学习笔记 | Python中的序列化

Python语言有个专门搞序列化的模块，pickle。用pickle序列化的基本模板如下：

import pickle

file_name = 'Pickle_name.txt'
dic = { 0:1 , 2:4}


with open(file_name,'wb') as f:
	f.write(pickle.dumps(dic))

with open(file_name,'rb') as f:
	dic = pickle.loads(f.read())
	print(dic)


dic = { 2:9 , 4:1}

with open(file_name,'wb') as f:
	pickle.dump(dic,f)

with open(file_name,'rb') as f:
	dic = pickle.load(f)
	print(dic)

pickle模块中的dumps()表示序列化，loads()表示反序列化。这两个方法在遇到file-like object时，可以简化为dump()和load()，简化方法如上。pickle的这些方法实际上是将Python对象序列化为bytes，再反序列化回Python对象，所以打开文件的模式也要是“wb”和“rb”。

Python语言特定的序列化模块是pickle，但如果要把序列化搞得更通用、更符合Web标准，就可以使用json模块。

import json

file_name = 'Json.json'
dic = { 0:1 , 2:4}

with open(file_name,'w') as f:
	f.write(json.dumps(dic))

with open(file_name,'r') as f:
	dic = json.loads(f.read())
	print(dic)


dic = { 2:9 , 4:1}

with open(file_name,'w') as f:
	json.dump(dic,f)

with open(file_name,'r') as f:
	dic = json.load(f)
	print(dic)

大同小异。不同点在于打开文件的模式是“w”和“r”。