CODEWITHSUNDEEP
pythonpython-2.7dictionarypermutation
Nicholas Kellas
Nicholas Kellas

Reputation: 29

Generating dictionaries that are permutations of larger dictionaries

I don't know if the title I gave does a good job of explaining what I need my program to do. I will have dictionaries of the form:

main_dictionary = {1: [ [a], [b], [c], ... ], 2: [ [a], [b], [c], ... ] , ... }

The dictionary can be arbitrarily long and have an arbitrary number of options for each key. I need the program to then make every permutation of this dictionary to test.

sub_dictionary = {1: [a], 2: [a], ... }

test_program(sub_dictionary)

sub_dictionary = {1: [a], 2: [b], ... }

test_program(sub_dictionary)

Upvotes: 1

Views: 72

Answers (1)

jpp
jpp

Reputation: 164823

Here is one way using itertools.product. The result is a list of "sub-dictionaries".

For simplicity, I have used a list of integers as values, but this can be replaced by values of your choice.

from itertools import product

d = {1: [3, 4, 5], 2: [6, 7, 8]}

values = list(zip(*sorted(d.items())))[1]

res = [dict(enumerate(x, 1)) for x in product(*values)]

If you need to test each dictionary individually, use a generator expression instead and iterate it:

for item in (dict(enumerate(x, 1)) for x in product(*values)):
    ...

If you have string keys:

res = [dict(zip(sorted(d), x)) for x in product(*values)]

Result:

[{1: 3, 2: 6},
 {1: 3, 2: 7},
 {1: 3, 2: 8},
 {1: 4, 2: 6},
 {1: 4, 2: 7},
 {1: 4, 2: 8},
 {1: 5, 2: 6},
 {1: 5, 2: 7},
 {1: 5, 2: 8}]

Upvotes: 1

Related Questions