如何从具有多个 arrays 的 json 文件创建 pandas dataframe

Question

I'm trying to create a pandas dataframe from a json file that I have of my Apple Health Data.

My json file looks like this:

{
  "data": {
    "workouts": [],
    "metrics": [
      {
        "name": "active_energy",
        "units": "kcal",
        "data": [
          {
            "qty": 213.881,
            "date": "2022-04-12 00:00:00 -0600"
          }
        ]
      },
      {
        "name": "apple_exercise_time",
        "units": "min",
        "data": [
          {
            "date": "2022-04-12 00:00:00 -0600",
            "qty": 6
          }
        ]
      },
      {
        "name": "sleep_analysis",
        "units": "min",
        "data": []
      }
    ]
  }
}

In this data, there is an empty array called workouts and another called metrics . I want to take the metrics array from this file and turn it into a pandas dataframe like this:

date	name	qty	units
2022-04-12	active_energy	213.881	kcal
2022-04-12	apple_excersise_time	6	min

Answer 1

Here's one way using a DataFrame constructor, explode and join :

tmp = pd.DataFrame(my_data['data']['metrics']).explode('data')
s = tmp['data'].dropna()
out = tmp.drop(columns='data').join(pd.DataFrame(s.tolist(), index=s.index))

Output:

                  name units      qty                       date
0        active_energy  kcal  213.881  2022-04-12 00:00:00 -0600
1  apple_exercise_time   min    6.000  2022-04-12 00:00:00 -0600
2       sleep_analysis   min      NaN                        NaN

Answer 2

Someone on SO shared this with me around a month ago. Sharing with you now.

import pandas as pd
df = pd.read_json("https://www.chsli.org/sites/default/files/transparency/111888924_GoodSamaritanHospitalMedicalCenter_standardcharges.json", lines=True)
print(df.head())
df.to_csv(r'C:\\your_path_here\\chsli.csv')

Result: