github/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-12-18 01:44:13 -05:00
Code Issues Projects Releases Packages Wiki Activity

improved performance by extracting accented chars to top level

Browse source
This commit is contained in:
Adam Thalhammer 2016-05-03 10:40:30 +10:00
parent 31c4448f6e
commit c587cbb793
1 changed files with 7 additions and 5 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

12
youtube_dl/utils.py
View file

@ -89,6 +89,11 @@
'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.
@ -365,11 +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):
accents = dict(zip('ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖØÙÚÛÜÝÞßàáâãäåæçèéêëìíîïðñòóôõöøùúûüýþÿ', if restricted and char in ACCENT_CHARS:
itertools.chain('AAAAAA', ['AE'], 'CEEEEIIIIDNOOOOOOUUUUYP', ['ss'], return ACCENT_CHARS[char]
'aaaaaa', ['ae'], 'ceeeeiiiionoooooouuuuypy')))
if restricted and char in accents:
return accents[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 == '"':