How to unique a dict by value?

Question

I want to unique duplicate values in a dict. It looks like this:

d = {
"a":1,
"b":2,
"c":2,
"d":3,
"e":4,
"f":5,
"g":1,
"h":2,
"i":2,
"j":1,
"k":1}

Here is what I did:

# sort and unique the dict values
obj = d.values()
K = []
K = sorted(list(zip(*[(x,K.append(x)) for x in obj if not x in K])[0]

V=[]
for v1 in L:  
     V.append([k for k, v in obj.iteritems() if v == v1][0])
d_out = dict(zip(K, V))  

1. So, will the K,V be in a right order? Also, it may a bit complex, can anyone give a simple solution to unique a dict by it's values?

2. Can the following be more simple?

for v1 in L:  
     V.append([k for k, v in obj.iteritems() if v == v1][0])

This is not working on my testing:

[V.append([k for k, v in obj.iteritems() if v == v1][0]) for v1 in L] 

3. I realized I may use the swap key value to achieve that (unique a dict by its value), but I have no idea how to select the key when swap caused a key conflict with this:

dict((value, key) for key, value in my_dict.iteritems())  

I know if swap it again the value will be unique, however, this just overwrites the key when a key conflict happens, giving no chance to make a selection. I feel confused why this gives no key conflict error? And can I do something to select the key beside the ugly way overwrite the new dict's key later?

4. I searched and find some "None" values for python dict are well discussed, anyone can give me a sample what it is used for and what it will impacted in using python dict?

Answer 1

try it out:

from collections import defaultdict
dout = defaultdict(dict)
for k,v in d.iteritems():
    dout[v] = k 
dout = dict(dout)
fdict = dict(zip(dout.values(), dout.keys()))

N.B: dictionary don't have duplicate key so input dictionary has no duplicate key hopefully it will works

Answer 2

1. A dict is not a sequence. There is no ordering.

2. You need a simpler overall approach.

3. A dict does not give a "key conflict error". It assumes that you want to overwrite the old value with the new value.

4. I don't understand what you're asking here.

The solution below is a more straightforward way of removing dupe values from a dictionary. Adjust either the sort or the insertion loop to control which keys should appear in the final dict.

d = {
    "a":1,
    "b":2,
    "c":2,
    "d":3,
    "e":4,
    "f":5,
    "g":1,
    "h":2,
    "i":2,
    "j":1,
    "k":1}

# Extract the dictionary into a list of (key, value) tuples.
t =  [(k, d[k]) for k in d]

# Sort the list -- by default it will sort by the key since it is
# first in the tuple.
t.sort()

# Reset the dictionary so it is ready to hold the new dataset.
d = {}

# Load key-values into the dictionary. Only the first value will be
# stored.
for k, v in t:
    if v in d.values():
        continue
    d[k] = v

print d

Answer 3

Maybe this is of some help:

import collections

d = ... # like above
d1 = collections.defaultdict(list)

for k, v in d.iteritems():
    d1[v].append(k)

print d1