Akkariin/youtube-dl
1
0
Fork 0
mirror of https://github.com/l1ving/youtube-dl synced 2025-03-13 14:37:17 +08:00
Code Issues Releases Wiki Activity

[youtube] Don't extract the wrong channel playlist for geoblocked channels; fixes #10009

Browse Source
This commit is contained in:
Ivan Kozik 2016-07-05 07:18:24 +00:00
parent 4080530624
commit 9c48ba9c96
1 changed files with 15 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

24
youtube_dl/extractor/youtube.py
View File

@ -1983,21 +1983,27 @@ class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
url = self._TEMPLATE_URL % channel_id
# Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
# Workaround by extracting as a playlist if managed to obtain channel playlist URL
# otherwise fallback on channel by page extraction
# On channel pages, YouTube has a 'Load more' button, but it serves
# only 35 pages of 30 items i.e. 1050 videos total (issue #5778).
# Work around this by grabbing the channel's upload playlist instead.
# If no channel ID is present, fall back to extracting 'Load more' data.
channel_page = self._download_webpage(
url + '?view=57', channel_id,
'Downloading channel page', fatal=False)
url, channel_id, 'Downloading channel page', fatal=False)
if channel_page is False:
channel_playlist_id = False
else:
channel_playlist_id = self._html_search_meta(
'channelId', channel_page, 'channel id', default=None)
if not channel_playlist_id:
channel_playlist_id = self._search_regex(
r'data-(?:channel-external-|yt)id="([^"]+)"',
channel_page, 'channel id', default=None)
# Extract the (hopefully) same ID from elsewhere on the page, just
# to make sure we don't start downloading the wrong channel when
# YouTube serves wonky HTML.
channel_playlist_id_confirmation = self._search_regex(
r'<link rel="alternate" type="application/rss\+xml" title="RSS"' +
r' href="https://www\.youtube\.com/feeds/videos\.xml\?channel_id=([^"]+)">',
channel_page, 'channel id', default=None)
if channel_playlist_id != channel_playlist_id_confirmation:
raise ExtractorError('[youtube] Channel playlist mismatch: %r != %r' % (
channel_playlist_id, channel_playlist_id_confirmation))
if channel_playlist_id and channel_playlist_id.startswith('UC'):
playlist_id = 'UU' + channel_playlist_id[2:]
return self.url_result(