Consolidate dictionary based on repeat values

I have a dictionary set up like this

{
     "key1" : [1,2,4],
     "key2" : [2,4],
     "key3" : [1,2,4],
  开发者_JAVA技巧   "key4" : [2,4],
     ....
}

What I want is something like this.

[
  [ 
     ["key1", "key3"],
     [1,2,4],
  ],
  [ 
     ["key2", "key4"],
     [2,4],
  ],
  .....
]

A list of keys and values based on unique value pairs. How can I do this in a pythonic way?

You can invert the dictionnary like this :

orig = {
     "key1" : [1,2,4],
     "key2" : [2,4],
     "key3" : [1,2,4],
     "key4" : [2,4],
}

new_dict = {}

for k, v in orig.iteritems():
    new_dict.setdefault(tuple(v), []).append(k)    #need to "freeze" the mutable type into an immutable to allow it to become a dictionnary key (hashable object)

# Here we have new_dict like this :
#new_dict = {
#    (2, 4): ['key2', 'key4'],
#    (1, 2, 4): ['key3', 'key1']
#}

# like sverre suggested :
final_output = [[k,v] for k,v in new_dict.iteritems()]

Here is a list comprehension to do the job cleanly:

[[[key for key in dictionary.keys() if dictionary[key] == value], value]
   for value in unique(list(dictionary.values()))]

Where unique can be a function that returns the unique elements of a list. There is no default for this, but there are many implementations (here are some).

Please find my sample code below if it is still actual for you:

orig = {
     "key1" : [1,2,4],
     "key2" : [2,4],
     "key3" : [1,2,4],
     "key4" : [2,4],
}

unique = map(list, set(map(tuple, orig.values())))
print map(lambda val: [val, filter(lambda key: orig[key] == val, orig)], unique)