在 Python 中选择性展平 JSON

Question

I'm extracting data from an API which returns non-standardised jsons, meaning that I have to write a custom function to parse each data stream as the JSON has a different structure of nested dictionaries and lists. The output I desire is a flattened list containing all of the relevant endpoint items. I've written functions for a few of these but am trying to create something more generalisable.

An example JSON would be of the form:

JSON = {
    'metadata' : {
        'meta_param' : 'meta_value',
    },
    'contents' : [
        { 
            'content' : [
                {
                    'attributes' : {
                        'series_name' : 'series_1a',
                    },
                    'data' : {
                        'index' : [0, 1, 2, 3, 4, 5],
                        'values' : [12, 84, 38, 3, 92, 67]
                    },
                },
                {
                    'attributes' : {
                        'series_name' : 'series_2a',
                    },
                    'data' : {
                        'index' : [0, 1, 2, 3, 4, 5],
                        'values' : [42, 73, 48, 20, 19, 61]
                    },
                },
            ],
        },
        {
            'content' : [
                {
                    'attributes' : {
                        'series_name' : 'series_1b',
                    },
                    'data' : {
                        'index' : [0, 1, 2, 3, 4, 5],
                        'values' : [13, 85, 39, 4, 93, 68]
                    },
                },
                {
                    'attributes' : {
                        'series_name' : 'series_2b',
                    },
                    'data' : {
                        'index' : [0, 1, 2, 3, 4, 5],
                        'values' : [43, 74, 49, 21, 20, 62]
                    },
                },
            ],
        },
    ]
}

And I'd write a custom function like this to convert it into a flat list

flatlist = []

for contents in JSON['contents']:
    for content in contents['content']:
        flatlist.append(content['data'])

flatlist -> 
[{'index': [0, 1, 2, 3, 4, 5], 'values': [12, 84, 38, 3, 92, 67]},
 {'index': [0, 1, 2, 3, 4, 5], 'values': [42, 73, 48, 20, 19, 61]},
 {'index': [0, 1, 2, 3, 4, 5], 'values': [13, 85, 39, 4, 93, 68]},
 {'index': [0, 1, 2, 3, 4, 5], 'values': [43, 74, 49, 21, 20, 62]}]

I'm trying to create a function which accepts a JSON of nested dictionaries and lists, as well as a 'routemap' of how to access specific items within it, and then returns a flat list of the specified items.

A routemap and its usage for the previous example would look like:

JSON_route = [
    'contents',
    'content',
    'data',
]

flatlist = json_to_flatlist(JSON, JSON_route)

flatlist -> 
[{'index': [0, 1, 2, 3, 4, 5], 'values': [12, 84, 38, 3, 92, 67]},
 {'index': [0, 1, 2, 3, 4, 5], 'values': [42, 73, 48, 20, 19, 61]},
 {'index': [0, 1, 2, 3, 4, 5], 'values': [13, 85, 39, 4, 93, 68]},
 {'index': [0, 1, 2, 3, 4, 5], 'values': [43, 74, 49, 21, 20, 62]}]

My key issue is that I'm not sure how to programmatically convert the JSON_route list into the nested for loops needed to extract the data.

Thanks for any help!

Answer 1

def json_to_flatlist(obj, route):
    if len(route) == 0:
        yield obj
    elif isinstance(obj, list):
        for item in obj:
            yield from json_to_flatlist(item, route)
    else:
        yield from json_to_flatlist(obj[route[0]], route[1:])

JSON_route = [
    'contents',
    'content',
    'data',
]

result = list(json_to_flatlist(JSON, JSON_route))
print(result)

Result (newlines added for clarity):

[
{'index': [0, 1, 2, 3, 4, 5], 'values': [12, 84, 38, 3, 92, 67]}, 
{'index': [0, 1, 2, 3, 4, 5], 'values': [42, 73, 48, 20, 19, 61]}, 
{'index': [0, 1, 2, 3, 4, 5], 'values': [13, 85, 39, 4, 93, 68]}, 
{'index': [0, 1, 2, 3, 4, 5], 'values': [43, 74, 49, 21, 20, 62]}
]