How to build recursive JSON hierarchy tree?-CodePudding

I would like to do json tree that inside have multiple of parent, you will see my example dataframe in column child_id , rows to be parent must be equal to 0 and it will be Color and Mobile and each children will bind id to parent.

My data

import pandas as pd

data = {
    'id': ['2', '13', '14', '15', '16', '17', '18'],
    'name': ['color', 'red', 'blue', 'ruby red', 'mobile', 'iphone', 'sumsung'],
    'child_id': ['0', '2', '2', '13', '0', '16', '16']
}
  

df = pd.DataFrame(data)
df

Expected Output

[
  {
    'name': 'color',
    'id': 2,
    'child_id': 0,
    'children': [
      {
        'name': 'red',
        'id': 13,
        'child_id': 2,
        'children': [
          {
            'name': 'ruby red',
            'id': 15,
            'child_id': 13,
          },
        ],
      },
      {
        'name': 'blue',
        'id': 14,
        'child_id': 2,
      },
    ],
  },
  {
    'name': 'mobile',
    'id': 16,
    'child_id': 0,
    'children': [
      {
        'name': 'iphone',
        'id': 17,
        'child_id': 16,
      },
      {
        'name': 'samsung',
        'id': 18,
        'child_id': 16,
      },
    ],
  },
]

CodePudding user response：

You can do this without pandas. Just iterate the initial data to create a dictionary, keyed by id, and with corresponding objects (dictionaries) as values.

Then iterate the data again to link these objects to their parents.

Here is how it could be done:

def makehiearchy(data):
    result = []
    d = { "0": { "children": result } }
    for id, name, child_id in zip(data["id"], data["name"], data["child_id"]):
        d[id] = { "name": name, "id": id, "child_id": child_id }
    
    for id, child_id in zip(data["id"], data["child_id"]):
        parent = d[child_id]
        if "children" not in parent:
            parent["children"] = []
        parent["children"].append(d[id])

    return result

# Example run
data = {
    'id': ['2', '13', '14', '15', '16', '17', '18'],
    'name': ['color', 'red', 'blue', 'ruby red', 'mobile', 'iphone', 'sumsung'],
    'child_id': ['0', '2', '2', '13', '0', '16', '16']
}
hierarchy = makehiearchy(data)