1 from __future__
import unicode_literals
6 from .common
import InfoExtractor
24 class LivestreamIE(InfoExtractor
):
25 IE_NAME
= 'livestream'
26 _VALID_URL
= r
'https?://(?:new\.)?livestream\.com/(?:accounts/(?P<account_id>\d+)|(?P<account_name>[^/]+))/(?:events/(?P<event_id>\d+)|(?P<event_name>[^/]+))(?:/videos/(?P<id>\d+))?'
28 'url': 'http://new.livestream.com/CoheedandCambria/WebsterHall/videos/4719370',
29 'md5': '53274c76ba7754fb0e8d072716f2292b',
33 'title': 'Live from Webster Hall NYC',
34 'timestamp': 1350008072,
35 'upload_date': '20121012',
39 'thumbnail': 're:^http://.*\.jpg$'
42 'url': 'http://new.livestream.com/tedx/cityenglish',
44 'title': 'TEDCity2.0 (English)',
47 'playlist_mincount': 4,
49 'url': 'http://new.livestream.com/chess24/tatasteelchess',
51 'title': 'Tata Steel Chess',
54 'playlist_mincount': 60,
56 'url': 'https://new.livestream.com/accounts/362/events/3557232/videos/67864563/player?autoPlay=false&height=360&mute=false&width=640',
57 'only_matching': True,
59 'url': 'http://livestream.com/bsww/concacafbeachsoccercampeonato2015',
60 'only_matching': True,
62 _API_URL_TEMPLATE
= 'http://livestream.com/api/accounts/%s/events/%s'
64 def _parse_smil_formats(self
, smil
, smil_url
, video_id
, namespace
=None, f4m_params
=None, transform_rtmp_url
=None):
65 base_ele
= find_xpath_attr(
66 smil
, self
._xpath
_ns
('.//meta', namespace
), 'name', 'httpBase')
67 base
= base_ele
.get('content') if base_ele
else 'http://livestreamvod-f.akamaihd.net/'
70 video_nodes
= smil
.findall(self
._xpath
_ns
('.//video', namespace
))
72 for vn
in video_nodes
:
73 tbr
= int_or_none(vn
.attrib
.get('system-bitrate'), 1000)
75 '%s%s?v=3.0.3&fp=WIN%%2014,0,0,145' % (base
, vn
.attrib
['src']))
76 if 'clipBegin' in vn
.attrib
:
77 furl
+= '&ssek=' + vn
.attrib
['clipBegin']
80 'format_id': 'smil_%d' % tbr
,
87 def _extract_video_info(self
, video_data
):
88 video_id
= compat_str(video_data
['id'])
91 ('sd', 'progressive_url'),
92 ('hd', 'progressive_url_hd'),
96 for format_id
, key
in FORMAT_KEYS
:
97 video_url
= video_data
.get(key
)
99 ext
= determine_ext(video_url
)
102 bitrate
= int_or_none(self
._search
_regex
(
103 r
'(\d+)\.%s' % ext
, video_url
, 'bitrate', default
=None))
106 'format_id': format_id
,
111 smil_url
= video_data
.get('smil_url')
113 formats
.extend(self
._extract
_smil
_formats
(smil_url
, video_id
))
115 m3u8_url
= video_data
.get('m3u8_url')
117 formats
.extend(self
._extract
_m
3u8_formats
(
118 m3u8_url
, video_id
, 'mp4', 'm3u8_native', m3u8_id
='hls', fatal
=False))
120 f4m_url
= video_data
.get('f4m_url')
122 formats
.extend(self
._extract
_f
4m
_formats
(
123 f4m_url
, video_id
, f4m_id
='hds', fatal
=False))
124 self
._sort
_formats
(formats
)
127 'author_id': comment
.get('author_id'),
128 'author': comment
.get('author', {}).get('full_name'),
129 'id': comment
.get('id'),
130 'text': comment
['text'],
131 'timestamp': parse_iso8601(comment
.get('created_at')),
132 } for comment
in video_data
.get('comments', {}).get('data', [])]
137 'title': video_data
['caption'],
138 'description': video_data
.get('description'),
139 'thumbnail': video_data
.get('thumbnail_url'),
140 'duration': float_or_none(video_data
.get('duration'), 1000),
141 'timestamp': parse_iso8601(video_data
.get('publish_at')),
142 'like_count': video_data
.get('likes', {}).get('total'),
143 'comment_count': video_data
.get('comments', {}).get('total'),
144 'view_count': video_data
.get('views'),
145 'comments': comments
,
148 def _extract_stream_info(self
, stream_info
):
149 broadcast_id
= stream_info
['broadcast_id']
150 is_live
= stream_info
.get('is_live')
153 smil_url
= stream_info
.get('play_url')
155 formats
.extend(self
._extract
_smil
_formats
(smil_url
, broadcast_id
))
157 entry_protocol
= 'm3u8' if is_live
else 'm3u8_native'
158 m3u8_url
= stream_info
.get('m3u8_url')
160 formats
.extend(self
._extract
_m
3u8_formats
(
161 m3u8_url
, broadcast_id
, 'mp4', entry_protocol
, m3u8_id
='hls', fatal
=False))
163 rtsp_url
= stream_info
.get('rtsp_url')
169 self
._sort
_formats
(formats
)
174 'title': self
._live
_title
(stream_info
['stream_title']) if is_live
else stream_info
['stream_title'],
175 'thumbnail': stream_info
.get('thumbnail_url'),
179 def _extract_event(self
, event_data
):
180 event_id
= compat_str(event_data
['id'])
181 account_id
= compat_str(event_data
['owner_account_id'])
182 feed_root_url
= self
._API
_URL
_TEMPLATE
% (account_id
, event_id
) + '/feed.json'
184 stream_info
= event_data
.get('stream_info')
186 return self
._extract
_stream
_info
(stream_info
)
190 for i
in itertools
.count(1):
191 if last_video
is None:
192 info_url
= feed_root_url
194 info_url
= '{root}?&id={id}&newer=-1&type=video'.format(
195 root
=feed_root_url
, id=last_video
)
196 videos_info
= self
._download
_json
(
197 info_url
, event_id
, 'Downloading page {0}'.format(i
))['data']
198 videos_info
= [v
['data'] for v
in videos_info
if v
['type'] == 'video']
201 for v
in videos_info
:
202 entries
.append(self
.url_result(
203 'http://livestream.com/accounts/%s/events/%s/videos/%s' % (account_id
, event_id
, v
['id']),
204 'Livestream', v
['id'], v
['caption']))
205 last_video
= videos_info
[-1]['id']
206 return self
.playlist_result(entries
, event_id
, event_data
['full_name'])
208 def _real_extract(self
, url
):
209 mobj
= re
.match(self
._VALID
_URL
, url
)
210 video_id
= mobj
.group('id')
211 event
= mobj
.group('event_id') or mobj
.group('event_name')
212 account
= mobj
.group('account_id') or mobj
.group('account_name')
213 api_url
= self
._API
_URL
_TEMPLATE
% (account
, event
)
215 video_data
= self
._download
_json
(
216 api_url
+ '/videos/%s' % video_id
, video_id
)
217 return self
._extract
_video
_info
(video_data
)
219 event_data
= self
._download
_json
(api_url
, video_id
)
220 return self
._extract
_event
(event_data
)
223 # The original version of Livestream uses a different system
224 class LivestreamOriginalIE(InfoExtractor
):
225 IE_NAME
= 'livestream:original'
226 _VALID_URL
= r
'''(?x)https?://original\.livestream\.com/
227 (?P<user>[^/\?#]+)(?:/(?P<type>video|folder)
228 (?:(?:\?.*?Id=|/)(?P<id>.*?)(&|$))?)?
231 'url': 'http://original.livestream.com/dealbook/video?clipId=pla_8aa4a3f1-ba15-46a4-893b-902210e138fb',
233 'id': 'pla_8aa4a3f1-ba15-46a4-893b-902210e138fb',
235 'title': 'Spark 1 (BitCoin) with Cameron Winklevoss & Tyler Winklevoss of Winklevoss Capital',
240 'url': 'https://original.livestream.com/newplay/folder?dirId=a07bf706-d0e4-4e75-a747-b021d84f2fd3',
242 'id': 'a07bf706-d0e4-4e75-a747-b021d84f2fd3',
244 'playlist_mincount': 4,
247 'url': 'http://original.livestream.com/znsbahamas',
248 'only_matching': True,
251 def _extract_video_info(self
, user
, video_id
):
252 api_url
= 'http://x%sx.api.channel.livestream.com/2.0/clipdetails?extendedInfo=true&id=%s' % (user
, video_id
)
253 info
= self
._download
_xml
(api_url
, video_id
)
255 item
= info
.find('channel').find('item')
256 title
= xpath_text(item
, 'title')
257 media_ns
= {'media': 'http://search.yahoo.com/mrss'}
258 thumbnail_url
= xpath_attr(
259 item
, xpath_with_ns('media:thumbnail', media_ns
), 'url')
260 duration
= float_or_none(xpath_attr(
261 item
, xpath_with_ns('media:content', media_ns
), 'duration'))
262 ls_ns
= {'ls': 'http://api.channel.livestream.com/2.0'}
263 view_count
= int_or_none(xpath_text(
264 item
, xpath_with_ns('ls:viewsCount', ls_ns
)))
269 'thumbnail': thumbnail_url
,
270 'duration': duration
,
271 'view_count': view_count
,
274 def _extract_video_formats(self
, video_data
, video_id
, entry_protocol
):
277 progressive_url
= video_data
.get('progressiveUrl')
280 'url': progressive_url
,
284 m3u8_url
= video_data
.get('httpUrl')
286 formats
.extend(self
._extract
_m
3u8_formats
(
287 m3u8_url
, video_id
, 'mp4', entry_protocol
, m3u8_id
='hls', fatal
=False))
289 rtsp_url
= video_data
.get('rtspUrl')
296 self
._sort
_formats
(formats
)
299 def _extract_folder(self
, url
, folder_id
):
300 webpage
= self
._download
_webpage
(url
, folder_id
)
301 paths
= orderedSet(re
.findall(
303 <li\s+class="folder">\s*<a\s+href="|
304 <a\s+href="(?=https?://livestre\.am/)
305 )([^"]+)"''', webpage
))
309 'url': compat_urlparse
.urljoin(url
, p
),
312 return self
.playlist_result(entries
, folder_id
)
314 def _real_extract(self
, url
):
315 mobj
= re
.match(self
._VALID
_URL
, url
)
316 user
= mobj
.group('user')
317 url_type
= mobj
.group('type')
318 content_id
= mobj
.group('id')
319 if url_type
== 'folder':
320 return self
._extract
_folder
(url
, content_id
)
322 # this url is used on mobile devices
323 stream_url
= 'http://x%sx.api.channel.livestream.com/3.0/getstream.json' % user
326 stream_url
+= '?id=%s' % content_id
327 info
= self
._extract
_video
_info
(user
, content_id
)
330 webpage
= self
._download
_webpage
(url
, content_id
)
332 'title': self
._og
_search
_title
(webpage
),
333 'description': self
._og
_search
_description
(webpage
),
334 'thumbnail': self
._search
_regex
(r
'channelLogo.src\s*=\s*"([^"]+)"', webpage
, 'thumbnail', None),
336 video_data
= self
._download
_json
(stream_url
, content_id
)
337 is_live
= video_data
.get('isLive')
338 entry_protocol
= 'm3u8' if is_live
else 'm3u8_native'
341 'title': self
._live
_title
(info
['title']) if is_live
else info
['title'],
342 'formats': self
._extract
_video
_formats
(video_data
, content_id
, entry_protocol
),
348 # The server doesn't support HEAD request, the generic extractor can't detect
350 class LivestreamShortenerIE(InfoExtractor
):
351 IE_NAME
= 'livestream:shortener'
352 IE_DESC
= False # Do not list
353 _VALID_URL
= r
'https?://livestre\.am/(?P<id>.+)'
355 def _real_extract(self
, url
):
356 mobj
= re
.match(self
._VALID
_URL
, url
)
357 id = mobj
.group('id')
358 webpage
= self
._download
_webpage
(url
, id)
362 'url': self
._og
_search
_url
(webpage
),