python 中 json 包用法简单总结

---

JSON包的引用

在文件头部引用json包

import json

---

python对象与JSON对象的互相转换

json对象的类型为’str’：

dic = {'b':'I', 'a':123, 'c':'100'}
j1 = json.dumps(dic)
print (j1)
# {"b": "I", "a": 123, "c": "100"}
print (type(j1))
# <class 'str'> 
print (j1[0],j1[1],j1[2])
# { " b

将sort_keys参数置为True时，生成的json对象是按键排序的，sort_keys默认为False：

dic = {'b':'I', 'a':123, 'c':'100'}
j2 = json.dumps(dic, sort_keys = True)
print (j2)
# {"a": 123, "b": "I", "c": "100"}

通过indent参数可以设置json对象的缩进格式：

dic = {'b':'I', 'a':123, 'c':'100'}
j3 = json.dumps(dic, sort_keys = True, indent = 4)
print (j3)
# {
#     "a": 123,
#     "b": "I",
#     "c": "100"
# }

通过separators参数可以设置json对象的分隔符：

dic = {'b':'I', 'a':123, 'c':'100'}
j4 = json.dumps(dic, sort_keys = True, separators = ('$','@'))
print (j4)
# {"a"@123$"b"@"I"$"c"@"100"}

列表也可以转化为json对象：

list1 = [1, 'big', [1, 'a', {'p':'+'}], (['t',2],{'1':'o'}), {'c':0,'d':1}]
j5 = json.dumps(list1)
print (j5)
# [1, "big", [1, "a", {"p": "+"}], [["t", 2], {"1": "o"}], {"c": 0, "d": 1}]

元组转化为json对象：

tuple1 = (1, 0)
j6 = json.dumps(tuple1)
print (j6)
# [1, 0]

将字典转换为json字符串时，key需为数字或字母，否则报错，可通过skipkeys参数跳过这些键：

dic1 = {1:'one', 2.3:'tPt', 'two':2, (3,4):'thr&four'}
j7 = json.dumps(dic1)
print (j7)
# TypeError: keys must be a string
j8 = json.dumps(dic1, skipkeys = True)
print (j8)
# {"1": "one", "2.3": "tPt", "two": 2}

dumps对中文使用ascii编码方式，通过将ensure_ascii参数设置为False可输出中文：

dic_info = {'name':'Elizabeth', 'husband':'达西', 'age':22}
j9 = json.dumps(dic_info)
print (j9)
# {"name": "Elizabeth", "husband": "\u8fbe\u897f", "age": 22}
j10 = json.dumps(dic_info,ensure_ascii=False)
print (j10)
# {"name": "Elizabeth", "husband": "达西", "age": 22}

将json对象解码为python对象：

dic = {'b':'I', 'a':123, 'c':'100'}
j1 = json.dumps(dic)
decode1 = json.loads(j1)
print (decode1)
# {'b': 'I', 'a': 123, 'c': '100'}

---

存取JSON文件

存json文件，将python对象存到json格式的文件中不再用dumps，而是要用dump：

dic_info = {'name':'Elizabeth', 'husband':'达西', 'age':22}
filew = open ('Elizabeth.json', 'w', encoding='utf-8')
json.dump(dic_info, filew)
filew.close()

json文件如下图所示：

读取json文件，从json文件中读取内容存入python对象，不再用loads而是要用load：

filer = open ('Elizabeth.json', 'r', encoding='utf-8')
Elizabeth = json.load(filer)
filer.close()
print (Elizabeth)
# {'name': 'Elizabeth', 'husband': '达西', 'age': 22}
print (type(Elizabeth))
# <class 'dict'>

---

python官方帮助文档

https://docs.python.org/3/library/json.html?highlight=json#module-json