Madeorsk
/
youtube-dl
mirror of https://codeberg.org/polarisfm/youtube-dl
1
0
Fork 0
Code Issues Releases Wiki Activity

Extract additional playlist page type and video link in that playlist 

Some user pages have additional playlist formats, e.g.:
http://www.youtube.com/user/stanforduniversity#g/c/9D558D49CA734A02
and
http://www.youtube.com/user/stanforduniversity#p/c/9D558D49CA734A02

There is also a related URL format which refers to a single video
within those playlists, where both playlist and video ids are
included in the URL:
http://www.youtube.com/user/stanforduniversity#p/c/9D558D49CA734A02/0/Ps8jOj7diA0

Extract playlist and turn the URL into a format that is already
understood.
The third format, the single video, should actually belong to
YoutubeIE and handled here temporarily until a better fix is created.
This commit is contained in:
Gergely Imreh 2011-02-09 13:22:17 +08:00
parent f74e22ae28
commit 86d758c21c
1 changed files with 25 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

37
youtube-dl
View File

@ -2096,7 +2096,8 @@ class YahooSearchIE(InfoExtractor):
class YoutubePlaylistIE(InfoExtractor):
"""Information Extractor for YouTube playlists."""
_VALID_URL = r'(?:http://)?(?:\w+\.)?youtube.com/(?:(?:view_play_list|my_playlists|artist)\?.*?(p|a)=|user/.*?/user/|p/)([^&]+).*'
_VALID_URL = r'(?:http://)?(?:\w+\.)?youtube.com/(?:(?:view_play_list|my_playlists|artist)\?.*?(?P<Pre1>p|a)=|user/.*?/user/|p/|user/.*?#(?P<Pre2>g|p)/c/)(?P<ID>[^&]+).*'
_COMBO_ID = r'(?:[^&]+)/(?:[^&]+)/(?P<ID>[^&]+)'
_TEMPLATE_URL = 'http://www.youtube.com/%s?%s=%s&page=%s&gl=US&hl=en'
_VIDEO_INDICATOR = r'/watch\?v=(.+?)&'
_MORE_PAGES_INDICATOR = r'(?m)>\s*Next\s*</a>'
@ -2125,17 +2126,28 @@ class YoutubePlaylistIE(InfoExtractor):
return
# Download playlist pages
# prefix is 'p' as default for playlists but there are other types that need extra care
playlist_prefix = mobj.group(1)
if playlist_prefix == 'a':
playlist_access = 'artist'
else:
playlist_access = 'view_play_list'
playlist_id = mobj.group(2)
playlist_prefix = mobj.group('Pre1')
playlist_altprefix = mobj.group('Pre2')
playlist_id = mobj.group('ID')
is_playlist = True
video_ids = []
pagenum = 1
while True:
# prefix is 'p' as default for playlists but there are other types that need extra care
if playlist_prefix == 'a':
playlist_access = 'artist'
else:
if playlist_altprefix == 'p':
# Not really a playlist but single video within the list:
ids = re.match(self._COMBO_ID, playlist_id)
if ids is not None:
is_playlist = False
video_ids = [ids.group('ID')]
if is_playlist:
playlist_prefix = 'p'
playlist_access = 'view_play_list'
while is_playlist:
self.report_download_page(playlist_id, pagenum)
request = urllib2.Request(self._TEMPLATE_URL % (playlist_access, playlist_prefix, playlist_id, pagenum))
try:
@ -2155,9 +2167,10 @@ class YoutubePlaylistIE(InfoExtractor):
break
pagenum = pagenum + 1
playliststart = self._downloader.params.get('playliststart', 1) - 1
playlistend = self._downloader.params.get('playlistend', -1)
video_ids = video_ids[playliststart:playlistend]
if is_playlist:
playliststart = self._downloader.params.get('playliststart', 1) - 1
playlistend = self._downloader.params.get('playlistend', -1)
video_ids = video_ids[playliststart:playlistend]
for id in video_ids:
self._youtube_ie.extract('http://www.youtube.com/watch?v=%s' % id)