-
Notifications
You must be signed in to change notification settings - Fork 8
/
Copy pathnep-0016-abstract-array.html
904 lines (701 loc) · 56.2 KB
/
nep-0016-abstract-array.html
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
<!DOCTYPE html>
<html lang="en" data-content_root="./" >
<head>
<meta charset="utf-8" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
<title>NEP 16 — An abstract base class for identifying “duck arrays” — NumPy Enhancement Proposals</title>
<script data-cfasync="false">
document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
</script>
<!--
this give us a css class that will be invisible only if js is disabled
-->
<noscript>
<style>
.pst-js-only { display: none !important; }
</style>
</noscript>
<!-- Loaded before other Sphinx assets -->
<link href="_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
<link href="_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
<link rel="stylesheet" type="text/css" href="_static/pygments.css?v=03e43079" />
<!-- So that users can add custom icons -->
<script src="_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
<!-- Pre-loaded scripts that we'll load fully later -->
<link rel="preload" as="script" href="_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
<link rel="preload" as="script" href="_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
<script src="_static/documentation_options.js?v=7f41d439"></script>
<script src="_static/doctools.js?v=888ff710"></script>
<script src="_static/sphinx_highlight.js?v=dc90522c"></script>
<script>DOCUMENTATION_OPTIONS.pagename = 'nep-0016-abstract-array';</script>
<link rel="icon" href="_static/favicon.ico"/>
<link rel="index" title="Index" href="genindex.html" />
<link rel="search" title="Search" href="search.html" />
<link rel="next" title="NEP 17 — Split out masked arrays" href="nep-0017-split-out-maskedarray.html" />
<link rel="prev" title="Rejected and Withdrawn NEPs" href="rejected.html" />
<meta name="viewport" content="width=device-width, initial-scale=1"/>
<meta name="docsearch:language" content="en"/>
<meta name="docsearch:version" content="" />
<meta name="docbuild:last-update" content="Apr 17, 2025"/>
</head>
<body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
<div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
<div id="pst-scroll-pixel-helper"></div>
<button type="button" class="btn rounded-pill" id="pst-back-to-top">
<i class="fa-solid fa-arrow-up"></i>Back to top</button>
<dialog id="pst-search-dialog">
<form class="bd-search d-flex align-items-center"
action="search.html"
method="get">
<i class="fa-solid fa-magnifying-glass"></i>
<input type="search"
class="form-control"
name="q"
placeholder="Search the docs ..."
aria-label="Search the docs ..."
autocomplete="off"
autocorrect="off"
autocapitalize="off"
spellcheck="false"/>
<span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
</form>
</dialog>
<div class="pst-async-banner-revealer d-none">
<aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
</div>
<header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
<div class="bd-header__inner bd-page-width">
<button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
<span class="fa-solid fa-bars"></span>
</button>
<div class="col-lg-3 navbar-header-items__start">
<div class="navbar-item">
<a class="navbar-brand logo" href="content.html">
<img src="_static/numpylogo.svg" class="logo__image only-light" alt="NumPy Enhancement Proposals - Home"/>
<img src="_static/numpylogo_dark.svg" class="logo__image only-dark pst-js-only" alt="NumPy Enhancement Proposals - Home"/>
</a></div>
</div>
<div class="col-lg-9 navbar-header-items">
<div class="me-auto navbar-header-items__center">
<div class="navbar-item">
<nav>
<ul class="bd-navbar-elements navbar-nav">
<li class="nav-item current active">
<a class="nav-link nav-internal" href="index.html">
Index
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="scope.html">
The Scope of NumPy
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="roadmap.html">
Current roadmap
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-external" href="https://github.com/numpy/numpy/issues?q=is%3Aopen+is%3Aissue+label%3A%2223+-+Wish+List%22">
Wishlist
</a>
</li>
</ul>
</nav></div>
</div>
<div class="navbar-header-items__end">
<div class="navbar-item navbar-persistent--container">
<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
<i class="fa-solid fa-magnifying-glass"></i>
<span class="search-button__default-text">Search</span>
<span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
</button>
</div>
<div class="navbar-item">
<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode" data-bs-placement="bottom" data-bs-toggle="tooltip">
<i class="theme-switch fa-solid fa-sun fa-lg" data-mode="light" title="Light"></i>
<i class="theme-switch fa-solid fa-moon fa-lg" data-mode="dark" title="Dark"></i>
<i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto" title="System Settings"></i>
</button></div>
<div class="navbar-item"><ul class="navbar-icon-links"
aria-label="Icon Links">
<li class="nav-item">
<a href="https://github.com/numpy/numpy" title="GitHub" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><i class="fa-brands fa-square-github fa-lg" aria-hidden="true"></i>
<span class="sr-only">GitHub</span></a>
</li>
</ul></div>
</div>
</div>
<div class="navbar-persistent--mobile">
<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
<i class="fa-solid fa-magnifying-glass"></i>
<span class="search-button__default-text">Search</span>
<span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
</button>
</div>
<button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
<span class="fa-solid fa-outdent"></span>
</button>
</div>
</header>
<div class="bd-container">
<div class="bd-container__inner bd-page-width">
<dialog id="pst-primary-sidebar-modal"></dialog>
<div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
<div class="sidebar-header-items sidebar-primary__section">
<div class="sidebar-header-items__center">
<div class="navbar-item">
<nav>
<ul class="bd-navbar-elements navbar-nav">
<li class="nav-item current active">
<a class="nav-link nav-internal" href="index.html">
Index
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="scope.html">
The Scope of NumPy
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="roadmap.html">
Current roadmap
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-external" href="https://github.com/numpy/numpy/issues?q=is%3Aopen+is%3Aissue+label%3A%2223+-+Wish+List%22">
Wishlist
</a>
</li>
</ul>
</nav></div>
</div>
<div class="sidebar-header-items__end">
<div class="navbar-item">
<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode" data-bs-placement="bottom" data-bs-toggle="tooltip">
<i class="theme-switch fa-solid fa-sun fa-lg" data-mode="light" title="Light"></i>
<i class="theme-switch fa-solid fa-moon fa-lg" data-mode="dark" title="Dark"></i>
<i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto" title="System Settings"></i>
</button></div>
<div class="navbar-item"><ul class="navbar-icon-links"
aria-label="Icon Links">
<li class="nav-item">
<a href="https://github.com/numpy/numpy" title="GitHub" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><i class="fa-brands fa-square-github fa-lg" aria-hidden="true"></i>
<span class="sr-only">GitHub</span></a>
</li>
</ul></div>
</div>
</div>
<div class="sidebar-primary-items__start sidebar-primary__section">
<div class="sidebar-primary-item">
<nav class="bd-docs-nav bd-links"
aria-label="Section Navigation">
<p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
<div class="bd-toc-item navbar-nav"><ul class="nav bd-sidenav">
<li class="toctree-l1"><a class="reference internal" href="scope.html">The Scope of NumPy</a></li>
<li class="toctree-l1"><a class="reference internal" href="roadmap.html">Current roadmap</a></li>
</ul>
<ul class="current nav bd-sidenav">
<li class="toctree-l1 has-children"><a class="reference internal" href="meta.html">Meta-NEPs (NEPs about NEPs or active Processes)</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l2"><a class="reference internal" href="nep-0000.html">NEP 0 — Purpose and process</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0023-backwards-compatibility.html">NEP 23 — Backwards compatibility and deprecation policy</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0036-fair-play.html">NEP 36 — Fair play</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0045-c_style_guide.html">NEP 45 — C style guide</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0046-sponsorship-guidelines.html">NEP 46 — NumPy sponsorship guidelines</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0048-spending-project-funds.html">NEP 48 — Spending NumPy project funds</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-template.html">NEP X — Template and instructions</a></li>
</ul>
</details></li>
<li class="toctree-l1 has-children"><a class="reference internal" href="provisional.html">Provisional NEPs (provisionally accepted; interface may change)</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="simple">
</ul>
</details></li>
<li class="toctree-l1 has-children"><a class="reference internal" href="accepted.html">Accepted NEPs (implementation in progress)</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l2"><a class="reference internal" href="nep-0041-improved-dtype-support.html">NEP 41 — First step towards a new datatype system</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0042-new-dtypes.html">NEP 42 — New and extensible DTypes</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0044-restructuring-numpy-docs.html">NEP 44 — Restructuring the NumPy documentation</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0051-scalar-representation.html">NEP 51 — Changing the representation of NumPy scalars</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0054-simd-cpp-highway.html">NEP 54 — SIMD infrastructure evolution: adopting Google Highway when moving to C++</a></li>
</ul>
</details></li>
<li class="toctree-l1 has-children"><a class="reference internal" href="open.html">Open NEPs (under consideration)</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l2"><a class="reference internal" href="nep-0043-extensible-ufuncs.html">NEP 43 — Enhancing the extensibility of UFuncs</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0053-c-abi-evolution.html">NEP 53 — Evolving the NumPy C-API for NumPy 2.0</a></li>
</ul>
</details></li>
<li class="toctree-l1 has-children"><a class="reference internal" href="finished.html">Finished NEPs</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l2"><a class="reference internal" href="nep-0001-npy-format.html">NEP 1 — A simple file format for NumPy arrays</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0005-generalized-ufuncs.html">NEP 5 — Generalized universal functions</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0007-datetime-proposal.html">NEP 7 — A proposal for implementing some date/time types in NumPy</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0010-new-iterator-ufunc.html">NEP 10 — Optimizing iterator/UFunc performance</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0013-ufunc-overrides.html">NEP 13 — A mechanism for overriding Ufuncs</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0014-dropping-python2.7-proposal.html">NEP 14 — Plan for dropping Python 2.7 support</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0015-merge-multiarray-umath.html">NEP 15 — Merging multiarray and umath</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0018-array-function-protocol.html">NEP 18 — A dispatch mechanism for NumPy's high level array functions</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0019-rng-policy.html">NEP 19 — Random number generator policy</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0020-gufunc-signature-enhancement.html">NEP 20 — Expansion of generalized universal function signatures</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0022-ndarray-duck-typing-overview.html">NEP 22 — Duck typing for NumPy arrays – high level overview</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0027-zero-rank-arrarys.html">NEP 27 — Zero rank arrays</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0028-website-redesign.html">NEP 28 — numpy.org website redesign</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0029-deprecation_policy.html">NEP 29 — Recommend Python and NumPy version support as a community policy standard</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0032-remove-financial-functions.html">NEP 32 — Remove the financial functions from NumPy</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0034-infer-dtype-is-object.html">NEP 34 — Disallow inferring ``dtype=object`` from sequences</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0035-array-creation-dispatch-with-array-function.html">NEP 35 — Array creation dispatching with __array_function__</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0038-SIMD-optimizations.html">NEP 38 — Using SIMD optimization instructions for performance</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0040-legacy-datatype-impl.html">NEP 40 — Legacy datatype implementation in NumPy</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0049.html">NEP 49 — Data allocation strategies</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0050-scalar-promotion.html">NEP 50 — Promotion rules for Python scalars</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0052-python-api-cleanup.html">NEP 52 — Python API cleanup for NumPy 2.0</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0055-string_dtype.html">NEP 55 — Add a UTF-8 variable-width string DType to NumPy</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0056-array-api-main-namespace.html">NEP 56 — Array API standard support in NumPy's main namespace</a></li>
</ul>
</details></li>
<li class="toctree-l1 has-children"><a class="reference internal" href="deferred.html">Deferred and Superseded NEPs</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l2"><a class="reference internal" href="nep-0002-warnfix.html">NEP 2 — A proposal to build numpy without warning with a big set of warning flags</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0003-math_config_clean.html">NEP 3 — Cleaning the math configuration of numpy.core</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0004-datetime-proposal3.html">NEP 4 — A (third) proposal for implementing some date/time types in NumPy</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0006-newbugtracker.html">NEP 6 — Replacing Trac with a different bug tracker</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0008-groupby_additions.html">NEP 8 — A proposal for adding groupby functionality to NumPy</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0009-structured_array_extensions.html">NEP 9 — Structured array extensions</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0011-deferred-ufunc-evaluation.html">NEP 11 — Deferred UFunc evaluation</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0012-missing-data.html">NEP 12 — Missing data functionality in NumPy</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0021-advanced-indexing.html">NEP 21 — Simplified and explicit advanced indexing</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0024-missing-data-2.html">NEP 24 — Missing data functionality - alternative 1 to NEP 12</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0025-missing-data-3.html">NEP 25 — NA support via special dtypes</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0026-missing-data-summary.html">NEP 26 — Summary of missing data NEPs and discussion</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0030-duck-array-protocol.html">NEP 30 — Duck typing for NumPy arrays - implementation</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0031-uarray.html">NEP 31 — Context-local and global overrides of the NumPy API</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0037-array-module.html">NEP 37 — A dispatch protocol for NumPy-like modules</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0047-array-api-standard.html">NEP 47 — Adopting the array API standard</a></li>
</ul>
</details></li>
<li class="toctree-l1 current active has-children"><a class="reference internal" href="rejected.html">Rejected and Withdrawn NEPs</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
<li class="toctree-l2 current active"><a class="current reference internal" href="#">NEP 16 — An abstract base class for identifying "duck arrays"</a></li>
<li class="toctree-l2"><a class="reference internal" href="nep-0017-split-out-maskedarray.html">NEP 17 — Split out masked arrays</a></li>
</ul>
</details></li>
</ul>
</div>
</nav></div>
</div>
<div class="sidebar-primary-items__end sidebar-primary__section">
<div class="sidebar-primary-item">
<div id="ethical-ad-placement"
class="flat"
data-ea-publisher="readthedocs"
data-ea-type="readthedocs-sidebar"
data-ea-manual="true">
</div></div>
</div>
</div>
<main id="main-content" class="bd-main" role="main">
<div class="bd-content">
<div class="bd-article-container">
<div class="bd-header-article d-print-none">
<div class="header-article-items header-article__inner">
<div class="header-article-items__start">
<div class="header-article-item">
<nav aria-label="Breadcrumb" class="d-print-none">
<ul class="bd-breadcrumbs">
<li class="breadcrumb-item breadcrumb-home">
<a href="content.html" class="nav-link" aria-label="Home">
<i class="fa-solid fa-home"></i>
</a>
</li>
<li class="breadcrumb-item"><a href="index.html" class="nav-link">Roadmap & NumPy enhancement proposals</a></li>
<li class="breadcrumb-item"><a href="rejected.html" class="nav-link">Rejected and Withdrawn NEPs</a></li>
<li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">NEP 16 — An abstract base class for identifying “duck arrays”</span></li>
</ul>
</nav>
</div>
</div>
</div>
</div>
<div id="searchbox"></div>
<article class="bd-article">
<section id="nep-16-an-abstract-base-class-for-identifying-duck-arrays">
<span id="nep16"></span><h1>NEP 16 — An abstract base class for identifying “duck arrays”<a class="headerlink" href="#nep-16-an-abstract-base-class-for-identifying-duck-arrays" title="Link to this heading">#</a></h1>
<dl class="field-list simple">
<dt class="field-odd">Author<span class="colon">:</span></dt>
<dd class="field-odd"><p>Nathaniel J. Smith <<a class="reference external" href="mailto:njs%40pobox.com">njs<span>@</span>pobox<span>.</span>com</a>></p>
</dd>
<dt class="field-even">Status<span class="colon">:</span></dt>
<dd class="field-even"><p>Withdrawn</p>
</dd>
<dt class="field-odd">Type<span class="colon">:</span></dt>
<dd class="field-odd"><p>Standards Track</p>
</dd>
<dt class="field-even">Created<span class="colon">:</span></dt>
<dd class="field-even"><p>2018-03-06</p>
</dd>
<dt class="field-odd">Resolution<span class="colon">:</span></dt>
<dd class="field-odd"><p><a class="github reference external" href="https://github.com/numpy/numpy/pull/12174">numpy/numpy#12174</a></p>
</dd>
</dl>
<div class="admonition note">
<p class="admonition-title">Note</p>
<p>This NEP has been withdrawn in favor of the protocol based approach
described in <a class="reference internal" href="nep-0022-ndarray-duck-typing-overview.html"><span class="doc">NEP 22 — Duck typing for NumPy arrays – high level overview</span></a></p>
</div>
<section id="abstract">
<h2>Abstract<a class="headerlink" href="#abstract" title="Link to this heading">#</a></h2>
<p>We propose to add an abstract base class <code class="docutils literal notranslate"><span class="pre">AbstractArray</span></code> so that
third-party classes can declare their ability to “quack like” an
<code class="docutils literal notranslate"><span class="pre">ndarray</span></code>, and an <code class="docutils literal notranslate"><span class="pre">asabstractarray</span></code> function that performs
similarly to <code class="docutils literal notranslate"><span class="pre">asarray</span></code> except that it passes through
<code class="docutils literal notranslate"><span class="pre">AbstractArray</span></code> instances unchanged.</p>
</section>
<section id="detailed-description">
<h2>Detailed description<a class="headerlink" href="#detailed-description" title="Link to this heading">#</a></h2>
<p>Many functions, in NumPy and in third-party packages, start with some
code like:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">def</span><span class="w"> </span><span class="nf">myfunc</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">):</span>
<span class="n">a</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">asarray</span><span class="p">(</span><span class="n">a</span><span class="p">)</span>
<span class="n">b</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">asarray</span><span class="p">(</span><span class="n">b</span><span class="p">)</span>
<span class="o">...</span>
</pre></div>
</div>
<p>This ensures that <code class="docutils literal notranslate"><span class="pre">a</span></code> and <code class="docutils literal notranslate"><span class="pre">b</span></code> are <code class="docutils literal notranslate"><span class="pre">np.ndarray</span></code> objects, so
<code class="docutils literal notranslate"><span class="pre">myfunc</span></code> can carry on assuming that they’ll act like ndarrays both
semantically (at the Python level), and also in terms of how they’re
stored in memory (at the C level). But many of these functions only
work with arrays at the Python level, which means that they don’t
actually need <code class="docutils literal notranslate"><span class="pre">ndarray</span></code> objects <em>per se</em>: they could work just as
well with any Python object that “quacks like” an ndarray, such as
sparse arrays, dask’s lazy arrays, or xarray’s labeled arrays.</p>
<p>However, currently, there’s no way for these libraries to express that
their objects can quack like an ndarray, and there’s no way for
functions like <code class="docutils literal notranslate"><span class="pre">myfunc</span></code> to express that they’d be happy with
anything that quacks like an ndarray. The purpose of this NEP is to
provide those two features.</p>
<p>Sometimes people suggest using <code class="docutils literal notranslate"><span class="pre">np.asanyarray</span></code> for this purpose, but
unfortunately its semantics are exactly backwards: it guarantees that
the object it returns uses the same memory layout as an <code class="docutils literal notranslate"><span class="pre">ndarray</span></code>,
but tells you nothing at all about its semantics, which makes it
essentially impossible to use safely in practice. Indeed, the two
<code class="docutils literal notranslate"><span class="pre">ndarray</span></code> subclasses distributed with NumPy – <code class="docutils literal notranslate"><span class="pre">np.matrix</span></code> and
<code class="docutils literal notranslate"><span class="pre">np.ma.masked_array</span></code> – do have incompatible semantics, and if they
were passed to a function like <code class="docutils literal notranslate"><span class="pre">myfunc</span></code> that doesn’t check for them
as a special-case, then it may silently return incorrect results.</p>
<section id="declaring-that-an-object-can-quack-like-an-array">
<h3>Declaring that an object can quack like an array<a class="headerlink" href="#declaring-that-an-object-can-quack-like-an-array" title="Link to this heading">#</a></h3>
<p>There are two basic approaches we could use for checking whether an
object quacks like an array. We could check for a special attribute on
the class:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">def</span><span class="w"> </span><span class="nf">quacks_like_array</span><span class="p">(</span><span class="n">obj</span><span class="p">):</span>
<span class="k">return</span> <span class="nb">bool</span><span class="p">(</span><span class="nb">getattr</span><span class="p">(</span><span class="nb">type</span><span class="p">(</span><span class="n">obj</span><span class="p">),</span> <span class="s2">"__quacks_like_array__"</span><span class="p">,</span> <span class="kc">False</span><span class="p">))</span>
</pre></div>
</div>
<p>Or, we could define an <a class="reference external" href="https://docs.python.org/3/library/collections.abc.html">abstract base class (ABC)</a>:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">def</span><span class="w"> </span><span class="nf">quacks_like_array</span><span class="p">(</span><span class="n">obj</span><span class="p">):</span>
<span class="k">return</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">obj</span><span class="p">,</span> <span class="n">AbstractArray</span><span class="p">)</span>
</pre></div>
</div>
<p>If you look at how ABCs work, this is essentially equivalent to
keeping a global set of types that have been declared to implement the
<code class="docutils literal notranslate"><span class="pre">AbstractArray</span></code> interface, and then checking it for membership.</p>
<p>Between these, the ABC approach seems to have a number of advantages:</p>
<ul class="simple">
<li><p>It’s Python’s standard, “one obvious way” of doing this.</p></li>
<li><p>ABCs can be introspected (e.g. <code class="docutils literal notranslate"><span class="pre">help(np.AbstractArray)</span></code> does
something useful).</p></li>
<li><p>ABCs can provide useful mixin methods.</p></li>
<li><p>ABCs integrate with other features like mypy type-checking,
<code class="docutils literal notranslate"><span class="pre">functools.singledispatch</span></code>, etc.</p></li>
</ul>
<p>One obvious thing to check is whether this choice affects speed. Using
the attached benchmark script on a CPython 3.7 prerelease (revision
c4d77a661138d, self-compiled, no PGO), on a Thinkpad T450s running
Linux, we find:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">np</span><span class="o">.</span><span class="n">asarray</span><span class="p">(</span><span class="n">ndarray_obj</span><span class="p">)</span> <span class="mi">330</span> <span class="n">ns</span>
<span class="n">np</span><span class="o">.</span><span class="n">asarray</span><span class="p">([])</span> <span class="mi">1400</span> <span class="n">ns</span>
<span class="n">Attribute</span> <span class="n">check</span><span class="p">,</span> <span class="n">success</span> <span class="mi">80</span> <span class="n">ns</span>
<span class="n">Attribute</span> <span class="n">check</span><span class="p">,</span> <span class="n">failure</span> <span class="mi">80</span> <span class="n">ns</span>
<span class="n">ABC</span><span class="p">,</span> <span class="n">success</span> <span class="n">via</span> <span class="n">subclass</span> <span class="mi">340</span> <span class="n">ns</span>
<span class="n">ABC</span><span class="p">,</span> <span class="n">success</span> <span class="n">via</span> <span class="n">register</span><span class="p">()</span> <span class="mi">700</span> <span class="n">ns</span>
<span class="n">ABC</span><span class="p">,</span> <span class="n">failure</span> <span class="mi">370</span> <span class="n">ns</span>
</pre></div>
</div>
<p>Notes:</p>
<ul>
<li><p>The first two lines are included to put the other lines in context.</p></li>
<li><p>This used 3.7 because both <code class="docutils literal notranslate"><span class="pre">getattr</span></code> and ABCs are receiving
substantial optimizations in this release, and it’s more
representative of the long-term future of Python. (Failed
<code class="docutils literal notranslate"><span class="pre">getattr</span></code> doesn’t necessarily construct an exception object
anymore, and ABCs were reimplemented in C.)</p></li>
<li><p>The “success” lines refer to cases where <code class="docutils literal notranslate"><span class="pre">quacks_like_array</span></code> would
return True. The “failure” lines are cases where it would return
False.</p></li>
<li><p>The first measurement for ABCs is subclasses defined like:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">class</span><span class="w"> </span><span class="nc">MyArray</span><span class="p">(</span><span class="n">AbstractArray</span><span class="p">):</span>
<span class="o">...</span>
</pre></div>
</div>
<p>The second is for subclasses defined like:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">class</span><span class="w"> </span><span class="nc">MyArray</span><span class="p">:</span>
<span class="o">...</span>
<span class="n">AbstractArray</span><span class="o">.</span><span class="n">register</span><span class="p">(</span><span class="n">MyArray</span><span class="p">)</span>
</pre></div>
</div>
<p>I don’t know why there’s such a large difference between these.</p>
</li>
</ul>
<p>In practice, either way we’d only do the full test after first
checking for well-known types like <code class="docutils literal notranslate"><span class="pre">ndarray</span></code>, <code class="docutils literal notranslate"><span class="pre">list</span></code>, etc. <a class="reference external" href="https://github.com/numpy/numpy/blob/main/numpy/core/src/private/get_attr_string.h">This
is how NumPy currently checks for other double-underscore attributes</a>
and the same idea applies here to either approach. So these numbers
won’t affect the common case, just the case where we actually have an
<code class="docutils literal notranslate"><span class="pre">AbstractArray</span></code>, or else another third-party object that will end up
going through <code class="docutils literal notranslate"><span class="pre">__array__</span></code> or <code class="docutils literal notranslate"><span class="pre">__array_interface__</span></code> or end up as an
object array.</p>
<p>So in summary, using an ABC will be slightly slower than using an
attribute, but this doesn’t affect the most common paths, and the
magnitude of slowdown is fairly small (~250 ns on an operation that
already takes longer than that). Furthermore, we can potentially
optimize this further (e.g. by keeping a tiny LRU cache of types that
are known to be AbstractArray subclasses, on the assumption that most
code will only use one or two of these types at a time), and it’s very
unclear that this even matters – if the speed of <code class="docutils literal notranslate"><span class="pre">asarray</span></code> no-op
pass-throughs were a bottleneck that showed up in profiles, then
probably we would have made them faster already! (It would be trivial
to fast-path this, but we don’t.)</p>
<p>Given the semantic and usability advantages of ABCs, this seems like
an acceptable trade-off.</p>
</section>
<section id="specification-of-asabstractarray">
<h3>Specification of <code class="docutils literal notranslate"><span class="pre">asabstractarray</span></code><a class="headerlink" href="#specification-of-asabstractarray" title="Link to this heading">#</a></h3>
<p>Given <code class="docutils literal notranslate"><span class="pre">AbstractArray</span></code>, the definition of <code class="docutils literal notranslate"><span class="pre">asabstractarray</span></code> is simple:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">def</span><span class="w"> </span><span class="nf">asabstractarray</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
<span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">AbstractArray</span><span class="p">):</span>
<span class="k">if</span> <span class="n">dtype</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">dtype</span> <span class="o">!=</span> <span class="n">a</span><span class="o">.</span><span class="n">dtype</span><span class="p">:</span>
<span class="k">return</span> <span class="n">a</span><span class="o">.</span><span class="n">astype</span><span class="p">(</span><span class="n">dtype</span><span class="p">)</span>
<span class="k">return</span> <span class="n">a</span>
<span class="k">return</span> <span class="n">asarray</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">)</span>
</pre></div>
</div>
<p>Things to note:</p>
<ul>
<li><p><code class="docutils literal notranslate"><span class="pre">asarray</span></code> also accepts an <code class="docutils literal notranslate"><span class="pre">order=</span></code> argument, but we don’t
include that here because it’s about details of memory
representation, and the whole point of this function is that you use
it to declare that you don’t care about details of memory
representation.</p></li>
<li><p>Using the <code class="docutils literal notranslate"><span class="pre">astype</span></code> method allows the <code class="docutils literal notranslate"><span class="pre">a</span></code> object to decide how to
implement casting for its particular type.</p></li>
<li><p>For strict compatibility with <code class="docutils literal notranslate"><span class="pre">asarray</span></code>, we skip calling
<code class="docutils literal notranslate"><span class="pre">astype</span></code> when the dtype is already correct. Compare:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">a</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="mi">10</span><span class="p">)</span>
<span class="go"># astype() always returns a view:</span>
<span class="gp">>>> </span><span class="n">a</span><span class="o">.</span><span class="n">astype</span><span class="p">(</span><span class="n">a</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span> <span class="ow">is</span> <span class="n">a</span>
<span class="go">False</span>
<span class="go"># asarray() returns the original object if possible:</span>
<span class="gp">>>> </span><span class="n">np</span><span class="o">.</span><span class="n">asarray</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">a</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span> <span class="ow">is</span> <span class="n">a</span>
<span class="go">True</span>
</pre></div>
</div>
</li>
</ul>
</section>
<section id="what-exactly-are-you-promising-if-you-inherit-from-abstractarray">
<h3>What exactly are you promising if you inherit from <code class="docutils literal notranslate"><span class="pre">AbstractArray</span></code>?<a class="headerlink" href="#what-exactly-are-you-promising-if-you-inherit-from-abstractarray" title="Link to this heading">#</a></h3>
<p>This will presumably be refined over time. The ideal of course is that
your class should be indistinguishable from a real <code class="docutils literal notranslate"><span class="pre">ndarray</span></code>, but
nothing enforces that except the expectations of users. In practice,
declaring that your class implements the <code class="docutils literal notranslate"><span class="pre">AbstractArray</span></code> interface
simply means that it will start passing through <code class="docutils literal notranslate"><span class="pre">asabstractarray</span></code>,
and so by subclassing it you’re saying that if some code works for
<code class="docutils literal notranslate"><span class="pre">ndarray</span></code>s but breaks for your class, then you’re willing to accept
bug reports on that.</p>
<p>To start with, we should declare <code class="docutils literal notranslate"><span class="pre">__array_ufunc__</span></code> to be an abstract
method, and add the <code class="docutils literal notranslate"><span class="pre">NDArrayOperatorsMixin</span></code> methods as mixin
methods.</p>
<p>Declaring <code class="docutils literal notranslate"><span class="pre">astype</span></code> as an <code class="docutils literal notranslate"><span class="pre">@abstractmethod</span></code> probably makes sense as
well, since it’s used by <code class="docutils literal notranslate"><span class="pre">asabstractarray</span></code>. We might also want to go
ahead and add some basic attributes like <code class="docutils literal notranslate"><span class="pre">ndim</span></code>, <code class="docutils literal notranslate"><span class="pre">shape</span></code>,
<code class="docutils literal notranslate"><span class="pre">dtype</span></code>.</p>
<p>Adding new abstract methods will be a bit tricky, because ABCs enforce
these at subclass time; therefore, simply adding a new
<cite>@abstractmethod</cite> will be a backwards compatibility break. If this
becomes a problem then we can use some hacks to implement an
<cite>@upcoming_abstractmethod</cite> decorator that only issues a warning if the
method is missing, and treat it like a regular deprecation cycle. (In
this case, the thing we’d be deprecating is “support for abstract
arrays that are missing feature X”.)</p>
</section>
<section id="naming">
<h3>Naming<a class="headerlink" href="#naming" title="Link to this heading">#</a></h3>
<p>The name of the ABC doesn’t matter too much, because it will only be
referenced rarely and in relatively specialized situations. The name
of the function matters a lot, because most existing instances of
<code class="docutils literal notranslate"><span class="pre">asarray</span></code> should be replaced by this, and in the future it’s what
everyone should be reaching for by default unless they have a specific
reason to use <code class="docutils literal notranslate"><span class="pre">asarray</span></code> instead. This suggests that its name really
should be <em>shorter</em> and <em>more memorable</em> than <code class="docutils literal notranslate"><span class="pre">asarray</span></code>… which
is difficult. I’ve used <code class="docutils literal notranslate"><span class="pre">asabstractarray</span></code> in this draft, but I’m not
really happy with it, because it’s too long and people are unlikely to
start using it by habit without endless exhortations.</p>
<p>One option would be to actually change <code class="docutils literal notranslate"><span class="pre">asarray</span></code>'s semantics so
that <em>it</em> passes through <code class="docutils literal notranslate"><span class="pre">AbstractArray</span></code> objects unchanged. But I’m
worried that there may be a lot of code out there that calls
<code class="docutils literal notranslate"><span class="pre">asarray</span></code> and then passes the result into some C function that
doesn’t do any further type checking (because it knows that its caller
has already used <code class="docutils literal notranslate"><span class="pre">asarray</span></code>). If we allow <code class="docutils literal notranslate"><span class="pre">asarray</span></code> to return
<code class="docutils literal notranslate"><span class="pre">AbstractArray</span></code> objects, and then someone calls one of these C
wrappers and passes it an <code class="docutils literal notranslate"><span class="pre">AbstractArray</span></code> object like a sparse
array, then they’ll get a segfault. Right now, in the same situation,
<code class="docutils literal notranslate"><span class="pre">asarray</span></code> will instead invoke the object’s <code class="docutils literal notranslate"><span class="pre">__array__</span></code> method, or
use the buffer interface to make a view, or pass through an array with
object dtype, or raise an error, or similar. Probably none of these
outcomes are actually desirable in most cases, so maybe making it a
segfault instead would be OK? But it’s dangerous given that we don’t
know how common such code is. OTOH, if we were starting from scratch
then this would probably be the ideal solution.</p>
<p>We can’t use <code class="docutils literal notranslate"><span class="pre">asanyarray</span></code> or <code class="docutils literal notranslate"><span class="pre">array</span></code>, since those are already
taken.</p>
<p>Any other ideas? <code class="docutils literal notranslate"><span class="pre">np.cast</span></code>, <code class="docutils literal notranslate"><span class="pre">np.coerce</span></code>?</p>
</section>
</section>
<section id="implementation">
<h2>Implementation<a class="headerlink" href="#implementation" title="Link to this heading">#</a></h2>
<ol class="arabic simple">
<li><p>Rename <code class="docutils literal notranslate"><span class="pre">NDArrayOperatorsMixin</span></code> to <code class="docutils literal notranslate"><span class="pre">AbstractArray</span></code> (leaving
behind an alias for backwards compatibility) and make it an ABC.</p></li>
<li><p>Add <code class="docutils literal notranslate"><span class="pre">asabstractarray</span></code> (or whatever we end up calling it), and
probably a C API equivalent.</p></li>
<li><p>Begin migrating NumPy internal functions to using
<code class="docutils literal notranslate"><span class="pre">asabstractarray</span></code> where appropriate.</p></li>
</ol>
</section>
<section id="backward-compatibility">
<h2>Backward compatibility<a class="headerlink" href="#backward-compatibility" title="Link to this heading">#</a></h2>
<p>This is purely a new feature, so there are no compatibility issues.
(Unless we decide to change the semantics of <code class="docutils literal notranslate"><span class="pre">asarray</span></code> itself.)</p>
</section>
<section id="rejected-alternatives">
<h2>Rejected alternatives<a class="headerlink" href="#rejected-alternatives" title="Link to this heading">#</a></h2>
<p>One suggestion that has come up is to define multiple abstract classes
for different subsets of the array interface. Nothing in this proposal
stops either NumPy or third-parties from doing this in the future, but
it’s very difficult to guess ahead of time which subsets would be
useful. Also, “the full ndarray interface” is something that existing
libraries are written to expect (because they work with actual
ndarrays) and test (because they test with actual ndarrays), so it’s
by far the easiest place to start.</p>
</section>
<section id="links-to-discussion">
<h2>Links to discussion<a class="headerlink" href="#links-to-discussion" title="Link to this heading">#</a></h2>
<ul class="simple">
<li><p><a class="reference external" href="https://mail.python.org/pipermail/numpy-discussion/2018-March/077767.html">https://mail.python.org/pipermail/numpy-discussion/2018-March/077767.html</a></p></li>
</ul>
</section>
<section id="appendix-benchmark-script">
<h2>Appendix: benchmark script<a class="headerlink" href="#appendix-benchmark-script" title="Link to this heading">#</a></h2>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span><span class="w"> </span><span class="nn">perf</span>
<span class="kn">import</span><span class="w"> </span><span class="nn">abc</span>
<span class="kn">import</span><span class="w"> </span><span class="nn">numpy</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">np</span>
<span class="k">class</span><span class="w"> </span><span class="nc">NotArray</span><span class="p">:</span>
<span class="k">pass</span>
<span class="k">class</span><span class="w"> </span><span class="nc">AttrArray</span><span class="p">:</span>
<span class="n">__array_implementer__</span> <span class="o">=</span> <span class="kc">True</span>
<span class="k">class</span><span class="w"> </span><span class="nc">ArrayBase</span><span class="p">(</span><span class="n">abc</span><span class="o">.</span><span class="n">ABC</span><span class="p">):</span>
<span class="k">pass</span>
<span class="k">class</span><span class="w"> </span><span class="nc">ABCArray1</span><span class="p">(</span><span class="n">ArrayBase</span><span class="p">):</span>
<span class="k">pass</span>
<span class="k">class</span><span class="w"> </span><span class="nc">ABCArray2</span><span class="p">:</span>
<span class="k">pass</span>
<span class="n">ArrayBase</span><span class="o">.</span><span class="n">register</span><span class="p">(</span><span class="n">ABCArray2</span><span class="p">)</span>
<span class="n">not_array</span> <span class="o">=</span> <span class="n">NotArray</span><span class="p">()</span>
<span class="n">attr_array</span> <span class="o">=</span> <span class="n">AttrArray</span><span class="p">()</span>
<span class="n">abc_array_1</span> <span class="o">=</span> <span class="n">ABCArray1</span><span class="p">()</span>
<span class="n">abc_array_2</span> <span class="o">=</span> <span class="n">ABCArray2</span><span class="p">()</span>
<span class="c1"># Make sure ABC cache is primed</span>
<span class="nb">isinstance</span><span class="p">(</span><span class="n">not_array</span><span class="p">,</span> <span class="n">ArrayBase</span><span class="p">)</span>
<span class="nb">isinstance</span><span class="p">(</span><span class="n">abc_array_1</span><span class="p">,</span> <span class="n">ArrayBase</span><span class="p">)</span>
<span class="nb">isinstance</span><span class="p">(</span><span class="n">abc_array_2</span><span class="p">,</span> <span class="n">ArrayBase</span><span class="p">)</span>
<span class="n">runner</span> <span class="o">=</span> <span class="n">perf</span><span class="o">.</span><span class="n">Runner</span><span class="p">()</span>
<span class="k">def</span><span class="w"> </span><span class="nf">t</span><span class="p">(</span><span class="n">name</span><span class="p">,</span> <span class="n">statement</span><span class="p">):</span>
<span class="n">runner</span><span class="o">.</span><span class="n">timeit</span><span class="p">(</span><span class="n">name</span><span class="p">,</span> <span class="n">statement</span><span class="p">,</span> <span class="nb">globals</span><span class="o">=</span><span class="nb">globals</span><span class="p">())</span>
<span class="n">t</span><span class="p">(</span><span class="s2">"np.asarray([])"</span><span class="p">,</span> <span class="s2">"np.asarray([])"</span><span class="p">)</span>
<span class="n">arrobj</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">array</span><span class="p">([])</span>
<span class="n">t</span><span class="p">(</span><span class="s2">"np.asarray(arrobj)"</span><span class="p">,</span> <span class="s2">"np.asarray(arrobj)"</span><span class="p">)</span>
<span class="n">t</span><span class="p">(</span><span class="s2">"attr, False"</span><span class="p">,</span>
<span class="s2">"getattr(not_array, '__array_implementer__', False)"</span><span class="p">)</span>
<span class="n">t</span><span class="p">(</span><span class="s2">"attr, True"</span><span class="p">,</span>
<span class="s2">"getattr(attr_array, '__array_implementer__', False)"</span><span class="p">)</span>
<span class="n">t</span><span class="p">(</span><span class="s2">"ABC, False"</span><span class="p">,</span> <span class="s2">"isinstance(not_array, ArrayBase)"</span><span class="p">)</span>
<span class="n">t</span><span class="p">(</span><span class="s2">"ABC, True, via inheritance"</span><span class="p">,</span> <span class="s2">"isinstance(abc_array_1, ArrayBase)"</span><span class="p">)</span>
<span class="n">t</span><span class="p">(</span><span class="s2">"ABC, True, via register"</span><span class="p">,</span> <span class="s2">"isinstance(abc_array_2, ArrayBase)"</span><span class="p">)</span>
</pre></div>
</div>
</section>
<section id="copyright">
<h2>Copyright<a class="headerlink" href="#copyright" title="Link to this heading">#</a></h2>
<p>This document has been placed in the public domain.</p>
</section>
</section>
</article>
</div>
<dialog id="pst-secondary-sidebar-modal"></dialog>
<div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
<div class="sidebar-secondary-item">
<div
id="pst-page-navigation-heading-2"
class="page-toc tocsection onthispage">
<i class="fa-solid fa-list"></i> On this page
</div>
<nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
<ul class="visible nav section-nav flex-column">
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#abstract">Abstract</a></li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#detailed-description">Detailed description</a><ul class="nav section-nav flex-column">
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#declaring-that-an-object-can-quack-like-an-array">Declaring that an object can quack like an array</a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#specification-of-asabstractarray">Specification of <code class="docutils literal notranslate"><span class="pre">asabstractarray</span></code></a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#what-exactly-are-you-promising-if-you-inherit-from-abstractarray">What exactly are you promising if you inherit from <code class="docutils literal notranslate"><span class="pre">AbstractArray</span></code>?</a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#naming">Naming</a></li>
</ul>
</li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#implementation">Implementation</a></li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#backward-compatibility">Backward compatibility</a></li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#rejected-alternatives">Rejected alternatives</a></li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#links-to-discussion">Links to discussion</a></li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#appendix-benchmark-script">Appendix: benchmark script</a></li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#copyright">Copyright</a></li>
</ul>
</nav></div>
</div></div>
</div>
<footer class="bd-footer-content">
</footer>
</main>
</div>
</div>
<!-- Scripts loaded after <body> so the DOM is not blocked -->
<script defer src="_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
<script defer src="_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
<footer class="bd-footer">
<div class="bd-footer__inner bd-page-width">
<div class="footer-items__start">
<div class="footer-item">
<p class="copyright">
© Copyright 2017-2025, NumPy Developers.
<br/>
</p>
</div>
<div class="footer-item">
<p class="sphinx-version">
Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 7.2.6.
<br/>
</p>
</div>
</div>
<div class="footer-items__end">
<div class="footer-item">
<p class="theme-version">
<!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
</p></div>
</div>
</div>
</footer>
</body>
</html>