python利用 pandas 库处理 json 文件数据

案例一

douban.json 文件内容截图如下：

python脚本内容：

import json
import pandas as pd 

df = pd.read_json("./douban.json", lines=True, encoding="utf-8")#lines是按行读取的意思
df.to_excel("./douban.xlsx")#输出表格douban.xlsx
print("over.....")

douban.xlsx 表格内容截图如下

案例二

ceshi.json 文件内容

[{"ttery":"[123]","issue":"20130801-3391"},{"ttery":"[123]","issue":"20130801-3390"},{"ttery":"[123]","issue":"20130801-3389"}]

python脚本内容：

# -*- coding: utf-8 -*-

import pandas as pd

file = open('ceshi.json', 'r', encoding='utf-8')

df = pd.read_json(file, orient='records')
df.to_excel('pandas处理ceshi-json.xlsx', index=False, columns=["ttery", "issue"])

输出的 “pandas处理ceshi-json.xlsx”文件内容：

案例三

Pandas pd.json_normalize() 读取半结构化 JSON

data = [
    {
        "state": "Florida",
        "shortname": "FL",
        "info": {"governor": "Rick Scott", "web": "gairuo.com"},
        "counties": [
            {"name": "Dade", "population": 12345},
            {"name": "Broward", "population": 40000},
            {"name": "Palm Beach", "population": 60000},
        ],
    },
    {
        "state": "Ohio",
        "shortname": "OH",
        "info": {"governor": "John Kasich", "web": "sin80.com"},
        "counties": [
            {"name": "Summit", "population": 1234},
            {"name": "Cuyahoga", "population": 1337},
        ],
    },
]

# 记录都在 counties，元数据取 state、shortname 和
# info 中的 web, info 中的 governor 被舍弃
result = pd.json_normalize(
    data,
    record_path="counties",
    meta=["state", "shortname", ["info", "web"]]
)
result
'''
         name  population    state shortname    info.web
0        Dade       12345  Florida        FL  gairuo.com
1     Broward       40000  Florida        FL  gairuo.com
2  Palm Beach       60000  Florida        FL  gairuo.com
3      Summit        1234     Ohio        OH   sin80.com
4    Cuyahoga        1337     Ohio        OH   sin80.com
'''