What is the best way of creating an alphabetically sorted list in Python?
Basic answer:
mylist = ["b", "C", "A"]
mylist.sort()
This modifies your original list (i.e. sorts in-place). To get a sorted copy of the list, without changing the original, use the sorted()
function:
for x in sorted(mylist):
print x
However, the examples above are a bit naive, because they don't take locale into account, and perform a case-sensitive sorting. You can take advantage of the optional parameter key
to specify custom sorting order (the alternative, using cmp
, is a deprecated solution, as it has to be evaluated multiple times - key
is only computed once per element).
So, to sort according to the current locale, taking language-specific rules into account (cmp_to_key
is a helper function from functools):
sorted(mylist, key=cmp_to_key(locale.strcoll))
And finally, if you need, you can specify a custom locale for sorting:
import locale
locale.setlocale(locale.LC_ALL, 'en_US.UTF-8') # vary depending on your lang/locale
assert sorted((u'Ab', u'ad', u'aa'),
key=cmp_to_key(locale.strcoll)) == [u'aa', u'Ab', u'ad']
Last note: you will see examples of case-insensitive sorting which use the lower()
method - those are incorrect, because they work only for the ASCII subset of characters. Those two are wrong for any non-English data:
# this is incorrect!
mylist.sort(key=lambda x: x.lower())
# alternative notation, a bit faster, but still wrong
mylist.sort(key=str.lower)
It is also worth noting the sorted()
function:
for x in sorted(list):
print x
This returns a new, sorted version of a list without changing the original list.
The proper way to sort strings is:
import locale
locale.setlocale(locale.LC_ALL, 'en_US.UTF-8') # vary depending on your lang/locale
assert sorted((u'Ab', u'ad', u'aa'), cmp=locale.strcoll) == [u'aa', u'Ab', u'ad']
# Without using locale.strcoll you get:
assert sorted((u'Ab', u'ad', u'aa')) == [u'Ab', u'aa', u'ad']
The previous example of mylist.sort(key=lambda x: x.lower())
will work fine for ASCII-only contexts.
But how does this handle language specific sorting rules? Does it take locale into account?
No, list.sort()
is a generic sorting function. If you want to sort according to the Unicode rules, you'll have to define a custom sort key function. You can try using the pyuca module, but I don't know how complete it is.
Old question, but if you want to do locale-aware sorting without setting locale.LC_ALL
you can do so by using the PyICU library as suggested by this answer:
import icu # PyICU
def sorted_strings(strings, locale=None):
if locale is None:
return sorted(strings)
collator = icu.Collator.createInstance(icu.Locale(locale))
return sorted(strings, key=collator.getSortKey)
Then call with e.g.:
new_list = sorted_strings(list_of_strings, "de_DE.utf8")
This worked for me without installing any locales or changing other system settings.
(This was already suggested in a comment above, but I wanted to give it more prominence, because I missed it myself at first.)