X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/a4f82408d9a3d921d9c2af7e4d757f44737dc7ea..76d85602f8a22ca3817c3a86f4f0e8969c0b02a9:/youtube_dl/extractor/livestream.py diff --git a/youtube_dl/extractor/livestream.py b/youtube_dl/extractor/livestream.py index c0c2d9b..6d7733e 100644 --- a/youtube_dl/extractor/livestream.py +++ b/youtube_dl/extractor/livestream.py @@ -2,12 +2,15 @@ from __future__ import unicode_literals import re import json +import itertools from .common import InfoExtractor -from ..utils import ( +from ..compat import ( compat_str, compat_urllib_parse_urlparse, compat_urlparse, +) +from ..utils import ( ExtractorError, find_xpath_attr, int_or_none, @@ -18,8 +21,8 @@ from ..utils import ( class LivestreamIE(InfoExtractor): IE_NAME = 'livestream' - _VALID_URL = r'http://new\.livestream\.com/.*?/(?P.*?)(/videos/(?P\d+))?/?$' - _TEST = { + _VALID_URL = r'https?://(?:new\.)?livestream\.com/.*?/(?P.*?)(/videos/(?P[0-9]+)(?:/player)?)?/?(?:$|[?#])' + _TESTS = [{ 'url': 'http://new.livestream.com/CoheedandCambria/WebsterHall/videos/4719370', 'md5': '53274c76ba7754fb0e8d072716f2292b', 'info_dict': { @@ -31,7 +34,27 @@ class LivestreamIE(InfoExtractor): 'view_count': int, 'thumbnail': 're:^http://.*\.jpg$' } - } + }, { + 'url': 'http://new.livestream.com/tedx/cityenglish', + 'info_dict': { + 'title': 'TEDCity2.0 (English)', + 'id': '2245590', + }, + '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', + 'only_matching': True, + }, { + 'url': 'http://livestream.com/bsww/concacafbeachsoccercampeonato2015', + 'only_matching': True, + }] def _parse_smil(self, video_id, smil_url): formats = [] @@ -105,81 +128,124 @@ class LivestreamIE(InfoExtractor): '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): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('id') event_name = mobj.group('event_name') webpage = self._download_webpage(url, video_id or event_name) + og_video = self._og_search_video_url( + webpage, 'player url', fatal=False, default=None) + if og_video is not None: + query_str = compat_urllib_parse_urlparse(og_video).query + query = compat_urlparse.parse_qs(query_str) + if 'play_url' in query: + api_url = query['play_url'][0].replace('.smil', '') + info = json.loads(self._download_webpage( + api_url, video_id, 'Downloading video info')) + return self._extract_video_info(info) + + config_json = self._search_regex( + r'window.config = ({.*?});', webpage, 'window config') + info = json.loads(config_json)['event'] + + def is_relevant(vdata, vid): + result = vdata['type'] == 'video' + if video_id is not None: + result = result and compat_str(vdata['data']['id']) == vid + return result + if video_id is None: # This is an event page: - config_json = self._search_regex( - r'window.config = ({.*?});', webpage, 'window config') - info = json.loads(config_json)['event'] - videos = [self._extract_video_info(video_data['data']) - for video_data in info['feed']['data'] - if video_data['type'] == 'video'] - return self.playlist_result(videos, info['id'], info['full_name']) + return self._extract_event(info) else: - og_video = self._og_search_video_url(webpage, 'player url') - query_str = compat_urllib_parse_urlparse(og_video).query - query = compat_urlparse.parse_qs(query_str) - api_url = query['play_url'][0].replace('.smil', '') - info = json.loads(self._download_webpage( - api_url, video_id, 'Downloading video info')) - return self._extract_video_info(info) + 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: + raise ExtractorError('Cannot find video %s' % video_id) + return videos[0] # The original version of Livestream uses a different system class LivestreamOriginalIE(InfoExtractor): IE_NAME = 'livestream:original' - _VALID_URL = r'''(?x)https?://www\.livestream\.com/ + _VALID_URL = r'''(?x)https?://original\.livestream\.com/ (?P[^/]+)/(?Pvideo|folder) (?:\?.*?Id=|/)(?P.*?)(&|$) ''' - _TEST = { - 'url': 'http://www.livestream.com/dealbook/video?clipId=pla_8aa4a3f1-ba15-46a4-893b-902210e138fb', + _TESTS = [{ + 'url': 'http://original.livestream.com/dealbook/video?clipId=pla_8aa4a3f1-ba15-46a4-893b-902210e138fb', 'info_dict': { 'id': 'pla_8aa4a3f1-ba15-46a4-893b-902210e138fb', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'Spark 1 (BitCoin) with Cameron Winklevoss & Tyler Winklevoss of Winklevoss Capital', }, - 'params': { - # rtmp - 'skip_download': True, + }, { + 'url': 'https://original.livestream.com/newplay/folder?dirId=a07bf706-d0e4-4e75-a747-b021d84f2fd3', + 'info_dict': { + 'id': 'a07bf706-d0e4-4e75-a747-b021d84f2fd3', }, - } + 'playlist_mincount': 4, + }] def _extract_video(self, user, video_id): api_url = 'http://x{0}x.api.channel.livestream.com/2.0/clipdetails?extendedInfo=true&id={1}'.format(user, video_id) info = self._download_xml(api_url, video_id) + # this url is used on mobile devices + stream_url = 'http://x{0}x.api.channel.livestream.com/3.0/getstream.json?id={1}'.format(user, video_id) + stream_info = self._download_json(stream_url, video_id) item = info.find('channel').find('item') ns = {'media': 'http://search.yahoo.com/mrss'} thumbnail_url = item.find(xpath_with_ns('media:thumbnail', ns)).attrib['url'] - # Remove the extension and number from the path (like 1.jpg) - path = self._search_regex(r'(user-files/.+)_.*?\.jpg$', thumbnail_url, 'path') return { 'id': video_id, 'title': item.find('title').text, - 'url': 'rtmp://extondemand.livestream.com/ondemand', - 'play_path': 'mp4:trans/dv15/mogulus-{0}.mp4'.format(path), - 'ext': 'flv', + 'url': stream_info['progressiveUrl'], 'thumbnail': thumbnail_url, } def _extract_folder(self, url, folder_id): webpage = self._download_webpage(url, folder_id) - urls = orderedSet(re.findall(r'\s*