[youtube:subscriptions] Fix extraction (fixes #2975)

This commit is contained in:
Jaime Marquínez Ferrándiz 2014-08-31 23:44:43 +02:00
parent 49e23e8b6a
commit 1ed5b5c9c8

View file

@ -1430,12 +1430,6 @@ class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
paging = mobj.group('paging') paging = mobj.group('paging')
return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE) return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE)
class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
_VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
_FEED_NAME = 'subscriptions'
_PLAYLIST_TITLE = u'Youtube Subscriptions'
class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor): class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
IE_DESC = u'YouTube.com recommended videos, "ytrec" keyword (requires authentication)' IE_DESC = u'YouTube.com recommended videos, "ytrec" keyword (requires authentication)'
_VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?' _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
@ -1468,6 +1462,43 @@ class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
return self.url_result(playlist_id, 'YoutubePlaylist') return self.url_result(playlist_id, 'YoutubePlaylist')
class YoutubeSubscriptionsIE(YoutubePlaylistIE):
IE_NAME = u'youtube:subscriptions'
IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
_VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
def _real_extract(self, url):
title = u'Youtube Subscriptions'
page = self._download_webpage('https://www.youtube.com/feed/subscriptions', title)
# The extraction process is the same as for playlists, but the regex
# for the video ids doesn't contain an index
ids = []
more_widget_html = content_html = page
for page_num in itertools.count(1):
matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
new_ids = orderedSet(matches)
ids.extend(new_ids)
mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
if not mobj:
break
more = self._download_json(
'https://youtube.com/%s' % mobj.group('more'), title,
'Downloading page #%s' % page_num,
transform_source=uppercase_escape)
content_html = more['content_html']
more_widget_html = more['load_more_widget_html']
return {
'_type': 'playlist',
'title': title,
'entries': self._ids_to_results(ids),
}
class YoutubeTruncatedURLIE(InfoExtractor): class YoutubeTruncatedURLIE(InfoExtractor):
IE_NAME = 'youtube:truncated_url' IE_NAME = 'youtube:truncated_url'
IE_DESC = False # Do not list IE_DESC = False # Do not list