administration/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-11-14 05:03:09 +00:00
Code Issues Projects Releases Wiki Activity

[youtube:playlist] Fix mixes extraction (fixes #2444)

Browse Source
This commit is contained in:
Jaime Marquínez Ferrándiz 2014-02-23 17:17:36 +01:00
parent f919201ecc
commit bc2f773b4f
1 changed files with 9 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
youtube_dl/extractor/youtube.py
View File

@ -1488,11 +1488,16 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
# the id of the playlist is just 'RD' + video_id # the id of the playlist is just 'RD' + video_id
url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id) url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
webpage = self._download_webpage(url, playlist_id, u'Downloading Youtube mix') webpage = self._download_webpage(url, playlist_id, u'Downloading Youtube mix')
title_span = (get_element_by_attribute('class', 'title long-title', webpage) or search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
get_element_by_attribute('class', 'title ', webpage)) title_span = (search_title('playlist-title') or
search_title('title long-title') or search_title('title'))
title = clean_html(title_span) title = clean_html(title_span)
video_re = r'data-index="\d+".*?href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s' % re.escape(playlist_id) video_re = r'''(?x)data-index="\d+".*?
ids = orderedSet(re.findall(video_re, webpage)) data-video-username="(.*?)".*?
href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id)
matches = orderedSet(re.findall(video_re, webpage, flags=re.DOTALL))
# Some of the videos may have beend deleted, their username field is empty
ids = [video_id for (username, video_id) in matches if username]
url_results = self._ids_to_results(ids) url_results = self._ids_to_results(ids)
return self.playlist_result(url_results, playlist_id, title) return self.playlist_result(url_results, playlist_id, title)