diff options
author | Adam Thalhammer <s3544305@student.rmit.edu.au> | 2016-05-02 13:21:39 +1000 |
---|---|---|
committer | Adam Thalhammer <s3544305@student.rmit.edu.au> | 2016-05-02 13:21:39 +1000 |
commit | 79a2e94e79e65cdf4898bc2dedb6a1bb4ca9af3c (patch) | |
tree | 97cd368a2089b073d2f0c1638bc66bd75fe11443 /youtube_dl | |
parent | 686cc8963441c37105c0447f31c5ea21405be05a (diff) |
Instead of replacing accented characters with an underscore when sanitizing file names in restricted mode, replace them with their non-accented equivalents fixes #9347
Diffstat (limited to 'youtube_dl')
-rw-r--r-- | youtube_dl/utils.py | 9 |
1 files changed, 7 insertions, 2 deletions
diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index 7bcc85e2b..f74f62268 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -14,8 +14,8 @@ import email.utils import errno import functools import gzip -import itertools import io +import itertools import json import locale import math @@ -24,8 +24,8 @@ import os import pipes import platform import re -import ssl import socket +import ssl import struct import subprocess import sys @@ -365,6 +365,11 @@ 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 """ def replace_insane(char): + accents = dict(zip('ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖØÙÚÛÜÝÞßàáâãäåæçèéêëìíîïðñòóôõöøùúûüýþÿ', + itertools.chain('AAAAAA', ['AE'], 'CEEEEIIIIDNOOOOOOUUUUYP', ['ss'], + 'aaaaaa', ['ae'], 'ceeeeiiiionoooooouuuuypy'))) + if restricted and char in accents: + return accents[char] if char == '?' or ord(char) < 32 or ord(char) == 127: return '' elif char == '"': |