Merge pull request #9367 from codesparkle/master

Feature: --restrict-filenames: replace accented characters by their unaccented counterpart instead of "_"
This commit is contained in:
Yen Chi Hsuan 2016-05-06 01:44:03 +08:00
commit 109db8ea64
2 changed files with 15 additions and 4 deletions

View File

@ -139,8 +139,8 @@ class TestUtil(unittest.TestCase):
self.assertEqual('yes_no', sanitize_filename('yes? no', restricted=True)) self.assertEqual('yes_no', sanitize_filename('yes? no', restricted=True))
self.assertEqual('this_-_that', sanitize_filename('this: that', restricted=True)) self.assertEqual('this_-_that', sanitize_filename('this: that', restricted=True))
tests = 'a\xe4b\u4e2d\u56fd\u7684c' tests = 'aäb\u4e2d\u56fd\u7684c'
self.assertEqual(sanitize_filename(tests, restricted=True), 'a_b_c') self.assertEqual(sanitize_filename(tests, restricted=True), 'aab_c')
self.assertTrue(sanitize_filename('\xf6', restricted=True) != '') # No empty filename self.assertTrue(sanitize_filename('\xf6', restricted=True) != '') # No empty filename
forbidden = '"\0\\/&!: \'\t\n()[]{}$;`^,#' forbidden = '"\0\\/&!: \'\t\n()[]{}$;`^,#'
@ -155,6 +155,10 @@ class TestUtil(unittest.TestCase):
self.assertTrue(sanitize_filename('-', restricted=True) != '') self.assertTrue(sanitize_filename('-', restricted=True) != '')
self.assertTrue(sanitize_filename(':', restricted=True) != '') self.assertTrue(sanitize_filename(':', restricted=True) != '')
self.assertEqual(sanitize_filename(
'ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖØÙÚÛÜÝÞßàáâãäåæçèéêëìíîïðñòóôõöøùúûüýþÿ', restricted=True),
'AAAAAAAECEEEEIIIIDNOOOOOOUUUUYPssaaaaaaaeceeeeiiiionoooooouuuuypy')
def test_sanitize_ids(self): def test_sanitize_ids(self):
self.assertEqual(sanitize_filename('_n_cd26wFpw', is_id=True), '_n_cd26wFpw') self.assertEqual(sanitize_filename('_n_cd26wFpw', is_id=True), '_n_cd26wFpw')
self.assertEqual(sanitize_filename('_BD_eEpuzXw', is_id=True), '_BD_eEpuzXw') self.assertEqual(sanitize_filename('_BD_eEpuzXw', is_id=True), '_BD_eEpuzXw')

View File

@ -14,8 +14,8 @@ import email.utils
import errno import errno
import functools import functools
import gzip import gzip
import itertools
import io import io
import itertools
import json import json
import locale import locale
import math import math
@ -24,8 +24,8 @@ import os
import pipes import pipes
import platform import platform
import re import re
import ssl
import socket import socket
import ssl
import struct import struct
import subprocess import subprocess
import sys import sys
@ -89,6 +89,11 @@ KNOWN_EXTENSIONS = (
'wav', 'wav',
'f4f', 'f4m', 'm3u8', 'smil') 'f4f', 'f4m', 'm3u8', 'smil')
# needed for sanitizing filenames in restricted mode
ACCENT_CHARS = dict(zip('ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖØÙÚÛÜÝÞßàáâãäåæçèéêëìíîïðñòóôõöøùúûüýþÿ',
itertools.chain('AAAAAA', ['AE'], 'CEEEEIIIIDNOOOOOOUUUUYP', ['ss'],
'aaaaaa', ['ae'], 'ceeeeiiiionoooooouuuuypy')))
def preferredencoding(): def preferredencoding():
"""Get preferred encoding. """Get preferred encoding.
@ -365,6 +370,8 @@ def sanitize_filename(s, restricted=False, is_id=False):
Set is_id if this is not an arbitrary string, but an ID that should be kept if possible Set is_id if this is not an arbitrary string, but an ID that should be kept if possible
""" """
def replace_insane(char): def replace_insane(char):
if restricted and char in ACCENT_CHARS:
return ACCENT_CHARS[char]
if char == '?' or ord(char) < 32 or ord(char) == 127: if char == '?' or ord(char) < 32 or ord(char) == 127:
return '' return ''
elif char == '"': elif char == '"':