[livestream] Extract all videos in events (fixes #5198)

The webpage only contains the most recent ones, but if you scroll down more will appear.
This commit is contained in:
Jaime Marquínez Ferrándiz 2015-03-14 12:06:01 +01:00
parent 9202b1b787
commit 082b1155a3

View File

@ -2,6 +2,7 @@
import re import re
import json import json
import itertools
from .common import InfoExtractor from .common import InfoExtractor
from ..compat import ( from ..compat import (
@ -40,6 +41,13 @@ class LivestreamIE(InfoExtractor):
'id': '2245590', 'id': '2245590',
}, },
'playlist_mincount': 4, 'playlist_mincount': 4,
}, {
'url': 'http://new.livestream.com/chess24/tatasteelchess',
'info_dict': {
'title': 'Tata Steel Chess',
'id': '3705884',
},
'playlist_mincount': 60,
}, { }, {
'url': 'https://new.livestream.com/accounts/362/events/3557232/videos/67864563/player?autoPlay=false&height=360&mute=false&width=640', 'url': 'https://new.livestream.com/accounts/362/events/3557232/videos/67864563/player?autoPlay=false&height=360&mute=false&width=640',
'only_matching': True, 'only_matching': True,
@ -117,6 +125,30 @@ def _extract_video_info(self, video_data):
'view_count': video_data.get('views'), 'view_count': video_data.get('views'),
} }
def _extract_event(self, info):
event_id = compat_str(info['id'])
account = compat_str(info['owner_account_id'])
root_url = (
'https://new.livestream.com/api/accounts/{account}/events/{event}/'
'feed.json'.format(account=account, event=event_id))
def _extract_videos():
last_video = None
for i in itertools.count(1):
if last_video is None:
info_url = root_url
else:
info_url = '{root}?&id={id}&newer=-1&type=video'.format(
root=root_url, id=last_video)
videos_info = self._download_json(info_url, event_id, 'Downloading page {0}'.format(i))['data']
videos_info = [v['data'] for v in videos_info if v['type'] == 'video']
if not videos_info:
break
for v in videos_info:
yield self._extract_video_info(v)
last_video = videos_info[-1]['id']
return self.playlist_result(_extract_videos(), event_id, info['full_name'])
def _real_extract(self, url): def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url) mobj = re.match(self._VALID_URL, url)
video_id = mobj.group('id') video_id = mobj.group('id')
@ -144,14 +176,13 @@ def is_relevant(vdata, vid):
result = result and compat_str(vdata['data']['id']) == vid result = result and compat_str(vdata['data']['id']) == vid
return result return result
videos = [self._extract_video_info(video_data['data'])
for video_data in info['feed']['data']
if is_relevant(video_data, video_id)]
if video_id is None: if video_id is None:
# This is an event page: # This is an event page:
return self.playlist_result( return self._extract_event(info)
videos, '%s' % info['id'], info['full_name'])
else: else:
videos = [self._extract_video_info(video_data['data'])
for video_data in info['feed']['data']
if is_relevant(video_data, video_id)]
if not videos: if not videos:
raise ExtractorError('Cannot find video %s' % video_id) raise ExtractorError('Cannot find video %s' % video_id)
return videos[0] return videos[0]