pandas で json の読み書き

json の作成

json_create.py

#! /usr/bin/python
# -*- coding: utf-8 -*-
#
#	json_create.py
#
#					Oct/01/2018
# ------------------------------------------------------------------
import sys
import pandas as pd
#
# ------------------------------------------------------------------
sys.stderr.write("*** 開始 ***\n")
#
df = pd.DataFrame({'t0921': ["宇都宮",41295,"2003-8-12"],
		't0922': ["小山",41298,"2003-8-12"],
		't0923': ["佐野",31925,"2003-5-20"],
		't0924': ["足利",57418,"2003-4-18"],
		't0925': ["日光",71696,"2003-7-9"]},
	index=['name', 'population', 'date_mod'])

print(df)
#
path="./cities.json"
#
df.to_json(path)
#
sys.stderr.write("*** 終了 ***\n")
# ------------------------------------------------------------------

作成される json

cities.json

{
  "t0921": {
    "name": "宇都宮",
    "population": 41295,
    "date_mod": "2003-8-12"
  },
  "t0922": {
    "name": "小山",
    "population": 41298,
    "date_mod": "2003-8-12"
  },
  "t0923": {
    "name": "佐野",
    "population": 31925,
    "date_mod": "2003-5-20"
  },
  "t0924": {
    "name": "足利",
    "population": 57418,
    "date_mod": "2003-4-18"
  },
  "t0925": {
    "name": "日光",
    "population": 71696,
    "date_mod": "2003-7-9"
  }
}

json の読み込み

json_read.py

#! /usr/bin/python
# -*- coding: utf-8 -*-
#
#	json_read.py
#
#					Oct/01/2018
# ------------------------------------------------------------------
import sys
import pandas as pd
#
# ------------------------------------------------------------------
sys.stderr.write("*** 開始 ***\n")
#
file_json = "./cities.json"
#
df=pd.read_json(file_json)
print(df)
#
df.to_json("tmp001.json")
#
sys.stderr.write("*** 終了 ***\n")
# ------------------------------------------------------------------

実行結果

$ ./json_read.py
*** 開始 ***
                t0921      t0922      t0923      t0924     t0925
date_mod    2003-8-12  2003-8-12  2003-5-20  2003-4-18  2003-7-9
name              宇都宮         小山         佐野         足利        日光
population      41295      41298      31925      57418     71696
*** 終了 ***

確認したバージョン

$ python
Python 3.10.9 (main, Dec 19 2022, 17:35:49) [GCC 12.2.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import pandas
>>> pandas.__version__
'1.5.3'