I have a dictionary where I want to return a dynamic list of the keys that would access any non-dict values. I do not know the structure of the dictionary beforehand and nesting could go n-levels.
{
"parent": {
"a": 1,
"b": {
"nested": "first"
}
}
}
From this dictionary, I want to return the following list of lists:
[
["parent", "a"],
["parent", "b", "nested"]
]
I plan to use this list with functools and reduce to get specific values out of the dictionary to build a new dictionary. I've tried a recursion function to lop through the dictionary items and check for isinstance(value, dict), but I'm having issues keeping the path together. Any help is much appreciated, thanks!
You can add the paths of the sub-dicts generated from recursive calls with the dict values to the keys of the current dict. Since you don't need leaf values, yield an empty list for them:
def paths(d):
if isinstance(d, dict):
yield from ([k, *p] for k, v in d.items() for p in paths(v))
else:
yield []
so that:
d = {
"parent": {
"a": 1,
"b": {
"nested": "first"
}
}
}
print(list(paths(d)))
outputs:
[['parent', 'a'], ['parent', 'b', 'nested']]