forked from flaggo/python3-source-code-analysis
-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathobject.html
More file actions
763 lines (462 loc) · 55.9 KB
/
object.html
File metadata and controls
763 lines (462 loc) · 55.9 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
<!DOCTYPE HTML>
<html lang="zh-hans" >
<head>
<meta charset="UTF-8">
<meta content="text/html; charset=utf-8" http-equiv="Content-Type">
<title>Python 对象初探 · Python 3 源码分析</title>
<meta http-equiv="X-UA-Compatible" content="IE=edge" />
<meta name="description" content="">
<meta name="generator" content="GitBook 3.2.3">
<meta name="author" content="Prodesire">
<link rel="stylesheet" href="../gitbook/style.css">
<link rel="stylesheet" href="../gitbook/gitbook-plugin-search-plus/search.css">
<link rel="stylesheet" href="../gitbook/gitbook-plugin-splitter/splitter.css">
<link rel="stylesheet" href="../gitbook/gitbook-plugin-tbfed-pagefooter/footer.css">
<link rel="stylesheet" href="../gitbook/gitbook-plugin-prism/prism-okaidia.css">
<link rel="stylesheet" href="../gitbook/gitbook-plugin-page-toc-button/plugin.css">
<link rel="stylesheet" href="../gitbook/gitbook-plugin-back-to-top-button/plugin.css">
<link rel="stylesheet" href="../gitbook/gitbook-plugin-highlight/website.css">
<link rel="stylesheet" href="../gitbook/gitbook-plugin-fontsettings/website.css">
<meta name="HandheldFriendly" content="true"/>
<meta name="viewport" content="width=device-width, initial-scale=1, user-scalable=no">
<meta name="apple-mobile-web-app-capable" content="yes">
<meta name="apple-mobile-web-app-status-bar-style" content="black">
<link rel="apple-touch-icon-precomposed" sizes="152x152" href="../gitbook/images/apple-touch-icon-precomposed-152.png">
<link rel="shortcut icon" href="../gitbook/images/favicon.ico" type="image/x-icon">
<link rel="next" href="long-object.html" />
<link rel="prev" href="../preface/modify-code.html" />
<style>
@media only screen and (max-width: 640px) {
.book-header .hidden-mobile {
display: none;
}
}
</style>
<script>
window["gitbook-plugin-github-buttons"] = {"repo":"flaggo/python3-source-code-analysis","types":["star"],"size":"small"};
</script>
</head>
<body>
<div class="book">
<div class="book-summary">
<div id="book-search-input" role="search">
<input type="text" placeholder="输入并搜索" />
</div>
<nav role="navigation">
<ul class="summary">
<li>
<a href="https://github.com/flaggo/python3-source-code-analysis" target="_blank" class="custom-link">主页</a>
</li>
<li class="divider"></li>
<li class="header">第 1 部分：序章</li>
<li class="chapter " data-level="1.1" data-path="../">
<a href="../">
<b>1.1.</b>
前言
</a>
</li>
<li class="chapter " data-level="1.2" data-path="../preface/code-organization.html">
<a href="../preface/code-organization.html">
<b>1.2.</b>
Python 源代码的组织
</a>
</li>
<li class="chapter " data-level="1.3" data-path="../preface/windows-build.html">
<a href="../preface/windows-build.html">
<b>1.3.</b>
Windows 环境下编译 Python
</a>
</li>
<li class="chapter " data-level="1.4" data-path="../preface/unix-linux-build.html">
<a href="../preface/unix-linux-build.html">
<b>1.4.</b>
UNIX/Linux 环境下编译 Python
</a>
</li>
<li class="chapter " data-level="1.5" data-path="../preface/modify-code.html">
<a href="../preface/modify-code.html">
<b>1.5.</b>
修改 Python 源码
</a>
</li>
<li class="header">第 2 部分：Python 内建对象</li>
<li class="chapter active" data-level="2.1" data-path="object.html">
<a href="object.html">
<b>2.1.</b>
Python 对象初探
</a>
</li>
<li class="chapter " data-level="2.2" data-path="long-object.html">
<a href="long-object.html">
<b>2.2.</b>
Python 整数对象
</a>
</li>
<li class="chapter " data-level="2.3" data-path="string-object.html">
<a href="string-object.html">
<b>2.3.</b>
Python 字符串 对象
</a>
</li>
<li class="chapter " data-level="2.4" data-path="list-object.html">
<a href="list-object.html">
<b>2.4.</b>
Python List 对象
</a>
</li>
<li class="chapter " data-level="2.5" data-path="dict-object.html">
<a href="dict-object.html">
<b>2.5.</b>
Python Dict 对象
</a>
</li>
<li class="chapter " data-level="2.6" data-path="set-object.html">
<a href="set-object.html">
<b>2.6.</b>
Python Set 对象
</a>
</li>
<li class="chapter " data-level="2.7" data-path="simple-implementation.html">
<a href="simple-implementation.html">
<b>2.7.</b>
实现简版 Python
</a>
</li>
<li class="header">第 3 部分：Python 虚拟机</li>
<li class="divider"></li>
<li>
<a href="https://www.gitbook.com" target="blank" class="gitbook-link">
本书使用 GitBook 发布
</a>
</li>
</ul>
</nav>
</div>
<div class="book-body">
<div class="body-inner">
<div class="book-header" role="navigation">
<!-- Title -->
<h1>
<i class="fa fa-circle-o-notch fa-spin"></i>
<a href=".." >Python 对象初探</a>
</h1>
</div>
<div class="page-wrapper" tabindex="-1" role="main">
<div class="page-inner">
<div id="book-search-results">
<div class="search-noresults">
<section class="normal markdown-section">
<h1 id="python-对象初探">Python 对象初探</h1>
<p>在Python的世界中，一切皆对象。不论是是整数，还是字符串，甚至连类型、函数等都是一种对象。</p>
<p>Python中有一些已经定义好的类型对象，如<code>int</code>类型、<code>str</code>类型和<code>dict</code>类型等，
我们把这些对象称为内建类型对象，这些类型对象通过“实例化”可以创建内建类型对象的实例对象，
如<code>int</code>对象、<code>str</code>对象和<code>dict</code>对象等。</p>
<p>Python还允许我们通过<code>class A(object)</code>这样的形式自定义类型对象，我们同样可以对自定义的
类型对象“实例化”从而生成“实例对象”。这些对象之间存在的各种复杂关系构成了“类型系统”和“对象系统”。</p>
<p>Python的类型系统和对象系统非常庞大复杂，我们将分成多篇来解读。本章我们着重于了解对象在Python内部是如何表示的。</p>
<h2 id="python-内的对象">Python 内的对象</h2>
<p>对于计算机而言，对象是一个抽象的概念，计算机不懂对象是什么也不能理解字符串、整数是什么，
它所知道的一切都是字节。在计算机中，一个对象的实际的表示就是一片被分配的内存空间，这些内存可能是连续的也可能是离散的，而这片内存所代表的一个整体就是一个对象，它可以存储一系列数据，我们可以对这些数据进行读写等一系列行为。</p>
<p>在Python中，对象就是以C语言中的结构体在堆上申请一块内存为代表的。一个对象一旦创建它的内存大小就不再改变，那些需要容纳变长数据的对象需要在内维护一个指向一块可变大小的内存区域的指针，这样做的好处是可以通过指针非常方便的管理维护对象。</p>
<h3 id="对象机制的基石-pyobject">对象机制的基石 PyObject</h3>
<p>在Python中，所有对象都拥有一些相同的内容，这些内容定义在<code>PyObject</code>中，
可以说 <code>PyObject</code> 是一切对象的基础，</p>
<p>它包含 一个用于垃圾回收的双向链表，一个引用计数变量 <code>ob_refcnt</code> 和 一个类型对象指针<code>ob_type</code></p>
<p><code>源文件：</code><a href="https://github.com/python/cpython/blob/v3.7.0/Include/object.h#L106" target="_blank">Include/object.h</a></p>
<pre class="language-"><code class="lang-c"><span class="token comment">// Include/object.h</span>
<span class="token macro property">#<span class="token directive keyword">define</span> _PyObject_HEAD_EXTRA \
struct _object *_ob_next; \
struct _object *_ob_prev;</span>
<span class="token keyword">typedef</span> <span class="token keyword">struct</span> _object <span class="token punctuation">{</span>
_PyObject_HEAD_EXTRA <span class="token comment">// 双向链表 垃圾回收 需要用到</span>
Py_ssize_t ob_refcnt<span class="token punctuation">;</span> <span class="token comment">// 引用计数</span>
<span class="token keyword">struct</span> _typeobject <span class="token operator">*</span>ob_type<span class="token punctuation">;</span> <span class="token comment">// 指向类型对象的指针，决定了对象的类型</span>
<span class="token punctuation">}</span> PyObject<span class="token punctuation">;</span>
</code></pre>
<p><img src="PyObject.jpg" alt="PyObject"></p>
<p>在Python的定义中，对于一个整型对象<code>longobj</code>，它的变量<code>ob_refcnt</code>与内存管理机制有关，它实现了基于
引用计数的垃圾回收机制。当一个新的对象引用该对象<code>longobj</code>时，<code>longobj</code>的引用计数增加，而当该引用被删除时候，<code>longobj</code>的引用计数减少。
当<code>longobj</code>的变量<code>ob_refcnt</code>减少为0时，该对象<code>longobj</code>从堆上被删除，释放出内存供其他对象使用。</p>
<p><code>ob_type</code>是一个指向<code>_typeobject</code>的结构体指针，它对应着Python内部的一类
特殊对象(类型对象)，类型对象用于表示一个对象的类型。</p>
<p>在<code>PyObject</code>中定义的内容每一个Python对象都必须有，这些内容在每一个Python
对象占用的内存空间的最开始的字节中。</p>
<h3 id="定长对象和变长对象">定长对象和变长对象</h3>
<p><code>源文件：</code><a href="https://github.com/python/cpython/blob/v3.7.0/Include/object.h#L106" target="_blank">Include/object.h</a></p>
<pre class="language-"><code class="lang-c"><span class="token comment">// Include/object.h</span>
<span class="token keyword">typedef</span> <span class="token keyword">struct</span> _object <span class="token punctuation">{</span>
_PyObject_HEAD_EXTRA
Py_ssize_t ob_refcnt<span class="token punctuation">;</span>
<span class="token keyword">struct</span> _typeobject <span class="token operator">*</span>ob_type<span class="token punctuation">;</span>
<span class="token punctuation">}</span> PyObject<span class="token punctuation">;</span>
<span class="token keyword">typedef</span> <span class="token keyword">struct</span> <span class="token punctuation">{</span>
PyObject ob_base<span class="token punctuation">;</span>
Py_ssize_t ob_size<span class="token punctuation">;</span> <span class="token comment">/* Number of items in variable part */</span>
<span class="token punctuation">}</span> PyVarObject<span class="token punctuation">;</span>
</code></pre>
<p>变长对象<code>PyVarObject</code>与<code>PyObject</code>的不同在于多了一个<code>ob_size</code>变量，<code>ob_size</code>用于保存元素的个数</p>
<p><img src="PyVarObject.jpg" alt="PyVarObject"></p>
<p>Python中的整数对象，除了<code>PyObject_VAR_HEAD</code>外还有一个额外的东西<code>ob_digit</code>,
这个整数对象的值就保存在<code>ob_digit</code>中。Python中的字符串对象、list对象、
dict对象等除了<code>PyObject</code>以外都有一些属于自己的特殊信息。</p>
<p><code>源文件：</code><a href="https://github.com/python/cpython/blob/v3.7.0/Include/object.h#L98" target="_blank">Include/object.h</a></p>
<pre class="language-"><code class="lang-c"><span class="token comment">// Include/object.h</span>
<span class="token macro property">#<span class="token directive keyword">define</span> PyObject_VAR_HEAD PyVarObject ob_base;</span>
</code></pre>
<p><code>源文件：</code><a href="https://github.com/python/cpython/blob/v3.7.0/Include/longobject.h#L10" target="_blank">Objects/longobject.h</a></p>
<pre class="language-"><code class="lang-c"><span class="token comment">// Include/longobject.h</span>
<span class="token keyword">typedef</span> <span class="token keyword">struct</span> _longobject PyLongObject<span class="token punctuation">;</span> <span class="token comment">/* Revealed in longintrepr.h */</span>
</code></pre>
<p><code>源文件：</code><a href="https://github.com/python/cpython/blob/v3.7.0/Include/longintrepr.h#L85" target="_blank">Include/longintrepr.h</a></p>
<pre class="language-"><code class="lang-c"><span class="token comment">// Include/longintrepr.h</span>
<span class="token keyword">struct</span> _longobject <span class="token punctuation">{</span>
PyObject_VAR_HEAD
digit ob_digit<span class="token punctuation">[</span><span class="token number">1</span><span class="token punctuation">]</span><span class="token punctuation">;</span>
<span class="token punctuation">}</span><span class="token punctuation">;</span>
</code></pre>
<h2 id="类型对象">类型对象</h2>
<p>Python中的对象开始的一些字节中的内容我们是可以明确知道的，双向链表 <code>ob_refcnt</code>和<code>ob_type</code>; 在创建对象需要知道分配多少内存空间，这个时候
我们需要从对象的<code>元信息</code>去找该对象到底该分配多少内存，而这个<code>元信息</code>就在
该对象的类型对象中，也就是<code>ob_type</code>所指向的对象中。</p>
<p>接下来我们就看看这个类型对象到底长什么样</p>
<p>在 <code>_typeobject</code> 的定义中包含许多信息，主要分类以下几类:</p>
<ul>
<li>类型名, tp_name, 主要用于 Python 内部调试用</li>
<li>创建该类型对象时分配的空间大小信息，即 <code>tp_basicsize</code> 和 <code>tp_itemsize</code></li>
<li>与该类型对象相关的操作信息(如 <code>tp_print</code> 这样的函数指针)</li>
<li>一些对象属性</li>
</ul>
<p><code>源文件：</code><a href="https://github.com/python/cpython/blob/v3.7.0/Include/object.h#L346" target="_blank">Include/object.h</a></p>
<pre class="language-"><code class="lang-c"><span class="token comment">// Include/object.h</span>
<span class="token keyword">typedef</span> <span class="token keyword">struct</span> _typeobject <span class="token punctuation">{</span>
PyObject_VAR_HEAD
<span class="token keyword">const</span> <span class="token keyword">char</span> <span class="token operator">*</span>tp_name<span class="token punctuation">;</span> <span class="token comment">/* For printing, in format "<module>.<name>" */</span> <span class="token comment">// 类型名</span>
Py_ssize_t tp_basicsize<span class="token punctuation">,</span> tp_itemsize<span class="token punctuation">;</span> <span class="token comment">/* For allocation */</span>
<span class="token comment">// 创建该类型对象分配的内存空间大小</span>
<span class="token comment">// 一堆方法定义，函数和指针</span>
<span class="token comment">/* Methods to implement standard operations */</span>
destructor tp_dealloc<span class="token punctuation">;</span>
printfunc tp_print<span class="token punctuation">;</span>
getattrfunc tp_getattr<span class="token punctuation">;</span>
setattrfunc tp_setattr<span class="token punctuation">;</span>
PyAsyncMethods <span class="token operator">*</span>tp_as_async<span class="token punctuation">;</span> <span class="token comment">/* formerly known as tp_compare (Python 2)
or tp_reserved (Python 3) */</span>
reprfunc tp_repr<span class="token punctuation">;</span>
<span class="token comment">/* Method suites for standard classes */</span>
<span class="token comment">// 标准类方法集</span>
PyNumberMethods <span class="token operator">*</span>tp_as_number<span class="token punctuation">;</span> <span class="token comment">// 数值对象操作</span>
PySequenceMethods <span class="token operator">*</span>tp_as_sequence<span class="token punctuation">;</span> <span class="token comment">// 序列对象操作</span>
PyMappingMethods <span class="token operator">*</span>tp_as_mapping<span class="token punctuation">;</span> <span class="token comment">// 字典对象操作</span>
<span class="token comment">// 更多标准操作</span>
<span class="token comment">/* More standard operations (here for binary compatibility) */</span>
hashfunc tp_hash<span class="token punctuation">;</span>
ternaryfunc tp_call<span class="token punctuation">;</span>
reprfunc tp_str<span class="token punctuation">;</span>
getattrofunc tp_getattro<span class="token punctuation">;</span>
setattrofunc tp_setattro<span class="token punctuation">;</span>
<span class="token punctuation">.</span><span class="token punctuation">.</span><span class="token punctuation">.</span><span class="token punctuation">.</span><span class="token punctuation">.</span><span class="token punctuation">.</span>
<span class="token punctuation">}</span> PyTypeObject<span class="token punctuation">;</span>
</code></pre>
<h3 id="对象的创建">对象的创建</h3>
<p>一般来说，Python会有两种方式创建对象</p>
<ul>
<li>通过 Python C API 来创建</li>
<li>通过类型对象创建</li>
</ul>
<h4 id="范型api-或称为-aol-abstract-object-layer">范型API 或称为 AOL (Abstract Object Layer)</h4>
<p>这类API通常形如<code>PyObject_XXX</code>这样的形式。可以应用在任何Python对象上，如<code>PyObject_Print</code></p>
<h4 id="与类型相关的api-或称为-col-concrete-object-layer">与类型相关的API 或称为 COL (Concrete Object Layer)</h4>
<p>这类API 通常只能作用于某一种类型的对象上，对于每一种内建对象
Python都提供了这样一组API。例如整数对象，我们可以利用如下的API创建</p>
<pre class="language-"><code class="lang-c">PyObject <span class="token operator">*</span>longObj <span class="token operator">=</span> <span class="token function">PyLong_FromLong</span><span class="token punctuation">(</span><span class="token number">10</span><span class="token punctuation">)</span><span class="token punctuation">;</span>
</code></pre>
<h3 id="对象的行为">对象的行为</h3>
<p>在 <code>PyTypeObject</code> 中定义了大量的函数指针。这些函数指针可以视为类型对象中
所定义的操作，这些操作直接决定着一个对象在运行时所表现出的行为。</p>
<p>比如 <code>PyTypeObject</code> 中的 <code>tp_hash</code> 指明了该类型对象如何生成其<code>hash</code>值。
在之前的<code>PyTypeObject</code>代码中，我们可以看到非常重要的三组操作族
<code>PyNumberMethods *tp_as_number</code>、<code>PySequenceMethods *tp_as_sequence</code>、<code>PyMappingMethods *tp_as_mapping</code>，</p>
<p><code>源文件：</code><a href="https://github.com/python/cpython/blob/v3.7.0/Include/object.h#L240" target="_blank">Include/object.h</a></p>
<pre class="language-"><code class="lang-c"><span class="token comment">// Include/object.h</span>
<span class="token keyword">typedef</span> PyObject <span class="token operator">*</span> <span class="token punctuation">(</span><span class="token operator">*</span>binaryfunc<span class="token punctuation">)</span><span class="token punctuation">(</span>PyObject <span class="token operator">*</span><span class="token punctuation">,</span> PyObject <span class="token operator">*</span><span class="token punctuation">)</span><span class="token punctuation">;</span>
<span class="token keyword">typedef</span> <span class="token keyword">struct</span> <span class="token punctuation">{</span>
binaryfunc nb_matrix_multiply<span class="token punctuation">;</span>
binaryfunc nb_inplace_matrix_multiply<span class="token punctuation">;</span>
<span class="token punctuation">.</span><span class="token punctuation">.</span><span class="token punctuation">.</span><span class="token punctuation">.</span><span class="token punctuation">.</span><span class="token punctuation">.</span>
<span class="token punctuation">}</span> PyNumberMethods<span class="token punctuation">;</span>
</code></pre>
<p>在<code>PyNumberMethods</code>定义了一个数值对象该支持的操作。一个数值对象如 整数对象，那么它的类型对象 <code>PyLong_Type</code>中,<code>tp_as_number.nb_add</code>
就指定了它进行加法操作时的具体行为。<code>PyMappingMethods</code>与<code>PySequenceMethods</code>与此相同。</p>
<p>在以下代码中可以看出<code>PyLong_Type</code>中的<code>tp_as_number</code>项指向的是<code>long_as_number</code></p>
<p><code>源文件：</code><a href="https://github.com/python/cpython/blob/v3.7.0/Objects/longobject.c#L5342" target="_blank">Objects/longobject.h</a></p>
<pre class="language-"><code class="lang-c"><span class="token comment">// Objects/longobject.c</span>
<span class="token keyword">static</span> PyNumberMethods long_as_number <span class="token operator">=</span> <span class="token punctuation">{</span>
<span class="token punctuation">(</span>binaryfunc<span class="token punctuation">)</span>long_add<span class="token punctuation">,</span> <span class="token comment">/*nb_add*/</span>
<span class="token punctuation">(</span>binaryfunc<span class="token punctuation">)</span>long_sub<span class="token punctuation">,</span> <span class="token comment">/*nb_subtract*/</span>
<span class="token punctuation">(</span>binaryfunc<span class="token punctuation">)</span>long_mul<span class="token punctuation">,</span> <span class="token comment">/*nb_multiply*/</span>
<span class="token punctuation">.</span><span class="token punctuation">.</span><span class="token punctuation">.</span><span class="token punctuation">.</span><span class="token punctuation">.</span><span class="token punctuation">.</span>
<span class="token punctuation">}</span><span class="token punctuation">;</span>
PyTypeObject PyLong_Type <span class="token operator">=</span> <span class="token punctuation">{</span>
<span class="token function">PyVarObject_HEAD_INIT</span><span class="token punctuation">(</span><span class="token operator">&</span>PyType_Type<span class="token punctuation">,</span> <span class="token number">0</span><span class="token punctuation">)</span>
<span class="token string">"int"</span><span class="token punctuation">,</span> <span class="token comment">/* tp_name */</span>
<span class="token function">offsetof</span><span class="token punctuation">(</span>PyLongObject<span class="token punctuation">,</span> ob_digit<span class="token punctuation">)</span><span class="token punctuation">,</span> <span class="token comment">/* tp_basicsize */</span>
<span class="token keyword">sizeof</span><span class="token punctuation">(</span>digit<span class="token punctuation">)</span><span class="token punctuation">,</span> <span class="token comment">/* tp_itemsize */</span>
long_dealloc<span class="token punctuation">,</span> <span class="token comment">/* tp_dealloc */</span>
<span class="token number">0</span><span class="token punctuation">,</span> <span class="token comment">/* tp_print */</span>
<span class="token number">0</span><span class="token punctuation">,</span> <span class="token comment">/* tp_getattr */</span>
<span class="token number">0</span><span class="token punctuation">,</span> <span class="token comment">/* tp_setattr */</span>
<span class="token number">0</span><span class="token punctuation">,</span> <span class="token comment">/* tp_reserved */</span>
long_to_decimal_string<span class="token punctuation">,</span> <span class="token comment">/* tp_repr */</span>
<span class="token operator">&</span>long_as_number<span class="token punctuation">,</span> <span class="token comment">/* tp_as_number */</span>
<span class="token number">0</span><span class="token punctuation">,</span> <span class="token comment">/* tp_as_sequence */</span>
<span class="token number">0</span><span class="token punctuation">,</span> <span class="token comment">/* tp_as_mapping */</span>
<span class="token punctuation">.</span><span class="token punctuation">.</span><span class="token punctuation">.</span><span class="token punctuation">.</span><span class="token punctuation">.</span><span class="token punctuation">.</span>
<span class="token punctuation">}</span><span class="token punctuation">;</span>
</code></pre>
<h3 id="类型的类型">类型的类型</h3>
<p>在 <code>PyTypeObjet</code> 定义开始有 <code>PyOject_VAR_HEAD</code>，这意味着 Python 中的类型实际上也是一个对象。在Python中一切东西都是对象，那么类型对象的类型是什么呢？
对于其他对象，可以通过与其关联的对象确定其类型，那么通过什么来确定一个对象是类型对象呢？
答案就是 <code>PyType_Type</code></p>
<p><code>源文件：</code><a href="https://github.com/python/cpython/blob/v3.7.0/Objects/typeobject.c#L3540" target="_blank">Objects/typeobject.c</a></p>
<pre class="language-"><code class="lang-c"><span class="token comment">// Objects/typeobject.c</span>
PyTypeObject PyType_Type <span class="token operator">=</span> <span class="token punctuation">{</span>
<span class="token function">PyVarObject_HEAD_INIT</span><span class="token punctuation">(</span><span class="token operator">&</span>PyType_Type<span class="token punctuation">,</span> <span class="token number">0</span><span class="token punctuation">)</span>
<span class="token string">"type"</span><span class="token punctuation">,</span> <span class="token comment">/* tp_name */</span>
<span class="token keyword">sizeof</span><span class="token punctuation">(</span>PyHeapTypeObject<span class="token punctuation">)</span><span class="token punctuation">,</span> <span class="token comment">/* tp_basicsize */</span>
<span class="token keyword">sizeof</span><span class="token punctuation">(</span>PyMemberDef<span class="token punctuation">)</span><span class="token punctuation">,</span> <span class="token comment">/* tp_itemsize */</span>
<span class="token punctuation">.</span><span class="token punctuation">.</span><span class="token punctuation">.</span><span class="token punctuation">.</span><span class="token punctuation">.</span><span class="token punctuation">.</span>
<span class="token punctuation">}</span><span class="token punctuation">;</span>
</code></pre>
<p><code>PyType_Type</code> 在Python的类型机制中至关重要，所有用户自定义 <code>class</code> 所
对应的 <code>PyTypeObject</code> 对象都是通过 <code>PyType_Type</code>创建的</p>
<p>接下来我们看 <code>PyLong_Type</code> 是怎么与 <code>PyType_Type</code> 建立联系的。
前面提到，在Python中，每一个对象都将自己的引用计数、类型信息保存在开始的部分中。
为了方便对这部分内存初始化，Python中提供了几个有用的宏:</p>
<p><code>源文件：</code><a href="https://github.com/python/cpython/blob/v3.7.0/Include/object.h#L69" target="_blank">Include/object.h</a></p>
<pre class="language-"><code class="lang-c"><span class="token comment">// Include/object.h</span>
<span class="token macro property">#<span class="token directive keyword">ifdef</span> Py_TRACE_REFS</span>
<span class="token macro property">#<span class="token directive keyword">define</span> _PyObject_EXTRA_INIT 0, 0,</span>
<span class="token macro property">#<span class="token directive keyword">else</span></span>
<span class="token macro property">#<span class="token directive keyword">define</span> _PyObject_EXTRA_INIT</span>
<span class="token macro property">#<span class="token directive keyword">endif</span></span>
<span class="token macro property">#<span class="token directive keyword">define</span> PyObject_HEAD_INIT(type) \
{ _PyObject_EXTRA_INIT \
1, type },</span>
</code></pre>
<p>再回顾下 <code>PyObject</code>和<code>PyVarObject</code>的定义，初始化动作就一目了然了。
实际上，这些宏在各种内建类型对象的初始化中被大量使用。
以<code>PyLong_Type</code>为例，可以更清晰的看到一般的类型对象和<code>PyType_type</code>之间的关系</p>
<p><code>源文件：</code><a href="https://github.com/python/cpython/blob/v3.7.0/Objects/longobject.c#L5379" target="_blank">Objects/longobject.c</a></p>
<pre class="language-"><code class="lang-c"><span class="token comment">// Objects/longobject.c</span>
PyTypeObject PyLong_Type <span class="token operator">=</span> <span class="token punctuation">{</span>
<span class="token function">PyVarObject_HEAD_INIT</span><span class="token punctuation">(</span><span class="token operator">&</span>PyType_Type<span class="token punctuation">,</span> <span class="token number">0</span><span class="token punctuation">)</span>
<span class="token string">"int"</span><span class="token punctuation">,</span> <span class="token comment">/* tp_name */</span>
<span class="token function">offsetof</span><span class="token punctuation">(</span>PyLongObject<span class="token punctuation">,</span> ob_digit<span class="token punctuation">)</span><span class="token punctuation">,</span> <span class="token comment">/* tp_basicsize */</span>
<span class="token keyword">sizeof</span><span class="token punctuation">(</span>digit<span class="token punctuation">)</span><span class="token punctuation">,</span> <span class="token comment">/* tp_itemsize */</span>
<span class="token punctuation">.</span><span class="token punctuation">.</span><span class="token punctuation">.</span><span class="token punctuation">.</span><span class="token punctuation">.</span><span class="token punctuation">.</span>
<span class="token punctuation">}</span><span class="token punctuation">;</span>
</code></pre>
<p>下图是对象运行时的图像表现</p>
<p><img src="object_runtime_relation.jpg" alt=""></p>
<h2 id="python-对象的多态性">Python 对象的多态性</h2>
<p>Python创建一个对象比如 PyLongObject 时，会分配内存进行初始化，然后
Python内部会用 <code>PyObject*</code> 变量来维护这个对象，其他对象也与此类似，
所以在 Python 内部各个函数之间传递的都是一种范型指针 <code>PyObject*</code>
我们不知道这个指针所指的对象是什么类型，只能通过所指对象的 ob_type 域
动态进行判断，而Python正是通过 <code>ob_type</code> 实现了多态机制</p>
<p>考虑以下的 Print 函数</p>
<pre class="language-"><code class="lang-c"><span class="token keyword">void</span> <span class="token function">Print</span><span class="token punctuation">(</span>PyObject<span class="token operator">*</span> object<span class="token punctuation">)</span>
<span class="token punctuation">{</span>
object<span class="token operator">-></span>ob_type<span class="token operator">-></span><span class="token function">tp_print</span><span class="token punctuation">(</span>object<span class="token punctuation">)</span><span class="token punctuation">;</span>
<span class="token punctuation">}</span>
</code></pre>
<p>如果传递给 Print 函数的指针是一个 <code>PyLongObject*</code>，那么它会调用 PyLongOjbect 对象对应的类型对象中定义的 输出操作</p>
<p>如果指针是一个 <code>PyUnicodeObject*</code>，那么就会调用 PyUnicodeObject 对象对应的类型对象中定义的输出操作</p>
<h2 id="引用计数">引用计数</h2>
<blockquote>
<p>Python 通过引用计数来管理维护对象在内存中的存在与否
Python 中的每个东西都有一个对象， 都有<code>ob_refcnt</code> 变量
这个变量维护对象的引用计数，从而最终决定该对象的创建与销毁</p>
</blockquote>
<p>在Python中，主要通过 <code>Py_INCREF(op)</code>与<code>Py_DECREF(op)</code> 这两个宏
来增加和减少对一个对象的引用计数。当一个对象的引用计数减少到0之后，
<code>Py_DECREF</code>将调用该对象的<code>tp_dealloc</code>来释放对象所占用的内存和系统资源</p>
<p><code>源文件：</code><a href="https://github.com/python/cpython/blob/v3.7.0/Include/object.h#L777" target="_blank">Include/object.h</a></p>
<pre class="language-"><code class="lang-c"><span class="token comment">// Include/object.h</span>
<span class="token macro property">#<span class="token directive keyword">define</span> _Py_NewReference(op) ( \
_Py_INC_TPALLOCS(op) _Py_COUNT_ALLOCS_COMMA \
_Py_INC_REFTOTAL _Py_REF_DEBUG_COMMA \
Py_REFCNT(op) = 1)</span>
<span class="token macro property">#<span class="token directive keyword">define</span> Py_INCREF(op) ( \
_Py_INC_REFTOTAL _Py_REF_DEBUG_COMMA \
((PyObject *)(op))->ob_refcnt++)</span>
<span class="token macro property">#<span class="token directive keyword">define</span> Py_DECREF(op) \
do { \
PyObject *_py_decref_tmp = (PyObject *)(op); \
if (_Py_DEC_REFTOTAL _Py_REF_DEBUG_COMMA \
--(_py_decref_tmp)->ob_refcnt != 0) \
_Py_CHECK_REFCNT(_py_decref_tmp) \
else \
_Py_Dealloc(_py_decref_tmp); \
} while (0)</span>
</code></pre>
<p>一个对象引用计数减为0时，该对象的<code>析构函数</code>被调用，但调用<code>析构函数</code>并不意味着最终一定会调用 <code>free</code> 释放内存空间。
因为频繁的申请、释放内存会大大降低Python的执行效率。
因此Python中大量采用了内存对象池的技术，使得对象释放的空间归还给内存池而不是直接<code>free</code>，后续使用可先从对象池中获取</p>
<h2 id="python-对象的分类">Python 对象的分类</h2>
<ul>
<li>Fundamental 对象: 类型对象</li>
<li>Numeric 对象: 数值对象</li>
<li>Sequence 对象: 容纳其他对象的序列集合对象</li>
<li>Mapping 对象: 类似 C++中的 map 的关联对象</li>
<li>Internal 对象: Python 虚拟机在运行时内部使用的对象</li>
</ul>
<p><img src="object_category.jpg" alt=""></p>
<footer class="page-footer"><span class="copyright">Copyright © Prodesire 2018 all right reserved，powered by Gitbook</span><span class="footer-modification">该文件修订时间：
2018-12-21 14:07:47
</span></footer>
</section>
</div>
<div class="search-results">
<div class="has-results">
<h1 class="search-results-title"><span class='search-results-count'></span> results matching "<span class='search-query'></span>"</h1>
<ul class="search-results-list"></ul>
</div>
<div class="no-results">
<h1 class="search-results-title">No results matching "<span class='search-query'></span>"</h1>
</div>
</div>
</div>
</div>
</div>
</div>
<a href="../preface/modify-code.html" class="navigation navigation-prev " aria-label="Previous page: 修改 Python 源码">
<i class="fa fa-angle-left"></i>
</a>
<a href="long-object.html" class="navigation navigation-next " aria-label="Next page: Python 整数对象">
<i class="fa fa-angle-right"></i>
</a>
</div>
<script>
var gitbook = gitbook || [];
gitbook.push(function() {
gitbook.page.hasChanged({"page":{"title":"Python 对象初探","level":"2.1","depth":1,"next":{"title":"Python 整数对象","level":"2.2","depth":1,"path":"objects/long-object.md","ref":"objects/long-object.md","articles":[]},"previous":{"title":"修改 Python 源码","level":"1.5","depth":1,"path":"preface/modify-code.md","ref":"preface/modify-code.md","articles":[]},"dir":"ltr"},"config":{"plugins":["-search","search-plus@^0.0.11","-sharing","sharing-plus","github@^2.0.0","github-buttons@2.1.0","edit-link@^2.0.2","splitter","tbfed-pagefooter","prism","page-toc-button","back-to-top-button"],"root":".","styles":{"website":"styles/website.css","pdf":"styles/pdf.css","epub":"styles/epub.css","mobi":"styles/mobi.css","ebook":"styles/ebook.css","print":"styles/print.css"},"pluginsConfig":{"tbfed-pagefooter":{"copyright":"Copyright © Prodesire 2018","modify_label":"该文件修订时间：","modify_format":"YYYY-MM-DD HH:mm:ss"},"prism":{"lang":{"console":"bash","shell":"bash"},"css":["prismjs/themes/prism-okaidia.css"]},"github":{"url":"https://github.com/flaggo/python3-source-code-analysis"},"splitter":{},"sharing-plus":{"qq":false,"all":["facebook","google","twitter","instapaper","linkedin","pocket","stumbleupon"],"douban":false,"facebook":true,"weibo":false,"instapaper":false,"whatsapp":false,"hatenaBookmark":false,"twitter":true,"messenger":false,"line":false,"vk":false,"pocket":true,"google":false,"viber":false,"stumbleupon":false,"qzone":false,"linkedin":false},"lunr":{"maxIndexSize":1000000,"ignoreSpecialCharacters":false},"fontsettings":{"theme":"white","family":"sans","size":2},"highlight":{},"anchor-navigation-ex":{"isRewritePageTitle":false,"tocLevel1Icon":"fa fa-hand-o-right","tocLevel2Icon":"fa fa-hand-o-right","tocLevel3Icon":"fa fa-hand-o-right"},"page-toc-button":{},"back-to-top-button":{},"github-buttons":{"repo":"flaggo/python3-source-code-analysis","types":["star"],"size":"small"},"sharing":{"weibo":true,"douban":true,"linkedin":true,"facebook":true,"google":true,"twitter":true,"all":["weibo","douban","linkedin","facebook","google","twitter"]},"edit-link":{"label":"编辑此页面","base":"https://github.com/flaggo/python3-source-code-analysis/edit/master"},"theme-default":{"styles":{"website":"styles/website.css","pdf":"styles/pdf.css","epub":"styles/epub.css","mobi":"styles/mobi.css","ebook":"styles/ebook.css","print":"styles/print.css"},"showLevel":true},"search-plus":{}},"theme":"default","author":"Prodesire","pdf":{"pageNumbers":true,"fontSize":12,"fontFamily":"Arial","paperSize":"a4","chapterMark":"pagebreak","pageBreaksBefore":"/","margin":{"right":62,"left":62,"top":56,"bottom":56}},"structure":{"langs":"LANGS.md","readme":"README.md","glossary":"GLOSSARY.md","summary":"SUMMARY.md"},"variables":{},"title":"Python 3 源码分析","language":"zh-hans","output.name":"site","links":{"sidebar":{"主页":"https://github.com/flaggo/python3-source-code-analysis"}},"gitbook":"3.2.3","description":"致力于分析Python 3.7.0 的源码实现"},"file":{"path":"objects/object.md","mtime":"2018-12-21T06:07:47.792Z","type":"markdown"},"gitbook":{"version":"3.2.3","time":"2018-12-21T06:08:46.669Z"},"basePath":"..","book":{"language":""}});
});
</script>
</div>
<script src="../gitbook/gitbook.js"></script>
<script src="../gitbook/theme.js"></script>
<script src="../gitbook/gitbook-plugin-search-plus/jquery.mark.min.js"></script>
<script src="../gitbook/gitbook-plugin-search-plus/search.js"></script>
<script src="../gitbook/gitbook-plugin-sharing-plus/buttons.js"></script>
<script src="../gitbook/gitbook-plugin-github/plugin.js"></script>
<script src="../gitbook/gitbook-plugin-github-buttons/plugin.js"></script>
<script src="../gitbook/gitbook-plugin-edit-link/plugin.js"></script>
<script src="../gitbook/gitbook-plugin-splitter/splitter.js"></script>
<script src="../gitbook/gitbook-plugin-page-toc-button/plugin.js"></script>
<script src="../gitbook/gitbook-plugin-back-to-top-button/plugin.js"></script>
<script src="../gitbook/gitbook-plugin-lunr/lunr.min.js"></script>
<script src="../gitbook/gitbook-plugin-lunr/search-lunr.js"></script>
<script src="../gitbook/gitbook-plugin-fontsettings/fontsettings.js"></script>
</body>
</html>