Combinations from dictionary with list values using Python

I have the following incoming value:

variants = {
  "debug" : ["on", "off"],
  "locale" : ["de_DE", "en_US", "fr_FR"],
  ...
}

I want to process them so I get the following result:

combinations = [
  [{"debug":"on"},{"locale":"de_DE"}],
  [{"debug":"on"},{"locale":"en_US"}],
  [{"debug":"on"},{"locale":"fr_FR"}],
  [{开发者_StackOverflow中文版"debug":"off"},{"locale":"de_DE"}],
  [{"debug":"off"},{"locale":"en_US"}],
  [{"debug":"off"},{"locale":"fr_FR"}]
]

This should work with arbitrary length of keys in the dictionary. Played with itertools in Python, but did not found anything matching these requirements.

import itertools as it

varNames = sorted(variants)
combinations = [dict(zip(varNames, prod)) for prod in it.product(*(variants[varName] for varName in varNames))]

Hm, this returns:

[{'debug': 'on', 'locale': 'de_DE'},
 {'debug': 'on', 'locale': 'en_US'},
 {'debug': 'on', 'locale': 'fr_FR'},
 {'debug': 'off', 'locale': 'de_DE'},
 {'debug': 'off', 'locale': 'en_US'},
 {'debug': 'off', 'locale': 'fr_FR'}]

which is probably not exactly, what you want. Let me adapt it...

combinations = [ [ {varName: val} for varName, val in zip(varNames, prod) ] for prod in it.product(*(variants[varName] for varName in varNames))]

returns now:

[[{'debug': 'on'}, {'locale': 'de_DE'}],
 [{'debug': 'on'}, {'locale': 'en_US'}],
 [{'debug': 'on'}, {'locale': 'fr_FR'}],
 [{'debug': 'off'}, {'locale': 'de_DE'}],
 [{'debug': 'off'}, {'locale': 'en_US'}],
 [{'debug': 'off'}, {'locale': 'fr_FR'}]]

Voilà ;-)

combinations = [[{key: value} for (key, value) in zip(variants, values)] 
                for values in itertools.product(*variants.values())]

[[{'debug': 'on'}, {'locale': 'de_DE'}],
 [{'debug': 'on'}, {'locale': 'en_US'}],
 [{'debug': 'on'}, {'locale': 'fr_FR'}],
 [{'debug': 'off'}, {'locale': 'de_DE'}],
 [{'debug': 'off'}, {'locale': 'en_US'}],
 [{'debug': 'off'}, {'locale': 'fr_FR'}]]

This is what I use:

from itertools import product

def dictproduct(dct):
    for t in product(*dct.itervalues()):
        yield dict(zip(dct.iterkeys(), t))

which applied to your example gives:

>>> list(dictproduct({"debug":["on", "off"], "locale":["de_DE", "en_US", "fr_FR"]}))
[{'debug': 'on', 'locale': 'de_DE'},
 {'debug': 'on', 'locale': 'en_US'},
 {'debug': 'on', 'locale': 'fr_FR'},
 {'debug': 'off', 'locale': 'de_DE'},
 {'debug': 'off', 'locale': 'en_US'},
 {'debug': 'off', 'locale': 'fr_FR'}]

I find this is more readable than the one liners above.

Also, it returns an iterator like itertools.product so it leaves it up to the user whether to instantiate a list or just consume the values one at a time.

I assume you want the cartesian product of all the keys? So if you had another entry, "foo", with values [1, 2, 3], then you'd have 18 total entries?

First, put the values in a list, where each entry is one of the possible variants in that spot. In your case, we want:

[[{'debug': 'on'}, {'debug': 'off'}], [{'locale': 'de_DE'}, {'locale': 'en_US'}, {'locale': 'fr_FR'}]]

To do that:

>>> stuff = []
>>> for k,v in variants.items():
    blah = []
    for i in v:
        blah.append({k:i})
    stuff.append(blah)


>>> stuff
[[{'debug': 'on'}, {'debug': 'off'}], [{'locale': 'de_DE'}, {'locale': 'en_US'}, {'locale': 'fr_FR'}]]

Next we can use a Cartesian product function to expand it...

>>> def cartesian_product(lists, previous_elements = []):
if len(lists) == 1:
    for elem in lists[0]:
        yield previous_elements + [elem, ]
else:
    for elem in lists[0]:
        for x in cartesian_product(lists[1:], previous_elements + [elem, ]):
            yield x


>>> list(cartesian_product(stuff))
[[{'debug': 'on'}, {'locale': 'de_DE'}], [{'debug': 'on'}, {'locale': 'en_US'}], [{'debug': 'on'}, {'locale': 'fr_FR'}], [{'debug': 'off'}, {'locale': 'de_DE'}], [{'debug': 'off'}, {'locale': 'en_US'}], [{'debug': 'off'}, {'locale': 'fr_FR'}]]

Note that this doesn't copy the dicts, so all the {'debug': 'on'} dicts are the same.