jcgruenhage/youtube-dl
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity

[youtube:channel] Fix extraction (fixes #4435)

Browse source 
It uses now the same pagination system as playlists
This commit is contained in:
Jaime Marquínez Ferrándiz 2014-12-12 22:23:54 +01:00
parent baa7081d68
commit 23d3608c6b
1 changed files with 13 additions and 8 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

21
youtube_dl/extractor/youtube.py
View file

@ -1269,8 +1269,6 @@ class YoutubeTopListIE(YoutubePlaylistIE):
class YoutubeChannelIE(InfoExtractor): class YoutubeChannelIE(InfoExtractor):
IE_DESC = 'YouTube.com channels' IE_DESC = 'YouTube.com channels'
_VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)' _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
_MORE_PAGES_INDICATOR = 'yt-uix-load-more'
_MORE_PAGES_URL = 'https://www.youtube.com/c4_browse_ajax?action_load_more_videos=1&flow=list&paging=%s&view=0&sort=da&channel_id=%s'
IE_NAME = 'youtube:channel' IE_NAME = 'youtube:channel'
_TESTS = [{ _TESTS = [{
'note': 'paginated channel', 'note': 'paginated channel',
@ -1307,20 +1305,27 @@ class YoutubeChannelIE(InfoExtractor):
return self.playlist_result(entries, channel_id) return self.playlist_result(entries, channel_id)
def _entries(): def _entries():
more_widget_html = content_html = channel_page
for pagenum in itertools.count(1): for pagenum in itertools.count(1):
url = self._MORE_PAGES_URL % (pagenum, channel_id)
page = self._download_json(
url, channel_id, note='Downloading page #%s' % pagenum,
transform_source=uppercase_escape)
ids_in_page = self.extract_videos_from_page(page['content_html']) ids_in_page = self.extract_videos_from_page(content_html)
for video_id in ids_in_page: for video_id in ids_in_page:
yield self.url_result( yield self.url_result(
video_id, 'Youtube', video_id=video_id) video_id, 'Youtube', video_id=video_id)
if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']: mobj = re.search(
r'data-uix-load-more-href="/?(?P<more>[^"]+)"',
more_widget_html)
if not mobj:
break break
more = self._download_json(
'https://youtube.com/%s' % mobj.group('more'), channel_id,
'Downloading page #%s' % (pagenum + 1),
transform_source=uppercase_escape)
content_html = more['content_html']
more_widget_html = more['load_more_widget_html']
return self.playlist_result(_entries(), channel_id) return self.playlist_result(_entries(), channel_id)