Python编译、运行、反编译pyc文件

Python 并非完全是解释性语言，它也存在编译。先将源码文件 *.py 编译为 *.pyc，然后由 Python 的虚拟机执行。相对于 *.py 文件来说，编译为 *.pyc本质上和 *.py 没有太大区别，只是提高了模块的加载速度，并没有提高代码的执行速度。

生成 *.pyc 文件

要编译为 *.pyc 文件，需要安装compile模块：

pip install compile

Collecting compileDownloading compile-1.0.3.tar.gz (6.0 kB)
Building wheels for collected packages: compileBuilding wheel for compile (setup.py) ... doneCreated wheel for compile: filename=compile-1.0.3-py3-none-any.whl size=6547 sha256=d4ee7be709d3280af20a669a934784c1abe2051797b99ca81eaaef22baf116e2Stored in directory: c:\users\anonym\appdata\local\pip\cache\wheels\ec\c9\48\b5caf5b67a5ae3bd2dc854aefe766aafeaad244b863792ef32
Successfully built compile
Installing collected packages: compile
Successfully installed compile-1.0.3

生成单个文件

python -m py_compile test.py

生成多个文件

python -m compile all E:\Python\Paddle\PaddleClas\ppcls\modeling\architectures

反编译 `pyc`

首先安装库 uncompyle:

pip install uncompyle

Collecting uncompyleDownloading uncompyle-2.0.0-py2.py3-none-any.whl (2.2 kB)
Collecting uncompyle6Downloading uncompyle6-3.7.4-py3-none-any.whl (316 kB)|████████████████████████████████| 316 kB 10 kB/s
Collecting spark-parser<1.9.0,>=1.8.9Downloading spark_parser-1.8.9-py38-none-any.whl (18 kB)
Collecting xdis<5.1.0,>=5.0.4Downloading xdis-5.0.9-py2.py3-none-any.whl (129 kB)|████████████████████████████████| 129 kB 1.4 kB/s
Requirement already satisfied: click in c:\program files\python\python38\lib\site-packages (from spark-parser<1.9.0,>=1.8.9->uncompyle6->uncompyle) (8.0.1)
Requirement already satisfied: six>=1.10.0 in c:\program files\python\python38\lib\site-packages (from xdis<5.1.0,>=5.0.4->uncompyle6->uncompyle) (1.16.0)
Requirement already satisfied: colorama in c:\program files\python\python38\lib\site-packages (from click->spark-parser<1.9.0,>=1.8.9->uncompyle6->uncompyle) (0.4.4)
Installing collected packages: xdis, spark-parser, uncompyle6, uncompyle
Successfully installed spark-parser-1.8.9 uncompyle-2.0.0 uncompyle6-3.7.4 xdis-5.0.9

命令行如下：

uncompyle6 .pyc > .py

关于效率

关于效率，Python 文档 “Compiled” Python files 中这样介绍：

A program doesn’t run any faster when it is read from a .pyc or .pyo file than when it is read from a .py file; the only thing that’s faster about .pyc or .pyo files is the speed with which they are loaded.

也就是说，".py"和"pyc"运行速度相差无几，加载速度 *.pyc 稍占优势。

当然，除此之外，还有一个很大的优点：隐藏源码！