CODEWITHSUNDEEP
pythondictionary
user2943615
user2943615

Reputation: 113

How to update a key in a dictionary without overwriting the old value

input = { "a" : { "x": 1, "y": 2 },
     "b" : { "x": 3, "z": 4 } }
output = {'y': {'a': 2},
                'x': {'a': 1, 'b': 3},
                'z': {'b': 4} }

I need to make a new dictionary given the input and it should come out as the output. So far I have written:

def pivot_nested_dict(nested_dict):
5       stateDict={}
6       for i in nested_dict:
7           dict2=nested_dict[i]
8           for j in dict2:
9               stateDict[j]= {i: dict2[j]}
10      return stateDict
11  
12  print pivot_nested_dict(input)

I almost have the answer, output={'y': {'a': 2}, 'x': {'b': 3}, 'z': {'b': 4}}. So, the x key is overwriting the 'a': 1. How do I keep both the a and the b?

Upvotes: 1

Views: 6836

Answers (4)

kiriloff
kiriloff

Reputation: 26333

I have this one

input = { "a" : { "x": 1, "y": 2 },
     "b" : { "x": 3, "z": 4 } }

output = {}

for k,v in zip(input.keys(), input.values()):
    for kk, vv in zip(v.keys(), v.values()):
        if kk in output.keys():
            output[kk][k] = vv
        else:
            output[kk] = {k : vv}

Upvotes: 1

Sravan K Ghantasala
Sravan K Ghantasala

Reputation: 1318

I ve got this little script yo get your job done.
NOTE: Provided the input is correct.

code:

output = {}
for d in input.items():
    for item in d:
        if type(item) == dict:
            for i in item.keys():
                d = {k:item[i]}
                output[i] = {k:item[i]} if not i in output.keys() else dict(output[i].items() + d.items())
        else: k=item

input: 

input = { "a" : { "x": 1, "y": 2 ,'z':15},"b" : { "x": 3, "z": 4 } }

output:

{'y': {'a': 2}, 'x': {'a': 1, 'b': 3}, 'z': {'a': 15, 'b': 4}}

Hope this helps :)

Upvotes: 1

Christian Tapia
Christian Tapia

Reputation: 34146

What you need to this is check if the key is already in stateDict, if it is, then add a new element to the nested dictionary:

def pivot_nested_dict(nested_dict):
    stateDict = {}
    for i in nested_dict:
        dict2 = nested_dict[i]
        for j in dict2:
            if j not in stateDict:
                stateDict[j] = {i: dict2[j]}
            else:
                stateDict[j][i] = dict2[j]
    return stateDict

print pivot_nested_dict(inp)

Upvotes: 1

user764357
user764357

Reputation:

Short of duplicating the dictionary, there is no way to retain the original value on a dictionary once you change the value that corresponds to the given key.

For example, the below overwriting will always happen no matter how complex the data structure.

x = {'a':1}
x['a'] = 2
print x
>>> {'a': 2}

You can duplicate it, but the original value will need to be managed manually.

x = {'a':1}
x,oldx = x,x.copy()
oldx['a'] = 2
print x, oldx
>>> {'a': 2} {'a': 1}

You could take advantage of the fact that values can be any type of structure including arrays or dictionaries, to store each successive change of value.

Upvotes: 1

Related Questions