts(dropna=False),
690+
Series([2, 1], index=[5., 10.3]))
661691

662692
tm.assert_series_equal(
663-
pd.Series([10.3, 5., 5., None]).value_counts(dropna=True),
664-
pd.Series([2, 1], index=[5., 10.3]))
693+
Series([10.3, 5., 5., None]).value_counts(dropna=True),
694+
Series([2, 1], index=[5., 10.3]))
665695

666696
# 32-bit linux has a different ordering
667697
if not compat.is_platform_32bit():
668-
result = pd.Series([10.3, 5., 5., None]).value_counts(dropna=False)
669-
expected = pd.Series([2, 1, 1], index=[5., 10.3, np.nan])
698+
result = Series([10.3, 5., 5., None]).value_counts(dropna=False)
699+
expected = Series([2, 1, 1], index=[5., 10.3, np.nan])
670700
tm.assert_series_equal(result, expected)
671701

672702
def test_value_counts_normalized(self):
@@ -781,15 +811,15 @@ def test_numeric_object_likes(self):
781811
tm.assert_numpy_array_equal(res_false, exp_false)
782812

783813
# series
784-
for s in [pd.Series(case), pd.Series(case, dtype='category')]:
814+
for s in [Series(case), Series(case, dtype='category')]:
785815
res_first = s.duplicated(keep='first')
786-
tm.assert_series_equal(res_first, pd.Series(exp_first))
816+
tm.assert_series_equal(res_first, Series(exp_first))
787817

788818
res_last = s.duplicated(keep='last')
789-
tm.assert_series_equal(res_last, pd.Series(exp_last))
819+
tm.assert_series_equal(res_last, Series(exp_last))
790820

791821
res_false = s.duplicated(keep=False)
792-
tm.assert_series_equal(res_false, pd.Series(exp_false))
822+
tm.assert_series_equal(res_false, Series(exp_false))
793823

794824
def test_datetime_likes(self):
795825

@@ -798,8 +828,8 @@ def test_datetime_likes(self):
798828
td = ['1 days', '2 days', '1 days', 'NaT', '3 days',
799829
'2 days', '4 days', '1 days', 'NaT', '6 days']
800830

801-
cases = [np.array([pd.Timestamp(d) for d in dt]),
802-
np.array([pd.Timestamp(d, tz='US/Eastern') for d in dt]),
831+
cases = [np.array([Timestamp(d) for d in dt]),
832+
np.array([Timestamp(d, tz='US/Eastern') for d in dt]),
803833
np.array([pd.Period(d, freq='D') for d in dt]),
804834
np.array([np.datetime64(d) for d in dt]),
805835
np.array([pd.Timedelta(d) for d in td])]
@@ -833,16 +863,16 @@ def test_datetime_likes(self):
833863
tm.assert_numpy_array_equal(res_false, exp_false)
834864

835865
# series
836-
for s in [pd.Series(case), pd.Series(case, dtype='category'),
837-
pd.Series(case, dtype=object)]:
866+
for s in [Series(case), Series(case, dtype='category'),
867+
Series(case, dtype=object)]:
838868
res_first = s.duplicated(keep='first')
839-
tm.assert_series_equal(res_first, pd.Series(exp_first))
869+
tm.assert_series_equal(res_first, Series(exp_first))
840870

841871
res_last = s.duplicated(keep='last')
842-
tm.assert_series_equal(res_last, pd.Series(exp_last))
872+
tm.assert_series_equal(res_last, Series(exp_last))
843873

844874
res_false = s.duplicated(keep=False)
845-
tm.assert_series_equal(res_false, pd.Series(exp_false))
875+
tm.assert_series_equal(res_false, Series(exp_false))
846876

847877
def test_unique_index(self):
848878
cases = [pd.Index([1, 2, 3]), pd.RangeIndex(0, 3)]
@@ -984,7 +1014,7 @@ def test_lookup_overflow(self):
9841014
np.arange(len(xs), dtype=np.int64))
9851015

9861016
def test_get_unique(self):
987-
s = pd.Series([1, 2, 2**63, 2**63], dtype=np.uint64)
1017+
s = Series([1, 2, 2**63, 2**63], dtype=np.uint64)
9881018
exp = np.array([1, 2, 2**63], dtype=np.uint64)
9891019
self.assert_numpy_array_equal(s.unique(), exp)
9901020

0 commit comments

Comments
 (0)