Roxedus/ytdl
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

[youtube] Fix mix video regex

Browse Source 
Attributes' order in <li> is arbitrary and changes every time playlist
page is fetched, so we can't rely on `data-index` to be before
`data-video-username`.
This commit is contained in:
Sergey M. 2014-02-24 12:51:06 +07:00
parent 92661c994b
commit a2dafe2887
1 changed files with 2 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
youtube_dl/extractor/youtube.py
View File

@ -1492,11 +1492,10 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
title_span = (search_title('playlist-title') or
search_title('title long-title') or search_title('title'))
title = clean_html(title_span)
video_re = r'''(?x)data-index="\d+".*?
data-video-username="(.*?)".*?
video_re = r'''(?x)data-video-username="(.*?)".*?
href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?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
# Some of the videos may have been deleted, their username field is empty
ids = [video_id for (username, video_id) in matches if username]
url_results = self._ids_to_results(ids)