Browse Source

[extractor/common] Speed-up media tags regex (closes #11979)

master
Sergey M․ 7 years ago
parent
commit
2aec7256ae
No known key found for this signature in database GPG Key ID: 2C393E0F18A9236D
1 changed files with 6 additions and 1 deletions
  1. 7
      youtube_dl/extractor/common.py

7
youtube_dl/extractor/common.py

@ -1959,7 +1959,12 @@ class InfoExtractor(object):
media_tags = [(media_tag, media_type, '') media_tags = [(media_tag, media_type, '')
for media_tag, media_type for media_tag, media_type
in re.findall(r'(?s)(<(video|audio)[^>]*/>)', webpage)] in re.findall(r'(?s)(<(video|audio)[^>]*/>)', webpage)]
media_tags.extend(re.findall(r'(?s)(<(?P<tag>video|audio)[^>]*>)(.*?)</(?P=tag)>', webpage))
media_tags.extend(re.findall(
# We only allow video|audio followed by a whitespace or '>'.
# Allowing more characters may end up in significant slow down (see
# https://github.com/rg3/youtube-dl/issues/11979, example URL:
# http://www.porntrex.com/maps/videositemap.xml).
r'(?s)(<(?P<tag>video|audio)(?:\s+[^>]*)?>)(.*?)</(?P=tag)>', webpage))
for media_tag, media_type, media_content in media_tags: for media_tag, media_type, media_content in media_tags:
media_info = { media_info = {
'formats': [], 'formats': [],

Loading…
Cancel
Save