Split dictionary according to key values-CodePudding

I have a dictionary that looks like this:

dict = {
        "name 1": "abc..",
        "location 1": "abc..",
        "name 2": "abc..",
        "location 2": "abc"
        .
        .
}

My goal is to split it into n other sub dictionaries (depending on the range of integers in the first dictionary), in order to obtain a result like this:

#output
output_dict = {
        "dict 1": {
                   "name 1" : "abc..",
                   "location 1" : "abc.."
                   },
        "dict 2": {
                  "name 2": "abc",
                  "location 2": "abc"
                  }
        .
        .
        }

CodePudding user response：

You can go through the dictionary's items and progressively fill the output dictionary by building the sub-dictionary keys from the original keys:

d = {
        "name 1": "abc..",
        "location 1": "abc..",
        "name 2": "abc..",
        "location 2": "abc"
    }

d2 = dict()                       # empty output
for k,v in d.items():             # go through all keys and values
    dk = f"dict {k.split()[-1]}"  # build sub-dict key
    d2.setdefault(dk,dict())[k]=v # add key:value to sub-dict

print(d2)
{'dict 1': {'name 1': 'abc..', 'location 1': 'abc..'},
 'dict 2': {'name 2': 'abc..', 'location 2': 'abc'}}

If you're allowed to use libraries, groupby (from itertools) can help with this:

from itertools import groupby

d2 = {dk:dict(g) for dk,g 
      in groupby(d.items(),lambda kv:f"dict {kv[0].split()[-1]}")}

CodePudding user response：

Iterate over the values, and assign them accordingly:

new_dict = {}
for k, v in my_dict.items():
    if "dict "   k.split()[1] in new_dict:
        new_dict["dict "   k.split()[1]][k] = v
    else:
        new_dict["dict "   k.split()[1]] = {k: v}

>>> new_dict
{ 'dict 1': { 
    'name 1': 'abc..', 
    'location 1': 'abc..'}, 
  'dict 2': { 
    'name 2': 'abc..', 
    'location 2': 'abc'}
}