-
Notifications
You must be signed in to change notification settings - Fork 4.2k
Expand file tree
/
Copy pathtest_array.py
More file actions
4585 lines (3728 loc) · 150 KB
/
Copy pathtest_array.py
File metadata and controls
4585 lines (3728 loc) · 150 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
# Licensed to the Apache Software Foundation (ASF) under one
# or more contributor license agreements. See the NOTICE file
# distributed with this work for additional information
# regarding copyright ownership. The ASF licenses this file
# to you under the Apache License, Version 2.0 (the
# "License"); you may not use this file except in compliance
# with the License. You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
# KIND, either express or implied. See the License for the
# specific language governing permissions and limitations
# under the License.
from collections.abc import Iterable
import datetime
import decimal
import hypothesis as h
import hypothesis.strategies as st
import itertools
import pytest
import struct
import subprocess
import sys
import weakref
try:
import numpy as np
except ImportError:
np = None
import pyarrow as pa
import pyarrow.tests.strategies as past
from pyarrow.vendored.version import Version
import pyarrow.compute as pc
@pytest.mark.processes
def test_total_bytes_allocated():
code = """if 1:
import pyarrow as pa
assert pa.total_allocated_bytes() == 0
"""
res = subprocess.run([sys.executable, "-c", code],
universal_newlines=True, stderr=subprocess.PIPE)
if res.returncode != 0:
print(res.stderr, file=sys.stderr)
res.check_returncode() # fail
assert len(res.stderr.splitlines()) == 0
def test_weakref():
arr = pa.array([1, 2, 3])
wr = weakref.ref(arr)
assert wr() is not None
del arr
assert wr() is None
def test_getitem_NULL():
arr = pa.array([1, None, 2])
assert arr[1].as_py() is None
assert arr[1].is_valid is False
assert isinstance(arr[1], pa.Int64Scalar)
def test_constructor_raises():
# This could happen by wrong capitalization.
# ARROW-2638: prevent calling extension class constructors directly
with pytest.raises(TypeError):
pa.Array([1, 2])
def test_list_format():
arr = pa.array([["foo"], None, ["bar", "a longer string", None]])
result = arr.to_string()
expected = """\
[
[
"foo"
],
null,
[
"bar",
"a longer string",
null
]
]"""
assert result == expected
result = arr.to_string(element_size_limit=10)
expected = """\
[
[
"foo"
],
null,
[
"bar",
"a longer (... 7 chars omitted)",
null
]
]"""
assert result == expected
def test_string_format():
arr = pa.array(['', None, 'foo'])
result = arr.to_string()
expected = """\
[
"",
null,
"foo"
]"""
assert result == expected
def test_long_array_format():
arr = pa.array(range(100))
result = arr.to_string(window=2)
expected = """\
[
0,
1,
...
98,
99
]"""
assert result == expected
def test_indented_string_format():
arr = pa.array(['', None, 'foo'])
result = arr.to_string(indent=1)
expected = '[\n "",\n null,\n "foo"\n]'
assert result == expected
def test_top_level_indented_string_format():
arr = pa.array(['', None, 'foo'])
result = arr.to_string(top_level_indent=1)
expected = ' [\n "",\n null,\n "foo"\n ]'
assert result == expected
def test_binary_format():
arr = pa.array([b'\x00', b'', None, b'\x01foo', b'\x80\xff'])
result = arr.to_string()
expected = """\
[
00,
,
null,
01666F6F,
80FF
]"""
assert result == expected
def test_binary_total_values_length():
arr = pa.array([b'0000', None, b'11111', b'222222', b'3333333'],
type='binary')
large_arr = pa.array([b'0000', None, b'11111', b'222222', b'3333333'],
type='large_binary')
assert arr.total_values_length == 22
assert arr.slice(1, 3).total_values_length == 11
assert large_arr.total_values_length == 22
assert large_arr.slice(1, 3).total_values_length == 11
@pytest.mark.numpy
def test_to_numpy_zero_copy():
arr = pa.array(range(10))
np_arr = arr.to_numpy()
# check for zero copy (both arrays using same memory)
arrow_buf = arr.buffers()[1]
assert arrow_buf.address == np_arr.ctypes.data
arr = None
import gc
gc.collect()
# Ensure base is still valid
assert np_arr.base is not None
expected = np.arange(10)
np.testing.assert_array_equal(np_arr, expected)
@pytest.mark.numpy
def test_chunked_array_to_numpy_zero_copy():
elements = [[2, 2, 4], [4, 5, 100]]
chunked_arr = pa.chunked_array(elements)
msg = "zero_copy_only must be False for pyarrow.ChunkedArray.to_numpy"
with pytest.raises(ValueError, match=msg):
chunked_arr.to_numpy(zero_copy_only=True)
np_arr = chunked_arr.to_numpy()
expected = [2, 2, 4, 4, 5, 100]
np.testing.assert_array_equal(np_arr, expected)
@pytest.mark.numpy
def test_to_numpy_unsupported_types():
# ARROW-2871: Some primitive types are not yet supported in to_numpy
bool_arr = pa.array([True, False, True])
with pytest.raises(ValueError):
bool_arr.to_numpy()
result = bool_arr.to_numpy(zero_copy_only=False)
expected = np.array([True, False, True])
np.testing.assert_array_equal(result, expected)
null_arr = pa.array([None, None, None])
with pytest.raises(ValueError):
null_arr.to_numpy()
result = null_arr.to_numpy(zero_copy_only=False)
expected = np.array([None, None, None], dtype=object)
np.testing.assert_array_equal(result, expected)
arr = pa.array([1, 2, None])
with pytest.raises(ValueError, match="with 1 nulls"):
arr.to_numpy()
@pytest.mark.numpy
def test_to_numpy_writable():
arr = pa.array(range(10))
np_arr = arr.to_numpy()
# by default not writable for zero-copy conversion
with pytest.raises(ValueError):
np_arr[0] = 10
np_arr2 = arr.to_numpy(zero_copy_only=False, writable=True)
np_arr2[0] = 10
assert arr[0].as_py() == 0
# when asking for writable, cannot do zero-copy
with pytest.raises(ValueError):
arr.to_numpy(zero_copy_only=True, writable=True)
@pytest.mark.numpy
@pytest.mark.parametrize('unit', ['s', 'ms', 'us', 'ns'])
@pytest.mark.parametrize('tz', [None, "UTC"])
def test_to_numpy_datetime64(unit, tz):
arr = pa.array([1, 2, 3], pa.timestamp(unit, tz=tz))
expected = np.array([1, 2, 3], dtype=f"datetime64[{unit}]")
np_arr = arr.to_numpy()
np.testing.assert_array_equal(np_arr, expected)
@pytest.mark.numpy
@pytest.mark.parametrize('unit', ['s', 'ms', 'us', 'ns'])
def test_to_numpy_timedelta64(unit):
arr = pa.array([1, 2, 3], pa.duration(unit))
expected = np.array([1, 2, 3], dtype=f"timedelta64[{unit}]")
np_arr = arr.to_numpy()
np.testing.assert_array_equal(np_arr, expected)
@pytest.mark.numpy
def test_to_numpy_dictionary():
# ARROW-7591
arr = pa.array(["a", "b", "a"]).dictionary_encode()
expected = np.array(["a", "b", "a"], dtype=object)
np_arr = arr.to_numpy(zero_copy_only=False)
np.testing.assert_array_equal(np_arr, expected)
@pytest.mark.pandas
def test_to_pandas_zero_copy():
import gc
arr = pa.array(range(10))
for i in range(10):
series = arr.to_pandas()
# In Python 3.14 interpreter might avoid some
# reference count modifications
assert sys.getrefcount(series) in (1, 2)
series = None # noqa
assert sys.getrefcount(arr) in (1, 2)
for i in range(10):
arr = pa.array(range(10))
series = arr.to_pandas()
arr = None
gc.collect()
# Ensure base is still valid
# Because of py.test's assert inspection magic, if you put getrefcount
# on the line being examined, it will be 1 higher than you expect
base_refcount = sys.getrefcount(series.values.base)
assert base_refcount == 2
series.sum()
@pytest.mark.nopandas
@pytest.mark.pandas
def test_asarray():
# ensure this is tested both when pandas is present or not (ARROW-6564)
arr = pa.array(range(4))
# The iterator interface gives back an array of Int64Value's
np_arr = np.asarray([_ for _ in arr])
assert np_arr.tolist() == [0, 1, 2, 3]
assert np_arr.dtype == np.dtype('O')
assert isinstance(np_arr[0], pa.lib.Int64Value)
# Calling with the arrow array gives back an array with 'int64' dtype
np_arr = np.asarray(arr)
assert np_arr.tolist() == [0, 1, 2, 3]
assert np_arr.dtype == np.dtype('int64')
# An optional type can be specified when calling np.asarray
np_arr = np.asarray(arr, dtype='str')
assert np_arr.tolist() == ['0', '1', '2', '3']
# If PyArrow array has null values, numpy type will be changed as needed
# to support nulls.
arr = pa.array([0, 1, 2, None])
assert arr.type == pa.int64()
np_arr = np.asarray(arr)
elements = np_arr.tolist()
assert elements[:3] == [0., 1., 2.]
assert np.isnan(elements[3])
assert np_arr.dtype == np.dtype('float64')
# DictionaryType data will be converted to dense numpy array
arr = pa.DictionaryArray.from_arrays(
pa.array([0, 1, 2, 0, 1]), pa.array(['a', 'b', 'c']))
np_arr = np.asarray(arr)
assert np_arr.dtype == np.dtype('object')
assert np_arr.tolist() == ['a', 'b', 'c', 'a', 'b']
@pytest.mark.parametrize('ty', [
None,
pa.null(),
pa.int8(),
pa.string()
])
def test_nulls(ty):
arr = pa.nulls(3, type=ty)
expected = pa.array([None, None, None], type=ty)
assert len(arr) == 3
assert arr.equals(expected)
if ty is None:
assert arr.type == pa.null()
else:
assert arr.type == ty
def test_array_from_scalar():
pytz = pytest.importorskip("pytz")
today = datetime.date.today()
now = datetime.datetime.now()
now_utc = now.replace(tzinfo=pytz.utc)
now_with_tz = now_utc.astimezone(pytz.timezone('US/Eastern'))
oneday = datetime.timedelta(days=1)
cases = [
(None, 1, pa.array([None])),
(None, 10, pa.nulls(10)),
(-1, 3, pa.array([-1, -1, -1], type=pa.int64())),
(2.71, 2, pa.array([2.71, 2.71], type=pa.float64())),
("string", 4, pa.array(["string"] * 4)),
(
pa.scalar(8, type=pa.uint8()),
17,
pa.array([8] * 17, type=pa.uint8())
),
(pa.scalar(None), 3, pa.array([None, None, None])),
(pa.scalar(True), 11, pa.array([True] * 11)),
(today, 2, pa.array([today] * 2)),
(now, 10, pa.array([now] * 10)),
(
now_with_tz,
2,
pa.array(
[now_utc] * 2,
type=pa.timestamp('us', tz=pytz.timezone('US/Eastern'))
)
),
(now.time(), 9, pa.array([now.time()] * 9)),
(oneday, 4, pa.array([oneday] * 4)),
(False, 9, pa.array([False] * 9)),
([1, 2], 2, pa.array([[1, 2], [1, 2]])),
(
pa.scalar([-1, 3], type=pa.large_list(pa.int8())),
5,
pa.array([[-1, 3]] * 5, type=pa.large_list(pa.int8()))
),
({'a': 1, 'b': 2}, 3, pa.array([{'a': 1, 'b': 2}] * 3))
]
for value, size, expected in cases:
arr = pa.repeat(value, size)
assert len(arr) == size
assert arr.type.equals(expected.type)
assert arr.equals(expected)
if expected.type == pa.null():
assert arr.null_count == size
else:
assert arr.null_count == 0
def test_array_from_dictionary_scalar():
dictionary = ['foo', 'bar', 'baz']
arr = pa.DictionaryArray.from_arrays([2, 1, 2, 0], dictionary=dictionary)
result = pa.repeat(arr[0], 5)
expected = pa.DictionaryArray.from_arrays([2] * 5, dictionary=dictionary)
assert result.equals(expected)
result = pa.repeat(arr[3], 5)
expected = pa.DictionaryArray.from_arrays([0] * 5, dictionary=dictionary)
assert result.equals(expected)
def test_array_getitem():
arr = pa.array(range(10, 15))
lst = arr.to_pylist()
for idx in range(-len(arr), len(arr)):
assert arr[idx].as_py() == lst[idx]
for idx in range(-2 * len(arr), -len(arr)):
with pytest.raises(IndexError):
arr[idx]
for idx in range(len(arr), 2 * len(arr)):
with pytest.raises(IndexError):
arr[idx]
@pytest.mark.numpy
def test_array_getitem_numpy_scalars():
arr = pa.array(range(10, 15))
lst = arr.to_pylist()
# check that numpy scalars are supported
for idx in range(-len(arr), len(arr)):
assert arr[np.int32(idx)].as_py() == lst[idx]
def test_to_pylist_bulk_paths():
# GH-50326: to_pylist converts through scalar-free _getitem_py
# specializations; the result must match the per-scalar conversion
# exactly.
arrays = [
pa.array([[1, None, 3], None, [], [4]], type=pa.list_(pa.int32())),
pa.array([["a", None], None, [], ["bcd", ""]],
type=pa.list_(pa.string())),
pa.array([["a", None], None, [], ["bcd", ""]],
type=pa.large_list(pa.large_string())),
pa.array([[1, None], None, [3, 4]], type=pa.list_(pa.int32(), 2)),
pa.array([[[1], [2, None]], None, [None, [3]]],
type=pa.list_(pa.list_(pa.int32()))),
pa.array([[("k1", 1), ("k2", None)], None, []],
type=pa.map_(pa.string(), pa.int32())),
pa.array(["a", None, "", "\N{GRINNING FACE} \N{SNOWMAN}"],
type=pa.string()),
pa.array(["a", None, "", "\N{GRINNING FACE} \N{SNOWMAN}"],
type=pa.large_string()),
pa.array([b"a\x00b", None, b"", b"\xff"], type=pa.binary()),
pa.array([b"a\x00b", None, b""], type=pa.large_binary()),
# View types store short values inline and long values out-of-line;
# cover both, plus NUL bytes and non-ASCII data.
pa.array(["a", None, "", "\N{GRINNING FACE} \N{SNOWMAN}",
"long string exceeding the inline view size"],
type=pa.string_view()),
pa.array([b"a\x00b", None, b"", b"\xff",
b"long binary value exceeding the inline view size"],
type=pa.binary_view()),
pa.array([[b"x", None, b"\x00y"], None, []],
type=pa.list_(pa.binary())),
pa.array([1, None, -(2**62), 2**62], type=pa.int64()),
pa.array([0, None, 2**63 + 7], type=pa.uint64()),
pa.array([-128, 127, None], type=pa.int8()),
pa.array([1.5, None, -0.5], type=pa.float64()),
pa.array([1.5, None], type=pa.float32()),
pa.array([True, None, False], type=pa.bool_()),
pa.array([{"a": 1, "b": "x"}, None, {"a": None, "b": None}],
type=pa.struct([("a", pa.int32()), ("b", pa.string())])),
pa.array([], type=pa.list_(pa.int32())),
pa.array([None, None], type=pa.list_(pa.string())),
]
for arr in arrays:
for view in (arr, arr.slice(1), arr.slice(0, 2), arr.slice(2)):
assert view.to_pylist() == [x.as_py() for x in view]
# Values inside numeric lists must stay Python ints/None, never floats
result = pa.array([[1, None, 3]], type=pa.list_(pa.int32())).to_pylist()
assert result == [[1, None, 3]]
assert [type(x) for x in result[0]] == [int, type(None), int]
# Duplicate struct field names raise like StructScalar.as_py does
dup = pa.StructArray.from_arrays(
[pa.array([1, 2]), pa.array(["a", "b"])], names=["x", "x"])
with pytest.raises(ValueError, match="duplicate field names"):
dup.to_pylist()
def test_array_slice():
arr = pa.array(range(10))
sliced = arr.slice(2)
expected = pa.array(range(2, 10))
assert sliced.equals(expected)
sliced2 = arr.slice(2, 4)
expected2 = pa.array(range(2, 6))
assert sliced2.equals(expected2)
# 0 offset
assert arr.slice(0).equals(arr)
# Slice past end of array
assert len(arr.slice(len(arr))) == 0
assert len(arr.slice(len(arr) + 2)) == 0
assert len(arr.slice(len(arr) + 2, 100)) == 0
with pytest.raises(IndexError):
arr.slice(-1)
with pytest.raises(ValueError):
arr.slice(2, -1)
# Test slice notation
assert arr[2:].equals(arr.slice(2))
assert arr[2:5].equals(arr.slice(2, 3))
assert arr[-5:].equals(arr.slice(len(arr) - 5))
n = len(arr)
for start in range(-n * 2, n * 2):
for stop in range(-n * 2, n * 2):
res = arr[start:stop]
res.validate()
expected = arr.to_pylist()[start:stop]
assert res.to_pylist() == expected
if np is not None:
assert res.to_numpy().tolist() == expected
def test_array_slice_negative_step():
# ARROW-2714
values = list(range(20))
arr = pa.array(values)
chunked_arr = pa.chunked_array([arr])
cases = [
slice(None, None, -1), # GH-46606
slice(None, 6, -2),
slice(10, 6, -2),
slice(8, None, -2),
slice(2, 10, -2),
slice(10, 2, -2),
slice(None, None, 2),
slice(0, 10, 2),
slice(15, -25, -1), # GH-38768
slice(-22, -22, -1), # GH-40642
]
for case in cases:
result = arr[case]
expected = pa.array(values[case], type=arr.type)
assert result.equals(expected)
result = pa.record_batch([arr], names=['f0'])[case]
expected = pa.record_batch([expected], names=['f0'])
assert result.equals(expected)
result = chunked_arr[case]
expected = pa.chunked_array([values[case]], type=arr.type)
assert result.equals(expected)
def test_arange():
cases = [
(5, 103), # Default step
(-2, 128, 3),
(4, 103, 5),
(10, -7, -1),
(100, -20, -3),
(0, 0), # Empty array
(2, 10, -1), # Empty array
(10, 3, 1), # Empty array
]
for case in cases:
result = pa.arange(*case)
result.validate(full=True)
assert result.equals(pa.array(list(range(*case)), type=pa.int64()))
# Validate memory_pool keyword argument
result = pa.arange(-1, 101, memory_pool=pa.default_memory_pool())
result.validate(full=True)
assert result.equals(pa.array(list(range(-1, 101)), type=pa.int64()))
# Special case for invalid step (arange does not accept step of 0)
with pytest.raises(pa.ArrowInvalid):
pa.arange(0, 10, 0)
def test_array_diff():
# ARROW-6252
arr1 = pa.array(['foo'], type=pa.utf8())
arr2 = pa.array(['foo', 'bar', None], type=pa.utf8())
arr3 = pa.array([1, 2, 3])
arr4 = pa.array([[], [1], None], type=pa.list_(pa.int64()))
arr5 = pa.array([1.5, 3, 6], type=pa.float16())
arr6 = pa.array([1, 3], type=pa.float16())
assert arr1.diff(arr1) == ''
assert arr1.diff(arr2) == '''
@@ -1, +1 @@
+"bar"
+null
'''
assert arr1.diff(arr3).strip() == '# Array types differed: string vs int64'
assert arr1.diff(arr3).strip() == '# Array types differed: string vs int64'
assert arr1.diff(arr4).strip() == ('# Array types differed: string vs '
'list<item: int64>')
assert arr5.diff(arr5) == ''
assert arr5.diff(arr6) == '''
@@ -0, +0 @@
-1.5
+1
@@ -2, +2 @@
-6
'''
def test_array_iter():
arr = pa.array(range(10))
for i, j in zip(range(10), arr):
assert i == j.as_py()
assert isinstance(arr, Iterable)
def test_struct_array_slice():
# ARROW-2311: slicing nested arrays needs special care
ty = pa.struct([pa.field('a', pa.int8()),
pa.field('b', pa.float32())])
arr = pa.array([(1, 2.5), (3, 4.5), (5, 6.5)], type=ty)
assert arr[1:].to_pylist() == [{'a': 3, 'b': 4.5},
{'a': 5, 'b': 6.5}]
@pytest.mark.numpy
def test_array_factory_invalid_type():
class MyObject:
pass
arr = np.array([MyObject()])
with pytest.raises(ValueError):
pa.array(arr)
@pytest.mark.numpy
def test_array_ref_to_ndarray_base():
arr = np.array([1, 2, 3])
refcount = sys.getrefcount(arr)
arr2 = pa.array(arr) # noqa
assert sys.getrefcount(arr) == (refcount + 1)
def test_array_eq():
# ARROW-2150 / ARROW-9445: we define the __eq__ behavior to be
# data equality (not element-wise equality)
arr1 = pa.array([1, 2, 3], type=pa.int32())
arr2 = pa.array([1, 2, 3], type=pa.int32())
arr3 = pa.array([1, 2, 3], type=pa.int64())
assert (arr1 == arr2) is True
assert (arr1 != arr2) is False
assert (arr1 == arr3) is False
assert (arr1 != arr3) is True
assert (arr1 == 1) is False
assert (arr1 == None) is False # noqa: E711
@pytest.mark.numpy
def test_array_from_buffers():
values_buf = pa.py_buffer(np.int16([4, 5, 6, 7]))
nulls_buf = pa.py_buffer(np.uint8([0b00001101]))
arr = pa.Array.from_buffers(pa.int16(), 4, [nulls_buf, values_buf])
assert arr.type == pa.int16()
assert arr.to_pylist() == [4, None, 6, 7]
arr = pa.Array.from_buffers(pa.int16(), 4, [None, values_buf])
assert arr.type == pa.int16()
assert arr.to_pylist() == [4, 5, 6, 7]
arr = pa.Array.from_buffers(pa.int16(), 3, [nulls_buf, values_buf],
offset=1)
assert arr.type == pa.int16()
assert arr.to_pylist() == [None, 6, 7]
with pytest.raises(TypeError):
pa.Array.from_buffers(pa.int16(), 3, ['', ''], offset=1)
def test_string_binary_from_buffers():
array = pa.array(["a", None, "b", "c"])
buffers = array.buffers()
copied = pa.StringArray.from_buffers(
len(array), buffers[1], buffers[2], buffers[0], array.null_count,
array.offset)
assert copied.to_pylist() == ["a", None, "b", "c"]
binary_copy = pa.Array.from_buffers(pa.binary(), len(array),
array.buffers(), array.null_count,
array.offset)
assert binary_copy.to_pylist() == [b"a", None, b"b", b"c"]
copied = pa.StringArray.from_buffers(
len(array), buffers[1], buffers[2], buffers[0])
assert copied.to_pylist() == ["a", None, "b", "c"]
sliced = array[1:]
buffers = sliced.buffers()
copied = pa.StringArray.from_buffers(
len(sliced), buffers[1], buffers[2], buffers[0], -1, sliced.offset)
assert copied.to_pylist() == [None, "b", "c"]
assert copied.null_count == 1
# Slice but exclude all null entries so that we don't need to pass
# the null bitmap.
sliced = array[2:]
buffers = sliced.buffers()
copied = pa.StringArray.from_buffers(
len(sliced), buffers[1], buffers[2], None, -1, sliced.offset)
assert copied.to_pylist() == ["b", "c"]
assert copied.null_count == 0
def test_string_view_from_buffers():
array = pa.array(
[
"String longer than 12 characters",
None,
"short",
"Length is 12"
], type=pa.string_view())
buffers = array.buffers()
copied = pa.StringViewArray.from_buffers(
pa.string_view(), len(array), buffers)
copied.validate(full=True)
assert copied.to_pylist() == [
"String longer than 12 characters",
None,
"short",
"Length is 12"
]
match = r"number of buffers is at least 2"
with pytest.raises(ValueError, match=match):
pa.StringViewArray.from_buffers(
pa.string_view(), len(array), buffers[0:1])
@pytest.mark.parametrize('list_type_factory', [
pa.list_, pa.large_list, pa.list_view, pa.large_list_view])
def test_list_from_buffers(list_type_factory):
ty = list_type_factory(pa.int16())
array = pa.array([[0, 1, 2], None, [], [3, 4, 5]], type=ty)
assert array.type == ty
buffers = array.buffers()
with pytest.raises(ValueError):
# No children
pa.Array.from_buffers(ty, 4, buffers[:ty.num_buffers])
child = pa.Array.from_buffers(pa.int16(), 6, buffers[ty.num_buffers:])
copied = pa.Array.from_buffers(ty, 4, buffers[:ty.num_buffers], children=[child])
assert copied.equals(array)
with pytest.raises(ValueError):
# too many children
pa.Array.from_buffers(ty, 4, buffers[:ty.num_buffers],
children=[child, child])
def test_struct_from_buffers():
ty = pa.struct([pa.field('a', pa.int16()), pa.field('b', pa.utf8())])
array = pa.array([{'a': 0, 'b': 'foo'}, None, {'a': 5, 'b': ''}],
type=ty)
buffers = array.buffers()
with pytest.raises(ValueError):
# No children
pa.Array.from_buffers(ty, 3, [None, buffers[1]])
children = [pa.Array.from_buffers(pa.int16(), 3, buffers[1:3]),
pa.Array.from_buffers(pa.utf8(), 3, buffers[3:])]
copied = pa.Array.from_buffers(ty, 3, buffers[:1], children=children)
assert copied.equals(array)
with pytest.raises(ValueError):
# not enough many children
pa.Array.from_buffers(ty, 3, [buffers[0]],
children=children[:1])
def test_struct_from_arrays():
a = pa.array([4, 5, 6], type=pa.int64())
b = pa.array(["bar", None, ""])
c = pa.array([[1, 2], None, [3, None]])
expected_list = [
{'a': 4, 'b': 'bar', 'c': [1, 2]},
{'a': 5, 'b': None, 'c': None},
{'a': 6, 'b': '', 'c': [3, None]},
]
# From field names
arr = pa.StructArray.from_arrays([a, b, c], ["a", "b", "c"])
assert arr.type == pa.struct(
[("a", a.type), ("b", b.type), ("c", c.type)])
assert arr.to_pylist() == expected_list
with pytest.raises(ValueError):
pa.StructArray.from_arrays([a, b, c], ["a", "b"])
arr = pa.StructArray.from_arrays([], [])
assert arr.type == pa.struct([])
assert arr.to_pylist() == []
# From fields
fa = pa.field("a", a.type, nullable=False)
fb = pa.field("b", b.type)
fc = pa.field("c", c.type)
arr = pa.StructArray.from_arrays([a, b, c], fields=[fa, fb, fc])
assert arr.type == pa.struct([fa, fb, fc])
assert not arr.type[0].nullable
assert arr.to_pylist() == expected_list
# From structtype
structtype = pa.struct([fa, fb, fc])
arr = pa.StructArray.from_arrays([a, b, c], type=structtype)
assert arr.type == pa.struct([fa, fb, fc])
assert not arr.type[0].nullable
assert arr.to_pylist() == expected_list
with pytest.raises(ValueError):
pa.StructArray.from_arrays([a, b, c], fields=[fa, fb])
arr = pa.StructArray.from_arrays([], fields=[])
assert arr.type == pa.struct([])
assert arr.to_pylist() == []
# Inconsistent fields
fa2 = pa.field("a", pa.int32())
with pytest.raises(ValueError, match="int64 vs int32"):
pa.StructArray.from_arrays([a, b, c], fields=[fa2, fb, fc])
arrays = [a, b, c]
fields = [fa, fb, fc]
# With mask
mask = pa.array([True, False, False])
arr = pa.StructArray.from_arrays(arrays, fields=fields, mask=mask)
assert arr.to_pylist() == [None] + expected_list[1:]
arr = pa.StructArray.from_arrays(arrays, names=['a', 'b', 'c'], mask=mask)
assert arr.to_pylist() == [None] + expected_list[1:]
# Bad masks
with pytest.raises(TypeError, match='Mask must be'):
pa.StructArray.from_arrays(arrays, fields, mask=[True, False, False])
with pytest.raises(ValueError, match='not contain nulls'):
pa.StructArray.from_arrays(
arrays, fields, mask=pa.array([True, False, None]))
with pytest.raises(TypeError, match='Mask must be'):
pa.StructArray.from_arrays(
arrays, fields, mask=pa.chunked_array([mask]))
# Non-empty array with no fields https://github.com/apache/arrow/issues/15109
arr = pa.StructArray.from_arrays([], [], mask=mask)
assert arr.is_null() == mask
assert arr.to_pylist() == [None, {}, {}]
def test_struct_array_from_chunked():
# ARROW-11780
# Check that we don't segfault when trying to build
# a StructArray from a chunked array.
chunked_arr = pa.chunked_array([[1, 2, 3], [4, 5, 6]])
with pytest.raises(TypeError, match="Expected Array"):
pa.StructArray.from_arrays([chunked_arr], ["foo"])
@pytest.mark.parametrize("offset", (0, 1))
def test_dictionary_from_buffers(offset):
a = pa.array(["one", "two", "three", "two", "one"]).dictionary_encode()
b = pa.DictionaryArray.from_buffers(a.type, len(a)-offset,
a.indices.buffers(), a.dictionary,
offset=offset)
assert a[offset:] == b
@pytest.mark.numpy
def test_dictionary_from_numpy():
indices = np.repeat([0, 1, 2], 2)
dictionary = np.array(['foo', 'bar', 'baz'], dtype=object)
mask = np.array([False, False, True, False, False, False])
d1 = pa.DictionaryArray.from_arrays(indices, dictionary)
d2 = pa.DictionaryArray.from_arrays(indices, dictionary, mask=mask)
assert d1.indices.to_pylist() == indices.tolist()
assert d1.indices.to_pylist() == indices.tolist()
assert d1.dictionary.to_pylist() == dictionary.tolist()
assert d2.dictionary.to_pylist() == dictionary.tolist()
for i in range(len(indices)):
assert d1[i].as_py() == dictionary[indices[i]]
if mask[i]:
assert d2[i].as_py() is None
else:
assert d2[i].as_py() == dictionary[indices[i]]
@pytest.mark.numpy
def test_dictionary_to_numpy():
expected = pa.array(
["foo", "bar", None, "foo"]
).to_numpy(zero_copy_only=False)
a = pa.DictionaryArray.from_arrays(
pa.array([0, 1, None, 0]),
pa.array(['foo', 'bar'])
)
np.testing.assert_array_equal(a.to_numpy(zero_copy_only=False),
expected)
with pytest.raises(pa.ArrowInvalid):
# If this would be changed to no longer raise in the future,
# ensure to test the actual result because, currently, to_numpy takes
# for granted that when zero_copy_only=True there will be no nulls
# (it's the decoding of the DictionaryArray that handles the nulls and
# this is only activated with zero_copy_only=False)
a.to_numpy(zero_copy_only=True)
anonulls = pa.DictionaryArray.from_arrays(
pa.array([0, 1, 1, 0]),
pa.array(['foo', 'bar'])
)
expected = pa.array(
["foo", "bar", "bar", "foo"]
).to_numpy(zero_copy_only=False)
np.testing.assert_array_equal(anonulls.to_numpy(zero_copy_only=False),
expected)
with pytest.raises(pa.ArrowInvalid):
anonulls.to_numpy(zero_copy_only=True)
afloat = pa.DictionaryArray.from_arrays(
pa.array([0, 1, 1, 0]),
pa.array([13.7, 11.0])
)
expected = pa.array([13.7, 11.0, 11.0, 13.7]).to_numpy()
np.testing.assert_array_equal(afloat.to_numpy(zero_copy_only=True),
expected)
np.testing.assert_array_equal(afloat.to_numpy(zero_copy_only=False),
expected)
afloat2 = pa.DictionaryArray.from_arrays(
pa.array([0, 1, None, 0]),