Created
November 26, 2009 01:35
-
-
Save lqc/243163 to your computer and use it in GitHub Desktop.
Buggy locale sorting in python
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
# -*- coding: utf-8 -*- | |
from __future__ import print_function | |
import sys | |
import locale | |
locale.setlocale(locale.LC_ALL, '') | |
print("Python version:", sys.version_info) | |
print("LC_ALL =", locale.getlocale(locale.LC_ALL)) | |
print("LC_COLLATE =", locale.getlocale(locale.LC_COLLATE)) | |
encoding = locale.getpreferredencoding() | |
print("Prefered encoding is:", encoding) | |
print() | |
alphabet = [ | |
"a", "ą", "b", "c", "ć", | |
"d", "e", "ę", "f", "g", | |
"h", "i", "j", "k", "l", | |
"ł", "m", "n", "ń", "o", | |
"ó", "p", "r", "s", "ś", | |
"u", "v", "w", "y", "z", | |
"ź", "ż" | |
] | |
# Convert to unicode in python 2.x | |
if isinstance(alphabet[0], bytes): | |
alphabet = [ x.decode('utf-8') for x in alphabet ] | |
def run_test(label, sort_filter): | |
print(label) | |
try: | |
result = sort_filter(alphabet) | |
print(' '.join(result)) | |
if result != alphabet: | |
print("FAILED: Bad result.") | |
except TypeError as e: | |
print("FAILED: Exception:", e) | |
except Exception as e: | |
print("FAILED: Exception:", e) | |
finally: | |
print() | |
# the original | |
run_test("Sorted alphabet", lambda x: x) | |
# Most efficient would be using strxfrm as key on the original values | |
# | |
# This is broken on Windows (locale: Polish_Poland) | |
# - on 2.6.4 in should fail with UnicodeDecodeError, instead yields bad results | |
# - on 3.1.1 yields bad results | |
# | |
# On Linux (locale pl_PL.UTF8): | |
# - works in 3.1.1 and trunk | |
# - yields the expected Exception in 2.6.4 | |
run_test('Key=strxfrm(unicode)', lambda x: sorted(x, key=locale.strxfrm)) | |
# Second option is to use strcoll (you can't in py3k) | |
# | |
# Works in 2.6.4 fine everywhere (why doesn't this yield the UnicodeDecodeError ?!) | |
# | |
run_test('Cmp=strcoll(unicode)', lambda x: sorted(x, cmp=locale.strcoll)) | |
print("A is before Z", (locale.strcoll('a', 'z') < 0) ) | |
print("P is after K", (locale.strcoll('p', 'k') > 0) ) | |
print("Ą is before B", (locale.strcoll('ą', 'b') < 0) ) | |
print() | |
# Next guess is to use strxfrm on strings encoded in the native coding | |
# Works in 2.6.4 - both Linux and Windows | |
# 3.1.1-Win: wrong anwser | |
# 3.1.1-Linux: throws exception (as it should) | |
encoded_key = lambda x: locale.strxfrm(x.encode(encoding)) | |
run_test("Key=strxfrm(bytes_using_preferred_encoding)", lambda x: sorted(x, key=encoded_key)) | |
encoded_cmp = lambda x, y: locale.strcoll(x.encode(encoding), y.encode(encoding)) | |
run_test("Cmp=strcoll(bytes_using_preferred_encoding)", lambda x: sorted(x, cmp=encoded_cmp)) | |
print("A is before Z", (locale.strcoll('a'.encode(encoding), 'z'.encode(encoding)) < 0) ) | |
print("P is after K", (locale.strcoll('p'.encode(encoding), 'k'.encode(encoding)) > 0) ) | |
print("Ą is before B", (locale.strcoll('ą'.encode(encoding), 'b'.encode(encoding)) < 0) ) | |
print() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment