I want to generate all config from a dict like this:

dict = {
  "a": [1,2,3],
  "b":{
    "b1":[True, False],
    "b2":[0],
  }
}

List attributes are needed to be enumerated. And output is like this:

config = [{
  "a": 1,
  "b":{
    "b1":True,
    "b2":0,
  },
  {
  "a": 2,
  "b":{
    "b1":True,
    "b2":0,
  },
...
]

How can I reach this?

I think recursion is a good idea, but I don't know how to use it

CodePudding user response：

There is no value in using recursion for this. A straightforward list comprehension will suffice:

_dict = {
    "a": [1, 2, 3],
    "b": {
        "b1": [True, False],
        "b2": [0]
    }
}

config = [{'a': k, 'b': {'b1': True, 'b2': 0}} for k in _dict['a']]

print(config)

Output:

[{'a': 1, 'b': {'b1': True, 'b2': 0}}, {'a': 2, 'b': {'b1': True, 'b2': 0}}, {'a': 3, 'b': {'b1': True, 'b2': 0}}]

CodePudding user response：

def merge_dicts(*dict_args):
    result = {}
    for dictionary in dict_args:
        result.update(dictionary)
    return result

def dict2combinations(d):
    for k, v in d.items():
        if isinstance(v, dict):
            d[k] = dict2combinations(v)
    values = [[{k: i} for i in v] for k, v in d.items()]
    return [merge_dicts(*i) for i in product(*values)]

print(dict2combinations(config))

Output:

[{'a': 1, 'b': {'b1': True, 'b2': 0}}, {'a': 1, 'b': {'b1': False, 'b2': 0}}, {'a': 2, 'b': {'b1': True, 'b2': 0}}, {'a': 2, 'b': {'b1': False, 'b2': 0}}, {'a': 3, 'b': {'b1': True, 'b2': 0}}, {'a': 3, 'b': {'b1': False, 'b2': 0}}]