-
Notifications
You must be signed in to change notification settings - Fork 6
/
nep-0053-c-abi-evolution.html
873 lines (684 loc) · 46.7 KB
/
nep-0053-c-abi-evolution.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
<!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 53 — Evolving the NumPy C-API for NumPy 2.0 — 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=26a4bc78f4c0ddb94549" rel="stylesheet" />
<link href="_static/styles/pydata-sphinx-theme.css?digest=26a4bc78f4c0ddb94549" rel="stylesheet" />
<link rel="stylesheet" type="text/css" href="_static/pygments.css?v=fa44fd50" />
<!-- So that users can add custom icons -->
<script src="_static/scripts/fontawesome.js?digest=26a4bc78f4c0ddb94549"></script>
<!-- Pre-loaded scripts that we'll load fully later -->
<link rel="preload" as="script" href="_static/scripts/bootstrap.js?digest=26a4bc78f4c0ddb94549" />
<link rel="preload" as="script" href="_static/scripts/pydata-sphinx-theme.js?digest=26a4bc78f4c0ddb94549" />
<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-0053-c-abi-evolution';</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 54 — SIMD infrastructure evolution: adopting Google Highway when moving to C++?" href="nep-0054-simd-cpp-highway.html" />
<link rel="prev" title="NEP 43 — Enhancing the extensibility of UFuncs" href="nep-0043-extensible-ufuncs.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="Nov 29, 2024"/>
</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.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">
Wish list
</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">
Wish list
</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>
<li class="toctree-l1"><a class="reference external" href="https://github.com/numpy/numpy/issues?q=is%3Aopen+is%3Aissue+label%3A%2223+-+Wish+List%22">Wish list</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>
</ul>
</details></li>
<li class="toctree-l1 current active has-children"><a class="reference internal" href="open.html">Open NEPs (under consideration)</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"><a class="reference internal" href="nep-0043-extensible-ufuncs.html">NEP 43 — Enhancing the extensibility of UFuncs</a></li>
<li class="toctree-l2 current active"><a class="current reference internal" href="#">NEP 53 — Evolving the NumPy C-API for NumPy 2.0</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="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 has-children"><a class="reference internal" href="rejected.html">Rejected and Withdrawn 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-0016-abstract-array.html">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>
<div id="rtd-footer-container"></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="open.html" class="nav-link">Open NEPs (under consideration)</a></li>
<li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">NEP 53 — Evolving the NumPy C-API for NumPy 2.0</span></li>
</ul>
</nav>
</div>
</div>
</div>
</div>
<div id="searchbox"></div>
<article class="bd-article">
<section id="nep-53-evolving-the-numpy-c-api-for-numpy-2-0">
<span id="nep53"></span><h1>NEP 53 — Evolving the NumPy C-API for NumPy 2.0<a class="headerlink" href="#nep-53-evolving-the-numpy-c-api-for-numpy-2-0" 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>Sebastian Berg <<a class="reference external" href="mailto:sebastianb%40nvidia.com">sebastianb<span>@</span>nvidia<span>.</span>com</a>></p>
</dd>
<dt class="field-even">Status<span class="colon">:</span></dt>
<dd class="field-even"><p>Draft</p>
</dd>
<dt class="field-odd">Type<span class="colon">:</span></dt>
<dd class="field-odd"><p>Standard</p>
</dd>
<dt class="field-even">Created<span class="colon">:</span></dt>
<dd class="field-even"><p>2022-04-10</p>
</dd>
</dl>
<section id="abstract">
<h2>Abstract<a class="headerlink" href="#abstract" title="Link to this heading">#</a></h2>
<p>The NumPy C-API is used in downstream projects (often through Cython)
to extend NumPy functionality. Supporting these packages generally means
that it is slow to evolve our C-API and some changes are not possible in a
normal NumPy release because NumPy must guarantee backwards compatibility:
A downstream package compiled against an old NumPy version (e.g. 1.17)
will generally work with a new NumPy version (e.g. 1.25).</p>
<p>A NumPy 2.0 release allows to <em>partially</em> break this promise:
We can accept that a SciPy version compiled with NumPy 1.17 (e.g. SciPy 1.10)
will <em>not</em> work with NumPy 2.0.
However, it must still be easy to create a single SciPy binary that is
compatible with both NumPy 1.x and NumPy 2.0.</p>
<p>Given these constraints this NEP outlines a path forward to allow large changes
to our C-API. Similar to Python API changes proposed for NumPy 2.0 the NEP
aims to allow changes to an extend that <em>most</em> downstream packages are expected
to need no or only minor code changes.</p>
<p>The implementation of this NEP consists would consist of two steps:</p>
<ol class="arabic simple">
<li><p>As part of a general improvement, starting with NumPy 1.25 building with
NumPy will by default export an older API version to allow backwards
compatible builds with the newest available NumPy version.
(New API is not available unless opted-in.)</p></li>
<li><p>The NumPy 2.0 will:</p>
<ul class="simple">
<li><p>require recompilation of downstream packages against NumPy 2.0 to be
compatible with NumPy 2.0.</p></li>
<li><p>need a <code class="docutils literal notranslate"><span class="pre">numpy2_compat</span></code> as a dependency when running on NumPy 1.x.</p></li>
<li><p>require some downstream code changes to adapt to changed API.</p></li>
</ul>
</li>
</ol>
</section>
<section id="motivation-and-scope">
<h2>Motivation and scope<a class="headerlink" href="#motivation-and-scope" title="Link to this heading">#</a></h2>
<p>The NumPy API conists of more than 300 functions and numerous macros.
Many of these are outdated: some were only ever used within NumPy,
exist only for compatibility with NumPy’s predecessors, or have no or only
a single known downstream user (i.e. SciPy).</p>
<p>Further, many structs used by NumPy have always been public making it
impossible to change them outside of a major release.
Some changes have been planned for years and were the reason for
<code class="docutils literal notranslate"><span class="pre">NPY_NO_DEPRECATED_API</span></code> and further deprecations as explained in
<a class="reference external" href="https://numpy.org/devdocs/reference/c-api/deprecations.html#c-api-deprecations" title="(in NumPy v2.3.dev0)"><span>C API deprecations</span></a>.</p>
<p>While we probably have little reason to change the layout of the array struct
(<code class="docutils literal notranslate"><span class="pre">PyArrayObject_fields</span></code>) for example the development and improvement of
dtypes would be made easier by changing the <cite>PyArray_Descr</cite> struct.</p>
<p>This NEP proposes a few concrete changes to our C-API mainly as examples.
However, more changes will be handled on a case-by-case basis, and we do not
aim to provide a full list of changes in this NEP.</p>
<section id="adding-state-is-out-of-scope">
<h3>Adding state is out of scope<a class="headerlink" href="#adding-state-is-out-of-scope" title="Link to this heading">#</a></h3>
<p>New developments such as CPython’s support for subinterpreters and the
HPy API may require the NumPy C-API to evolve in a way that may require
(or at least prefer) state to be passed in.</p>
<p>As of now, we do not aim to include changes for this here. We cannot expect
users to do large code updates to pass in for example an <code class="docutils literal notranslate"><span class="pre">HPy</span></code> context
to many NumPy functions.</p>
<p>While we could introduce a second API for this purpose in NumPy 2.0,
we expect that this is unnecessary and that the provisions introduced here:</p>
<ul class="simple">
<li><p>the ability to compile with the newest NumPy version but be compatible with
older versions,</p></li>
<li><p>and the possibility of updating a <code class="docutils literal notranslate"><span class="pre">numpy2_compat</span></code> package.</p></li>
</ul>
<p>should allow to add such an API also in a minor release.</p>
</section>
</section>
<section id="usage-and-impact">
<h2>Usage and impact<a class="headerlink" href="#usage-and-impact" title="Link to this heading">#</a></h2>
<section id="backwards-compatible-builds">
<h3>Backwards compatible builds<a class="headerlink" href="#backwards-compatible-builds" title="Link to this heading">#</a></h3>
<p>Backwards compatible builds will be described in more details in the
documentation.
Briefly, we will allow users to use a definition like:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="c1">#define NPY_TARGET_VERSION NPY_1_22_API_VERSION</span>
</pre></div>
</div>
<p>to select the version they wish to compile for (lowest version to be
compatible with).
By default the backwards compatibility will be such that the resulting binary
is compatible with the oldest NumPy version which supports the same
version of Python: NumPy 1.19.x was the first to support Python 3.9 and
NumPy 1.25 supports Python 3.9 or greater, so NumPy 1.25 defaults to 1.19
compatibility.
Thus, users of <em>new</em> API may be required to add the define,
but users of who want to be compatible with older versions need not do
anything unless they wish to have exceptionally long compatibility.</p>
<p>The API additions in the past years were so limited that such a change
should be necessary at most for a hand-full of users worldwide.</p>
<p>This mechanism is much the same as the <a class="reference external" href="https://docs.python.org/3/c-api/stable.html">Python limited API</a> since NumPy’s
C-API has a similar need for ABI stability.</p>
</section>
<section id="breaking-the-c-api-and-changing-the-abi">
<h3>Breaking the C-API and changing the ABI<a class="headerlink" href="#breaking-the-c-api-and-changing-the-abi" title="Link to this heading">#</a></h3>
<p>NumPy has too many functions, many of which are aliases. The following
lists <em>examples</em> of things we plan to remove and users will have to adapt
to be compatible with NumPy 2.0:</p>
<ul class="simple">
<li><p><code class="docutils literal notranslate"><span class="pre">PyArray_Mean</span></code> and <code class="docutils literal notranslate"><span class="pre">PyArray_Std</span></code> are untested implementation similar to
<code class="docutils literal notranslate"><span class="pre">arr.mean()</span></code> and <code class="docutils literal notranslate"><span class="pre">arr.std()</span></code>. We are planning on removing these as they
can be replaced with method calls relatively easily.</p></li>
<li><p>The <code class="docutils literal notranslate"><span class="pre">MapIter</span></code> set of API functions (and struct) allows to implement
advanced indexing like semantics downstream. There was a single <em>historic</em>
known user of this (theano) and the use-case would be faster and easier to
implement in a different way. The API is complicated, requires reaching
deep into NumPy to be useful and its exposure makes the implementation
more difficult. Unless new important use cases are found, we propose to
remove it.</p></li>
</ul>
<p>An example for an ABI change is to change the layout of <code class="docutils literal notranslate"><span class="pre">PyArray_Descr</span></code>
(the struct of <code class="docutils literal notranslate"><span class="pre">np.dtype</span></code> instances) to allow a larger maximum itemsize and
new flags (useful for future custom user DTypes).
For this specific change, users who access the structs fields directly
will have to change their code. A downstream search shows that this should
not be very common, the main impacts are:</p>
<ul class="simple">
<li><p>Access of the <code class="docutils literal notranslate"><span class="pre">descr->elsize</span></code> field (and others) would have to be replaced
with a macro’s like <code class="docutils literal notranslate"><span class="pre">PyDataType_ITEMSIZE(descr)</span></code> (NumPy may include a
version check when needed).</p></li>
<li><p>Implementers of user defined dtypes, will have to change a few lines of code
and luckily, there are very few of such user defined dtypes.
(The details are that we rename the struct to <code class="docutils literal notranslate"><span class="pre">PyArray_DescrProto</span></code> for the
static definition and fetch the actual instance from NumPy explicitly.)</p></li>
</ul>
<p>A last example is increasing <code class="docutils literal notranslate"><span class="pre">NPY_MAXDIMS</span></code> to <code class="docutils literal notranslate"><span class="pre">64</span></code>.
<code class="docutils literal notranslate"><span class="pre">NPY_MAXDIMS</span></code> is mainly used to statically allocate scratch space:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">func</span><span class="p">(</span><span class="n">PyArrayObject</span> <span class="o">*</span><span class="n">arr</span><span class="p">)</span> <span class="p">{</span>
<span class="n">npy_intp</span> <span class="n">shape</span><span class="p">[</span><span class="n">NPY_MAXDIMS</span><span class="p">];</span>
<span class="o">/*</span> <span class="n">Work</span> <span class="k">with</span> <span class="n">a</span> <span class="n">shape</span> <span class="ow">or</span> <span class="n">strides</span> <span class="kn">from</span> <span class="nn">the</span> <span class="n">array</span> <span class="o">*/</span>
<span class="p">}</span>
</pre></div>
</div>
<p>If NumPy changed it to 64 in a minor release, this would lead to undefined
behavior if the code was compiled with <code class="docutils literal notranslate"><span class="pre">NPY_MAXDIMS=32</span></code> but a 40 dimensional
array is passed in.
But the larger value is also a correct maximum on previous versions of NumPy
making it generally safe for NumPy 2.0 change.
(One can imagine code that wants to know the actual runtime value.
We have not seen such code in practice, but it would need to be adjusted.)</p>
</section>
<section id="impact-on-cython-users">
<h3>Impact on Cython users<a class="headerlink" href="#impact-on-cython-users" title="Link to this heading">#</a></h3>
<p>Cython users may use the NumPy C-API via <code class="docutils literal notranslate"><span class="pre">cimport</span> <span class="pre">numpy</span> <span class="pre">as</span> <span class="pre">cnp</span></code>.
Due to the uncertainty of Cython development, there are two scenarios for
impact on Cython users.</p>
<p>If Cython 3 can be relied on, Cython users would be impacted <em>less</em> than C-API
users, because Cython 3 allows us to hide struct layout changes (i.e. changes
to <code class="docutils literal notranslate"><span class="pre">PyArray_Descr</span></code>).
If this is not the case and we must support Cython 0.29.x (which is the historic branch
before Cython 3), then Cython users will also have to use a function/macro like
<code class="docutils literal notranslate"><span class="pre">PyDataType_ITEMSIZE()</span></code> (or use the Python object). This is unfortunately less
typical in Cython code, but also unlikely to be a common pattern for dtype struct
fields/attributes.</p>
<p>A further impact is that some future API additions such as new classes may
need to placed in a distinct <code class="docutils literal notranslate"><span class="pre">.pyd</span></code> file to avoid Cython generating code
that would fail on older NumPy versions.</p>
</section>
<section id="end-user-and-packaging-impact">
<h3>End-user and packaging impact<a class="headerlink" href="#end-user-and-packaging-impact" title="Link to this heading">#</a></h3>
<p>Packaging in a way that is compatible with NumPy 2.0 will require a
recompilation of downstream libraries that rely on the NumPy C-API.
This may take some time, although hopefully the process will start before
NumPy 2.0 is itself released.</p>
<p>Further, to allow bigger changes more easily in NumPy 2.0, we expect to
create a <code class="docutils literal notranslate"><span class="pre">numpy2_compat</span></code> package.
When a library is build with NumPy 2.0 but wants to support NumPy 1.x it will
have to depend on <code class="docutils literal notranslate"><span class="pre">numpy2_compat</span></code>. End-users should not need to be aware
of this dependency and an informative error can be raised when the module
is missing.</p>
</section>
<section id="some-new-api-can-be-backported">
<h3>Some new API can be backported<a class="headerlink" href="#some-new-api-can-be-backported" title="Link to this heading">#</a></h3>
<p>One large advantage of allowing users to compile with the newest version of
NumPy is that in some cases we will be able to backport new API.
Some new API functions can be written in terms of old ones or included
directly.</p>
<div class="admonition note">
<p class="admonition-title">Note</p>
<p>It may be possible to make functions public that were present but
private in NumPy 1.x public via the compatible <code class="docutils literal notranslate"><span class="pre">numpy2_compat</span></code> package.</p>
</div>
<p>This means that at some new API additions could be made available to
downstreams users faster. They would require a new NumPy version for
<em>compilation</em> but their wheels can be backwards compatible with earlier
versions.</p>
</section>
</section>
<section id="implementation">
<h2>Implementation<a class="headerlink" href="#implementation" title="Link to this heading">#</a></h2>
<p>The first part of implementation (allowing building for an earlier API version)
is very straight forward since the NumPy C-API evolved slowly for
many years.
Some struct fields will be hidden by default and functions introduced in a
more recent version will be marked and hidden unless the
user opted in to a newer API version.
An implementation can be found in the <a class="reference external" href="https://github.com/numpy/numpy/pull/23528">PR 23528</a>.</p>
<p>The second part is mainly about identifying and implementing the desired
changes in a way that backwards compatibility will not be broken and API
breaks remain manageable for downstream libraries.
Every change we do must have a brief note on how to adapt to the
API change (i.e. alternative functions).</p>
<section id="numpy-2-compatibility-and-api-table-changes">
<h3>NumPy 2 compatibility and API table changes<a class="headerlink" href="#numpy-2-compatibility-and-api-table-changes" title="Link to this heading">#</a></h3>
<p>To allow changing the API table, NumPy 2.0 would ship a different table than
NumPy 1.x (a table is a list of functions and symbols).</p>
<p>For compatibility we would need to translate the 1.x table to the 2.0 table.
This could be done in headers only in theory, but this seems unwieldy.
We thus propose to add a <code class="docutils literal notranslate"><span class="pre">numpy2_compat</span></code> package. This package’s main
purpose would be to provide a translation of the 1.x table to the 2.x one
in a single place (filling in any necessary blanks).</p>
<p>Introducing this package solves the “transition” issue because it allows
a user to:</p>
<ul class="simple">
<li><p>Install a SciPy version that is compatible with 2.0 and 1.x</p></li>
<li><p>and keep using NumPy 1.x because of other packages they are using are not
yet compatible.</p></li>
</ul>
<p>The import of <code class="docutils literal notranslate"><span class="pre">numpy2_compat</span></code> (and an error when it is missing) will be
inserted by the NumPy headers as part of the <code class="docutils literal notranslate"><span class="pre">import_array()</span></code> call.</p>
</section>
</section>
<section id="alternatives">
<h2>Alternatives<a class="headerlink" href="#alternatives" title="Link to this heading">#</a></h2>
<p>There are always possibilities to decide not to do certain changes (e.g. due
to downstream users noting their continued need for it). For example, the
function <code class="docutils literal notranslate"><span class="pre">PyArray_Mean</span></code> could be replaced by one to call <code class="docutils literal notranslate"><span class="pre">array.mean()</span></code>
if necessary.</p>
<p>The NEP proposes to allow larger changes to our API table by introducing a
compatibility package <code class="docutils literal notranslate"><span class="pre">numpy2_compat</span></code>.
We could do many changes without introducing such a package.</p>
<p>The default API version could be chosen to be older or as the current one.
An older version would be aimed at libraries who want a larger compatibility
than NEP 29 suggests.
Choosing the current would default to removing unnecessary compatibility shims
for users who do not distribute wheels.
The suggested default chooses to favors libraries that distribute wheels and
wish a compatibility range similar to NEP 29. This is because compatibility
shims should be light-weight and we expect few libraries require a longer
compatibility.</p>
</section>
<section id="backward-compatibility">
<h2>Backward compatibility<a class="headerlink" href="#backward-compatibility" title="Link to this heading">#</a></h2>
<p>As mentioned above backwards compatibility is achieved by:</p>
<ol class="arabic simple">
<li><p>Forcing downstream to recompile with NumPy 2.0</p></li>
<li><p>Providing a <code class="docutils literal notranslate"><span class="pre">numpy2_compat</span></code> library.</p></li>
</ol>
<p>But relies on users to adapt to changed C-API as described in the Usage and
Impact section.</p>
</section>
<section id="discussion">
<h2>Discussion<a class="headerlink" href="#discussion" title="Link to this heading">#</a></h2>
<ul class="simple">
<li><p><a class="github reference external" href="https://github.com/numpy/numpy/issues/5888">numpy/numpy#5888</a> brought up previously that it
would be helpful to allow exporting of an older API version in our headers.
This was never implemented, instead we relied on <a class="reference external" href="https://github.com/scipy/oldest-supported-numpy">oldest-support-numpy</a>.</p></li>
<li><p>A first draft of this proposal was presented at the NumPy 2.0 planning
meeting 2023-04-03.</p></li>
</ul>
</section>
<section id="references-and-footnotes">
<h2>References and footnotes<a class="headerlink" href="#references-and-footnotes" title="Link to this heading">#</a></h2>
<aside class="footnote-list brackets">
<aside class="footnote brackets" id="id1" role="doc-footnote">
<span class="label"><span class="fn-bracket">[</span><a role="doc-backlink" href="#id2">1</a><span class="fn-bracket">]</span></span>
<p>Each NEP must either be explicitly labeled as placed in the public domain (see
this NEP as an example) or licensed under the <a class="reference external" href="https://www.opencontent.org/openpub/">Open Publication License</a>.</p>
</aside>
</aside>
</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. <a class="footnote-reference brackets" href="#id1" id="id2" role="doc-noteref"><span class="fn-bracket">[</span>1<span class="fn-bracket">]</span></a></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="#motivation-and-scope">Motivation and scope</a><ul class="nav section-nav flex-column">
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#adding-state-is-out-of-scope">Adding state is out of scope</a></li>
</ul>
</li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#usage-and-impact">Usage and impact</a><ul class="nav section-nav flex-column">
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#backwards-compatible-builds">Backwards compatible builds</a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#breaking-the-c-api-and-changing-the-abi">Breaking the C-API and changing the ABI</a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#impact-on-cython-users">Impact on Cython users</a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#end-user-and-packaging-impact">End-user and packaging impact</a></li>
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#some-new-api-can-be-backported">Some new API can be backported</a></li>
</ul>
</li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#implementation">Implementation</a><ul class="nav section-nav flex-column">
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#numpy-2-compatibility-and-api-table-changes">NumPy 2 compatibility and API table changes</a></li>
</ul>
</li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#alternatives">Alternatives</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="#discussion">Discussion</a></li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#references-and-footnotes">References and footnotes</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=26a4bc78f4c0ddb94549"></script>
<script defer src="_static/scripts/pydata-sphinx-theme.js?digest=26a4bc78f4c0ddb94549"></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 2017-2024, 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">
Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.0.
</p></div>
</div>
</div>
</footer>
</body>
</html>