Python sort list of lists over multiple levels and with a custom order

Question

I have a list of lists my_list e.g.

[
['item 3.1', 'item 3.2', 'item 3.3', 'item 3.4', 'item 3.5'],
['item 3.1', 'item 3.2', 'item 3.2', 'item 3.2', 'item 3.2'],
['item 3.1', 'item 3.2', 'item 3.2', 'item 3.6', 'item 3.2'],
['item 3.1', 'item 2.2', 'item 2.4', 'item 2.7', 'item 2.5'],
['item 2.1', 'item 2.2', 'item 2.3', 'item 2.4', 'item 2.5'],
['item 1.1', 'item 1.2', 'item 1.3', 'item 1.4', 'item 1.5'],
['item 4.1', 'item 4.2', 'item 4.3', 'item 4.4', 'item 4.5'],
['item 1.1', 'item 1.2', 'item 1.3', 'item 1.5', 'item 1.5']
]

I want to order this list on multiple levels with 1 of the levels being a customer order but don't know how to do this.

e.g. I want to order by the first item first then within that order, order them by the third item then within that order, order by the fourth item. But I want the first item to be ordered/sorted in this particular order rather than alphabetical:

item 4.1
item 2.1
item 3.1
item 1.1

so the final list (after sorting) would be:

[
['item 4.1', 'item 4.2', 'item 4.3', 'item 4.4', 'item 4.5']
['item 2.1', 'item 2.2', 'item 2.3', 'item 2.4', 'item 2.5'],
['item 3.1', 'item 3.2', 'item 3.2', 'item 3.2', 'item 3.2'],
['item 3.1', 'item 3.2', 'item 3.2', 'item 3.6', 'item 3.2'],
['item 3.1', 'item 3.2', 'item 3.3', 'item 3.4', 'item 3.5'],
['item 3.1', 'item 2.2', 'item 2.4', 'item 2.7', 'item 2.5'],
['item 1.1', 'item 1.2', 'item 1.3', 'item 1.4', 'item 1.5'],
['item 1.1', 'item 1.2', 'item 1.3', 'item 1.5', 'item 1.5'],
]

I know I can sort over multiple levels by doing:

s = sorted(my_list, key=itemgetter(0,2,3))

But this would sort them all alphabetically and what I don't know how to do is set the first column to be sorted by the custom order.

Answer 1

Ugly and inefficient, but probably working:

order = ['item 4.1', 'item 2.1', 'item 3.1', 'item 1.1']
my_list.sort(key=lambda x: order.index(x[0]))

reorders my_list to:

[['item 4.1', 'item 4.2', 'item 4.3', 'item 4.4', 'item 4.5'],
 ['item 2.1', 'item 2.2', 'item 2.3', 'item 2.4', 'item 2.5'],
 ['item 3.1', 'item 3.2', 'item 3.3', 'item 3.4', 'item 3.5'],
 ['item 3.1', 'item 3.2', 'item 3.2', 'item 3.2', 'item 3.2'],
 ['item 3.1', 'item 3.2', 'item 3.2', 'item 3.6', 'item 3.2'],
 ['item 3.1', 'item 2.2', 'item 2.4', 'item 2.7', 'item 2.5'],
 ['item 1.1', 'item 1.2', 'item 1.3', 'item 1.4', 'item 1.5'],
 ['item 1.1', 'item 1.2', 'item 1.3', 'item 1.5', 'item 1.5']]

If you want to sort by first column in a customized way, then by the third and fourth alphabetically, just expand the key:

my_list.sort(key=lambda x: (order.index(x[0]), x[2], x[3]))

which returns

[['item 4.1', 'item 4.2', 'item 4.3', 'item 4.4', 'item 4.5'],
 ['item 2.1', 'item 2.2', 'item 2.3', 'item 2.4', 'item 2.5'],
 ['item 3.1', 'item 2.2', 'item 2.4', 'item 2.7', 'item 2.5'],
 ['item 3.1', 'item 3.2', 'item 3.2', 'item 3.2', 'item 3.2'],
 ['item 3.1', 'item 3.2', 'item 3.2', 'item 3.6', 'item 3.2'],
 ['item 3.1', 'item 3.2', 'item 3.3', 'item 3.4', 'item 3.5'],
 ['item 1.1', 'item 1.2', 'item 1.3', 'item 1.4', 'item 1.5'],
 ['item 1.1', 'item 1.2', 'item 1.3', 'item 1.5', 'item 1.5']]