10

I have a tuple in the following format:

(639283, 298290710, 1385)
(639283, 298290712, 1389)
(639283, 298290715, 1395)
(745310, 470212995, 2061)
(745310, 470213821, 3713)
(745310, 470215360, 6791)
(745310, 470215361, 6793)
(745310, 470215363, 6797)
(911045, 374330803, 4905)
(911045, 374330804, 4907)
(911045, 374330807, 4913)
(911045, 374330808, 4915)
(911045, 374330809, 4917)

I want to convert into a nested dictionary like this:

{639283:{298290710:1385, 298290712:1389, 298290715:1395},745310:{470212995:2061,470213821:3713}............}

Is there a pythonic way of doing this? It seems pretty simple, but i can't seem to figure this out.

msakya
  • 9,311
  • 5
  • 23
  • 31

2 Answers2

11

You can use tuple unpacking combined with collections.defaultdict to make your life easier.

Create an outer defaultdict with dict as its default value. Then, you can simply loop through your list of tuples once, setting the values appropriately as you go.

from collections import defaultdict

d = defaultdict(dict) # dict where the default values are dicts.
for a, b, c in list_of_tuples: # Each tuple is "key1, key2, value"
    d[a][b] = c

Of course, you presumably know more about what these values actually represent, so you can give your dictionary, and the individual items, better, more descriptive names than a, b, c, and d.

Henry Keiter
  • 16,863
  • 7
  • 51
  • 80
  • 2
    I think both response are great, but this seems rather intuitive and clean. So, I am accepting this :) – msakya Feb 17 '14 at 20:24
10

You can use itertools.groupby to group the tuples based on the first item, and then iterate over those groups in a dict-comprehension to get the desired result.

>>> from operator import itemgetter
>>> from pprint import pprint
>>> from itertools import groupby
>>> d = {k: dict(x[1:] for x in g) for k, g in groupby(data, key=itemgetter(0))}
>>> pprint(d)
{639283: {298290710: 1385, 298290712: 1389, 298290715: 1395},
 745310: {470212995: 2061,
          470213821: 3713,
          470215360: 6791,
          470215361: 6793,
          470215363: 6797},
 911045: {374330803: 4905,
          374330804: 4907,
          374330807: 4913,
          374330808: 4915,
          374330809: 4917}}

Where data is your list of tuples or tuple of tuples.

Ashwini Chaudhary
  • 244,495
  • 58
  • 464
  • 504