Madeorsk/youtube-dl
1
0
Fork 0
mirror of https://codeberg.org/polarisfm/youtube-dl synced 2025-01-07 13:47:54 +01:00
Code Issues Releases Wiki Activity

[bbc] Fix errors

Browse Source 
Fixes errors on news articles, archive pages, etc
This commit is contained in:
ajj8 2019-12-15 00:53:30 +00:00 committed by GitHub
parent 73d8f3a634
commit 04eb2f977c
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 13 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

24
youtube_dl/extractor/bbc.py
View File

@ -608,9 +608,10 @@ class BBCIE(BBCCoUkIE):
_VALID_URL = r'https?://(?:www\.)?bbc\.(?:com|co\.uk)/(?:[^/]+/)+(?P<id>[^/#?]+)'
_MEDIASELECTOR_URLS = [
'https://open.live.bbc.co.uk/mediaselector/6/select/version/2.0/mediaset/iptv-all/vpid/%s/format/xml/',
# Provides HQ HLS streams but fails with geolocation in some cases when it's
# even not geo restricted at all
'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/iptv-all/vpid/%s',
'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/legacy-iptv-all/vpid/%s',
# Provides more formats, namely direct mp4 links, but fails on some videos with
# notukerror for non UK (?) users (e.g.
# http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
@ -977,20 +978,12 @@ class BBCIE(BBCCoUkIE):
if entries:
return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
# http://www.bbc.co.uk/learningenglish/chinese/features/lingohack/ep-181227
group_id = self._search_regex(
r'<div[^>]+\bclass=["\']video["\'][^>]+\bdata-pid=["\'](%s)' % self._ID_REGEX,
webpage, 'group id', default=None)
if playlist_id:
return self.url_result(
'https://www.bbc.co.uk/programmes/%s' % group_id,
ie=BBCCoUkIE.ie_key())
# single video story (e.g. http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
programme_id = self._search_regex(
[r'data-(?:video-player|media)-vpid="(%s)"' % self._ID_REGEX,
r'<param[^>]+name="externalIdentifier"[^>]+value="(%s)"' % self._ID_REGEX,
r'videoId\s*:\s*["\'](%s)["\']' % self._ID_REGEX],
r'videoId\s*:\s*["\'](%s)["\']' % self._ID_REGEX,
r'"vpid":"(%s)"' % self._ID_REGEX],
webpage, 'vpid', default=None)
if programme_id:
@ -1014,6 +1007,15 @@ class BBCIE(BBCCoUkIE):
'subtitles': subtitles,
}
# http://www.bbc.co.uk/learningenglish/chinese/features/lingohack/ep-181227
group_id = self._search_regex(
r'<div[^>]+\bclass=["\']video["\'][^>]+\bdata-pid=["\'](%s)' % self._ID_REGEX,
webpage, 'group id', default=None)
if playlist_id:
return self.url_result(
'https://www.bbc.co.uk/programmes/%s' % group_id,
ie=BBCCoUkIE.ie_key())
# Morph based embed (e.g. http://www.bbc.co.uk/sport/live/olympics/36895975)
# There are several setPayload calls may be present but the video
# seems to be always related to the first one