I have a JSON list data type data. I am using json_normalize() function to read it as a CSV format. In the data, I have two types of indexing and I need both indexings in my CSV file. But for me, its showing only one level of indexing

Data file:

data = [{
  "line": [
    {
      "index_line": 0,
      "tokens": [
        {
          "index_tokens": 1,
          "person": "A",
          "age": "23",
          "sex": "M",
          "add": ""
        },
        {
          "index_tokens": 2,
          "person": "B",
          "age": "22",
          "sex": "F",
          "add": ""
        }
        ]
    },
    {
      "index_line": 2,
      "tokens": [
        {
          "index_tokens": 1,
          "person": "C",
          "age": "23",
          "sex": "F",
          "add": ""
        },
        {
          "index_tokens": 2,
          "person": "D",
          "age": "21",
          "sex": "F",
          "add": ""
        }
        ]
    }
  ]
}
]

This command is showing the "index_line"

pd_data1 = pd.json_normalize(data , "line")
pd_data1

This command is showing the "index_tokens" not the "index_line". But I need both the index.

pd_data1 = pd.json_normalize(data , ["line" , ["tokens"]])
pd_data1

CodePudding user response：

Try:

df = pd.DataFrame(data[0]["line"]).explode("tokens")
df = pd.concat([df, df.pop("tokens").apply(pd.Series)], axis=1)
print(df)

Prints: