1 from __future__
import unicode_literals
7 from .common
import InfoExtractor
10 compat_urllib_parse_urlparse
,
22 class LivestreamIE(InfoExtractor
):
23 IE_NAME
= 'livestream'
24 _VALID_URL
= r
'https?://(?:new\.)?livestream\.com/.*?/(?P<event_name>.*?)(/videos/(?P<id>[0-9]+)(?:/player)?)?/?(?:$|[?#])'
26 'url': 'http://new.livestream.com/CoheedandCambria/WebsterHall/videos/4719370',
27 'md5': '53274c76ba7754fb0e8d072716f2292b',
31 'title': 'Live from Webster Hall NYC',
32 'upload_date': '20121012',
35 'thumbnail': 're:^http://.*\.jpg$'
38 'url': 'http://new.livestream.com/tedx/cityenglish',
40 'title': 'TEDCity2.0 (English)',
43 'playlist_mincount': 4,
45 'url': 'http://new.livestream.com/chess24/tatasteelchess',
47 'title': 'Tata Steel Chess',
50 'playlist_mincount': 60,
52 'url': 'https://new.livestream.com/accounts/362/events/3557232/videos/67864563/player?autoPlay=false&height=360&mute=false&width=640',
53 'only_matching': True,
55 'url': 'http://livestream.com/bsww/concacafbeachsoccercampeonato2015',
56 'only_matching': True,
59 def _parse_smil(self
, video_id
, smil_url
):
62 './/{http://www.w3.org/2001/SMIL20/Language}body/'
63 '{http://www.w3.org/2001/SMIL20/Language}switch')
64 smil_doc
= self
._download
_xml
(
66 note
='Downloading SMIL information',
67 errnote
='Unable to download SMIL information',
69 if smil_doc
is False: # Download failed
71 title_node
= find_xpath_attr(
72 smil_doc
, './/{http://www.w3.org/2001/SMIL20/Language}meta',
74 if title_node
is None:
75 self
.report_warning('Cannot find SMIL id')
76 switch_node
= smil_doc
.find(_SWITCH_XPATH
)
78 title_id
= title_node
.attrib
['content']
79 switch_node
= find_xpath_attr(
80 smil_doc
, _SWITCH_XPATH
, 'id', title_id
)
81 if switch_node
is None:
82 raise ExtractorError('Cannot find switch node')
83 video_nodes
= switch_node
.findall(
84 '{http://www.w3.org/2001/SMIL20/Language}video')
86 for vn
in video_nodes
:
87 tbr
= int_or_none(vn
.attrib
.get('system-bitrate'))
89 'http://livestream-f.akamaihd.net/%s?v=3.0.3&fp=WIN%%2014,0,0,145' %
91 if 'clipBegin' in vn
.attrib
:
92 furl
+= '&ssek=' + vn
.attrib
['clipBegin']
95 'format_id': 'smil_%d' % tbr
,
102 def _extract_video_info(self
, video_data
):
103 video_id
= compat_str(video_data
['id'])
106 ('sd', 'progressive_url'),
107 ('hd', 'progressive_url_hd'),
110 'format_id': format_id
,
111 'url': video_data
[key
],
113 } for i
, (format_id
, key
) in enumerate(FORMAT_KEYS
)
114 if video_data
.get(key
)]
116 smil_url
= video_data
.get('smil_url')
118 formats
.extend(self
._parse
_smil
(video_id
, smil_url
))
119 self
._sort
_formats
(formats
)
124 'title': video_data
['caption'],
125 'thumbnail': video_data
.get('thumbnail_url'),
126 'upload_date': video_data
['updated_at'].replace('-', '')[:8],
127 'like_count': video_data
.get('likes', {}).get('total'),
128 'view_count': video_data
.get('views'),
131 def _extract_event(self
, info
):
132 event_id
= compat_str(info
['id'])
133 account
= compat_str(info
['owner_account_id'])
135 'https://new.livestream.com/api/accounts/{account}/events/{event}/'
136 'feed.json'.format(account
=account
, event
=event_id
))
138 def _extract_videos():
140 for i
in itertools
.count(1):
141 if last_video
is None:
144 info_url
= '{root}?&id={id}&newer=-1&type=video'.format(
145 root
=root_url
, id=last_video
)
146 videos_info
= self
._download
_json
(info_url
, event_id
, 'Downloading page {0}'.format(i
))['data']
147 videos_info
= [v
['data'] for v
in videos_info
if v
['type'] == 'video']
150 for v
in videos_info
:
151 yield self
._extract
_video
_info
(v
)
152 last_video
= videos_info
[-1]['id']
153 return self
.playlist_result(_extract_videos(), event_id
, info
['full_name'])
155 def _real_extract(self
, url
):
156 mobj
= re
.match(self
._VALID
_URL
, url
)
157 video_id
= mobj
.group('id')
158 event_name
= mobj
.group('event_name')
159 webpage
= self
._download
_webpage
(url
, video_id
or event_name
)
161 og_video
= self
._og
_search
_video
_url
(
162 webpage
, 'player url', fatal
=False, default
=None)
163 if og_video
is not None:
164 query_str
= compat_urllib_parse_urlparse(og_video
).query
165 query
= compat_urlparse
.parse_qs(query_str
)
166 if 'play_url' in query
:
167 api_url
= query
['play_url'][0].replace('.smil', '')
168 info
= json
.loads(self
._download
_webpage
(
169 api_url
, video_id
, 'Downloading video info'))
170 return self
._extract
_video
_info
(info
)
172 config_json
= self
._search
_regex
(
173 r
'window.config = ({.*?});', webpage
, 'window config')
174 info
= json
.loads(config_json
)['event']
176 def is_relevant(vdata
, vid
):
177 result
= vdata
['type'] == 'video'
178 if video_id
is not None:
179 result
= result
and compat_str(vdata
['data']['id']) == vid
183 # This is an event page:
184 return self
._extract
_event
(info
)
186 videos
= [self
._extract
_video
_info
(video_data
['data'])
187 for video_data
in info
['feed']['data']
188 if is_relevant(video_data
, video_id
)]
190 raise ExtractorError('Cannot find video %s' % video_id
)
194 # The original version of Livestream uses a different system
195 class LivestreamOriginalIE(InfoExtractor
):
196 IE_NAME
= 'livestream:original'
197 _VALID_URL
= r
'''(?x)https?://original\.livestream\.com/
198 (?P<user>[^/]+)/(?P<type>video|folder)
199 (?:\?.*?Id=|/)(?P<id>.*?)(&|$)
202 'url': 'http://original.livestream.com/dealbook/video?clipId=pla_8aa4a3f1-ba15-46a4-893b-902210e138fb',
204 'id': 'pla_8aa4a3f1-ba15-46a4-893b-902210e138fb',
206 'title': 'Spark 1 (BitCoin) with Cameron Winklevoss & Tyler Winklevoss of Winklevoss Capital',
209 'url': 'https://original.livestream.com/newplay/folder?dirId=a07bf706-d0e4-4e75-a747-b021d84f2fd3',
211 'id': 'a07bf706-d0e4-4e75-a747-b021d84f2fd3',
213 'playlist_mincount': 4,
216 def _extract_video(self
, user
, video_id
):
217 api_url
= 'http://x{0}x.api.channel.livestream.com/2.0/clipdetails?extendedInfo=true&id={1}'.format(user
, video_id
)
219 info
= self
._download
_xml
(api_url
, video_id
)
220 # this url is used on mobile devices
221 stream_url
= 'http://x{0}x.api.channel.livestream.com/3.0/getstream.json?id={1}'.format(user
, video_id
)
222 stream_info
= self
._download
_json
(stream_url
, video_id
)
223 item
= info
.find('channel').find('item')
224 ns
= {'media': 'http://search.yahoo.com/mrss'}
225 thumbnail_url
= item
.find(xpath_with_ns('media:thumbnail', ns
)).attrib
['url']
229 'title': item
.find('title').text
,
230 'url': stream_info
['progressiveUrl'],
231 'thumbnail': thumbnail_url
,
234 def _extract_folder(self
, url
, folder_id
):
235 webpage
= self
._download
_webpage
(url
, folder_id
)
236 paths
= orderedSet(re
.findall(
238 <li\s+class="folder">\s*<a\s+href="|
239 <a\s+href="(?=https?://livestre\.am/)
240 )([^"]+)"''', webpage
))
247 'url': compat_urlparse
.urljoin(url
, p
),
251 def _real_extract(self
, url
):
252 mobj
= re
.match(self
._VALID
_URL
, url
)
253 id = mobj
.group('id')
254 user
= mobj
.group('user')
255 url_type
= mobj
.group('type')
256 if url_type
== 'folder':
257 return self
._extract
_folder
(url
, id)
259 return self
._extract
_video
(user
, id)
262 # The server doesn't support HEAD request, the generic extractor can't detect
264 class LivestreamShortenerIE(InfoExtractor
):
265 IE_NAME
= 'livestream:shortener'
266 IE_DESC
= False # Do not list
267 _VALID_URL
= r
'https?://livestre\.am/(?P<id>.+)'
269 def _real_extract(self
, url
):
270 mobj
= re
.match(self
._VALID
_URL
, url
)
271 id = mobj
.group('id')
272 webpage
= self
._download
_webpage
(url
, id)
276 'url': self
._og
_search
_url
(webpage
),