How to create subset lists with underscores from a list Python

Question

I have a list called animals,

animals = ["B_FOX", "A_CAT", "A_DOG", "A_MOUSE", 
         "B_DOG", "B_MOUSE", "C_DUCK", "C_FOX", "C_BIRD"]

and would like the following outputs:

 A = ["A_CAT", "A_DOG", "A_MOUSE"]
 B = ["B_DOG", "B_MOUSE", "B_FOX"]
 C = ["C_DUCK", "C_FOX", "C_BIRD"]

I can only get a subset list of only the letters or the animals like this:

  [species.split("_",1)[1] for species in animals]
  ['FOX', 'CAT', 'DOG', 'MOUSE', 'DOG', 'MOUSE', 'DUCK', 'FOX', 'BIRD']

  [letters.split("_",1)[0] for letters in animals]
  ['B', 'A', 'A', 'A', 'B', 'B', 'C', 'C', 'C']

Not sure if I've worded the question correctly. Any help in solving this tricky problem would be greatly appreciated!

explain how you would imagine proceeding from here in words and you'll find help. — Mad Physicist, Aug 29 '17 at 02:42

score 3 · Answer 1 · answered Aug 29 '17 at 02:43

3

You could build separate lists, one for each initial letter, however, that would be tricky if you have many letters. You can use a defaultdict instead:

from collections import defaultdict

d = defaultdict(list)
animals = ["B_FOX", "A_CAT", "A_DOG", "A_MOUSE", 
     "B_DOG", "B_MOUSE", "C_DUCK", "C_FOX", "C_BIRD"]

for animal in animals:
   d[animal[0]].append(animal)
print(dict(d))

Output:

{'A': ['A_CAT', 'A_DOG', 'A_MOUSE'], 'C': ['C_DUCK', 'C_FOX', 'C_BIRD'], 'B': ['B_FOX', 'B_DOG', 'B_MOUSE']}

answered Aug 29 '17 at 02:43

Ajax1234

69,937
8
61
102

Hey Ajax, thanks for the help! The output is cool, seems to work. – MichaelRSF Aug 29 '17 at 02:52
@MichaelRSF glad to help! – Ajax1234 Aug 29 '17 at 02:54

pylang · Answer 2 · 2017-08-29T03:23:11.847

3

Try an itertools.groupby according to the first letter:

import operator as op
import itertools as it


animals = [
    "B_FOX", "A_CAT", "A_DOG", "A_MOUSE", 
    "B_DOG", "B_MOUSE", "C_DUCK", "C_FOX", "C_BIRD"
]

A, B, C = [list(g) for _, g in it.groupby(sorted(animals), key=op.itemgetter(0))]

Outputs:

A
# ['A_CAT', 'A_DOG', 'A_MOUSE']

B
# ['B_DOG', 'B_FOX', 'B_MOUSE']

C
# ['C_BIRD', 'C_DUCK', 'C_FOX']

Here is a post on how groupby works.

edited Aug 29 '17 at 03:23

answered Aug 29 '17 at 02:47

pylang

40,867
14
129
121

3

That's very neat, I wasn't aware of the operator module til now. Thanks pylang! – MichaelRSF Aug 29 '17 at 02:55
The other key option is a regular function, such as `def f(x): x[0]` or `lambda x: x[0]` – pylang Aug 29 '17 at 02:57
This would be last on readability and ease of understanding but it definitely wins for the shortest solution ;-) – Hubert Grzeskowiak Aug 29 '17 at 03:06
@Hubert Grzeskowiak, I might agree on understanding, which is how `groupby` is. It is one of the tougher and trickier itertools to grok. But, I think it reads fine. I'll add a link for clarity. Thanks. – pylang Aug 29 '17 at 03:10
Thanks for your solution as well, Hubert! Much appreciated for all the help :) – MichaelRSF Aug 29 '17 at 03:11

Hubert Grzeskowiak · Accepted Answer · 2017-08-29T03:04:16.183

You can unpack the values of the prefix and the name both from one call to split:

groups = {}
for animal in animals:
    prefix, name = animal.split("_")
    if prefix not in groups:
        groups[prefix] = []
    groups[prefix].append(animal)

print groups

{'A': ['A_CAT', 'A_DOG', 'A_MOUSE'], 'C': ['C_DUCK', 'C_FOX', 'C_BIRD'], 'B': ['B_FOX', 'B_DOG', 'B_MOUSE']}

If required, you can later still unpack the dict into single variables:

A = groups["A"]
B = groups["B"]
C = groups["C"]

If you want to get rid of the prefixes:

groups = {}
for animal in animals:
    prefix, name = animal.split("_")
    if prefix not in groups:
        groups[prefix] = []
    groups[prefix].append(name)

How to create subset lists with underscores from a list Python

3 Answers3