Skip to content

Commit 2bae5e6

Browse files
hsz1273327hsz1273327
committed
update
1 parent c056555 commit 2bae5e6

File tree

4 files changed

+209
-0
lines changed

4 files changed

+209
-0
lines changed

‎src/.DS_Store‎

0 Bytes
Binary file not shown.

‎src/术语表/.DS_Store‎

6 KB
Binary file not shown.

‎src/虚拟机篇/python的虚拟机.ipynb‎

Lines changed: 57 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,57 @@
1+
{
2+
"cells": [
3+
{
4+
"cell_type": "markdown",
5+
"metadata": {},
6+
"source": [
7+
"# CPython的虚拟机\n",
8+
"\n",
9+
"CPython 使用一个基于栈的虚拟机.也就是说它完全面向栈数据结构的(你可以\"推入(push)\"一个东西到栈顶，或者从栈顶上\"弹出(pop)\"一个东西来).\n",
10+
"\n",
11+
"CPython 使用三种类型的栈:\n",
12+
"\n",
13+
"+ 调用栈(call stack)\n",
14+
"\n",
15+
" 这是运行 Python 程序的主要结构.它为每个当前活动的函数调用使用了一个东西--\"帧(frame)\".栈底是程序的入口点.每个函数调用推送一个新的帧到调用栈,每当函数调用返回后,这个帧被销毁.\n",
16+
" \n",
17+
"+ 计算栈(evaluation stack)\n",
18+
"\n",
19+
" 也称为数据栈(data stack).在每个帧中有一个计算栈.这个栈就是 Python 函数运行的地方.运行的 Python 代码大多数是由推入到这个栈中的东西组成的,这个栈负责操作它们.然后在返回后销毁它们。\n",
20+
"\n",
21+
"+ 块栈(block stack)\n",
22+
"\n",
23+
" 在每个帧中还有一个块栈.它被 Python 用于去跟踪某些类型的控制结构--循环、`try / except 块`、以及 `with 块`，全部推入到块栈中,当你退出这些控制结构时块栈被销毁.这将帮助 Python 了解任意给定时刻哪个块是活动的.比如，一个 continue 或者 break 语句可能影响正确的块.\n",
24+
" \n",
25+
" \n"
26+
]
27+
},
28+
{
29+
"cell_type": "code",
30+
"execution_count": null,
31+
"metadata": {},
32+
"outputs": [],
33+
"source": []
34+
}
35+
],
36+
"metadata": {
37+
"kernelspec": {
38+
"display_name": "Python 3",
39+
"language": "python",
40+
"name": "python3"
41+
},
42+
"language_info": {
43+
"codemirror_mode": {
44+
"name": "ipython",
45+
"version": 3
46+
},
47+
"file_extension": ".py",
48+
"mimetype": "text/x-python",
49+
"name": "python",
50+
"nbconvert_exporter": "python",
51+
"pygments_lexer": "ipython3",
52+
"version": "3.6.4"
53+
}
54+
},
55+
"nbformat": 4,
56+
"nbformat_minor": 2
57+
}

‎src/虚拟机篇/编译与反编译.ipynb‎

Lines changed: 152 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,152 @@
1+
{
2+
"cells": [
3+
{
4+
"cell_type": "markdown",
5+
"metadata": {},
6+
"source": [
7+
"# 编译与反编译\n",
8+
"\n",
9+
"首先说明下本文的python特指官方的Cpython实现.\n",
10+
"\n",
11+
"python外界的定义是解释型语言,其原理是将代码放入解释器,解释一句执行一句.然而实际上python的执行流程会比这个略微复杂一些.python源码到执行的中间会有一个步骤生成字节码,实际执行的其实是这个字节码.因此,如果想跳过第一步,我们可以对源码进行编译,编译为字节码后直接执行字节码."
12+
]
13+
},
14+
{
15+
"cell_type": "markdown",
16+
"metadata": {},
17+
"source": [
18+
"## python的字节码\n",
19+
"\n",
20+
"cpython实际上运行的是字节码,我们所写的python源文件之所以可以被执行,其原因是当python程序加载时,源文件会被先翻译为字节码.然后保存在同级文件的`__pycache__`文件夹下.下次执行是,如果`__pycache__`下的字节码文件和源码文件修改时间匹配一致就会直接执行`__pycache__`中缓存的字节码,不一致则会将对应的字节码替换.\n",
21+
"\n",
22+
"\n",
23+
"python的字节码在旧版本中有两种,一种是常规的`.pyc`文件,其加载速度相对于之前的.py文件有所提高,而且还可以实现源码隐藏,以及一定程度上的反编译;另一种是经过优化的`.pyo`文件(相比于`.pyc`文件更小),也可以提高加载速度.\n",
24+
"\n",
25+
"而现在则统一都是以`.pyc`作为后缀,优化的等级则显式的放在编译后文件的命名上."
26+
]
27+
},
28+
{
29+
"cell_type": "markdown",
30+
"metadata": {},
31+
"source": [
32+
"### 编译模块到`.pyc`\n",
33+
"\n",
34+
"编译到`.pyc`我们可以使用两个标准库提供的工具:\n",
35+
"\n",
36+
"+ `py_compile`用于编译单独的源码文件\n",
37+
"\n",
38+
"+ `compileall`用于将某个文件夹下的源码文件递归地进行编译\n",
39+
"\n",
40+
"#### **单文件编译**\n",
41+
"\n",
42+
"使用`py_compile`单独编译一个文件的话,可以使用命令行工具\n",
43+
"\n",
44+
"```shell\n",
45+
"python -m py_compile /path/to/需要生成.pyc的脚本.py\n",
46+
"```\n",
47+
"即可.\n",
48+
"\n",
49+
"也可以在python脚本下使用,调用`py_compile`:\n",
50+
"\n",
51+
"```python\n",
52+
"py_compile.compile(file[, cfile[, dfile[, doraise]]])\n",
53+
"```\n",
54+
"\n",
55+
"其中\n",
56+
"+ `file`,表示需要生成`.pyc`或`.pyo`文件的源码路径;\n",
57+
"+ `cfile`，表示需要生成`.pyc`或`.pyo`文件的目标文件路径.它默认是以`.pyc`为扩展名的形如`xxxx.cpython-36.pyc`这样形式的字符串.此外，当且仅当所使用的解释器允许编译成.pyo文件,才能以`.pyo`结尾,这个后文说.\n",
58+
"+ `dfile`,表示编译出错时,将报错信息中的名字`file`替换为`dfile`\n",
59+
"+ `doraise`，设置是否忽略异常.若为`True`,则抛出`PyCompileError`异常;否则直接将错误信息写入`sys.stderr`.\n",
60+
"\n",
61+
"\n",
62+
"类似的,编译整个文件夹可以使用命令行工具`compileall`\n",
63+
"\n",
64+
"```shell\n",
65+
"python -m compileall <dir>\n",
66+
"```\n",
67+
"\n",
68+
"常用的参数有`-b`表示写入它们源文件同级并且同名不同后缀\n",
69+
"\n",
70+
"也可以在python脚本下使用,调用`compileall`:\n",
71+
"\n",
72+
"```python\n",
73+
"import compileall\n",
74+
"\n",
75+
"compileall.compile_dir('Lib/', force=True)\n",
76+
"```\n",
77+
"compile_dir参数有:\n",
78+
"\n",
79+
"+ `dir`目标文件夹\n",
80+
"+ `maxlevels=10`文件夹最大递归深度\n",
81+
"+ `ddir=None`检查ddir中二进制文件的时间戳,如果与目标文件不一致才会编译\n",
82+
"+ `force=False`强制编译\n",
83+
"+ `rx=None`使用`re`查找目标文件夹下要编译的文件\n",
84+
"+ `quiet=0`是否输出编译过程\n",
85+
"+ `legacy=False`字节码文件是否被写入它们源文件同级并且同名(python2的默认编译行为),它们可能覆盖由另一版本的Python创建的字节码文件\n",
86+
"+ `optimize=-1`优化等级,\n",
87+
"+ `workers=1`多进程编译\n",
88+
"\n",
89+
"\n",
90+
"#### **编译优化**\n",
91+
"\n",
92+
"使用命令行工具编译是否优化的标志位在python一级,使用`-O`表示优化等级为1,`-OO`表示优化等级为2级(相比1级去掉了文档字符串),经过优化的字节码文件会被默认的命名为`xxxx.cpython-36.opt-2.pyc`这样的形式.\n",
93+
"\n",
94+
"单文件的1级优化编译\n",
95+
"\n",
96+
"```shell\n",
97+
"python -O -m py_compile /path/to/需要生成.pyo的脚本.py\n",
98+
"```\n",
99+
"\n",
100+
"文件夹下递归2级优化编译:\n",
101+
"\n",
102+
"```shell\n",
103+
"python -O -m compileall -b <path>\n",
104+
"```\n",
105+
"\n",
106+
"需要注意编译后的字节码文件并不会提高运行时的执行速度,只会提高模块的加载速度."
107+
]
108+
},
109+
{
110+
"cell_type": "markdown",
111+
"metadata": {},
112+
"source": [
113+
"## 反编译\n",
114+
"\n",
115+
"指望通过编译python源文件来反编译是不现实的,使用`uncompyle6`工具可以非常轻易的将`.pyc`反编译为源码`.py`文件.`uncompyle6`可以直接使用`pip`工具安装."
116+
]
117+
},
118+
{
119+
"cell_type": "markdown",
120+
"metadata": {},
121+
"source": [
122+
"## 结语\n",
123+
"\n",
124+
"python的编译功能看起来挺鸡肋的,但还是可以在一些特定环境下发挥用处的,比如:\n",
125+
"\n",
126+
"+ 硬盘空间资源紧张的硬件上,像树莓派,结合`zipapp`,可执行文件可以被控制在一个很小的程度.\n",
127+
"+ 吓唬外行,通常知道点python的也不会知道可以`编译`,编译后的字节码人类也是没法直接解读的,一定程度上可以防止源码泄漏\n"
128+
]
129+
}
130+
],
131+
"metadata": {
132+
"kernelspec": {
133+
"display_name": "Python 3",
134+
"language": "python",
135+
"name": "python3"
136+
},
137+
"language_info": {
138+
"codemirror_mode": {
139+
"name": "ipython",
140+
"version": 3
141+
},
142+
"file_extension": ".py",
143+
"mimetype": "text/x-python",
144+
"name": "python",
145+
"nbconvert_exporter": "python",
146+
"pygments_lexer": "ipython3",
147+
"version": "3.6.4"
148+
}
149+
},
150+
"nbformat": 4,
151+
"nbformat_minor": 2
152+
}

0 commit comments

Comments
 (0)