4 from .common
import InfoExtractor
23 class LivestreamIE(InfoExtractor
):
24 IE_NAME
= 'livestream'
25 _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+))?'
27 'url': 'http://new.livestream.com/CoheedandCambria/WebsterHall/videos/4719370',
28 'md5': '53274c76ba7754fb0e8d072716f2292b',
32 'title': 'Live from Webster Hall NYC',
33 'timestamp': 1350008072,
34 'upload_date': '20121012',
38 'thumbnail': r
're:^http://.*\.jpg$'
41 'url': 'http://new.livestream.com/tedx/cityenglish',
43 'title': 'TEDCity2.0 (English)',
46 'playlist_mincount': 4,
48 'url': 'http://new.livestream.com/chess24/tatasteelchess',
50 'title': 'Tata Steel Chess',
53 'playlist_mincount': 60,
55 'url': 'https://new.livestream.com/accounts/362/events/3557232/videos/67864563/player?autoPlay=false&height=360&mute=false&width=640',
56 'only_matching': True,
58 'url': 'http://livestream.com/bsww/concacafbeachsoccercampeonato2015',
59 'only_matching': True,
61 _API_URL_TEMPLATE
= 'http://livestream.com/api/accounts/%s/events/%s'
63 def _parse_smil_formats(self
, smil
, smil_url
, video_id
, namespace
=None, f4m_params
=None, transform_rtmp_url
=None):
64 base_ele
= find_xpath_attr(
65 smil
, self
._xpath
_ns
('.//meta', namespace
), 'name', 'httpBase')
66 base
= base_ele
.get('content') if base_ele
is not None else 'http://livestreamvod-f.akamaihd.net/'
69 video_nodes
= smil
.findall(self
._xpath
_ns
('.//video', namespace
))
71 for vn
in video_nodes
:
72 tbr
= int_or_none(vn
.attrib
.get('system-bitrate'), 1000)
74 update_url_query(compat_urlparse
.urljoin(base
, vn
.attrib
['src']), {
76 'fp': 'WIN% 14,0,0,145',
78 if 'clipBegin' in vn
.attrib
:
79 furl
+= '&ssek=' + vn
.attrib
['clipBegin']
82 'format_id': 'smil_%d' % tbr
,
85 'preference': -1000, # Strictly inferior than all other formats?
89 def _extract_video_info(self
, video_data
):
90 video_id
= compat_str(video_data
['id'])
93 ('sd', 'progressive_url'),
94 ('hd', 'progressive_url_hd'),
98 for format_id
, key
in FORMAT_KEYS
:
99 video_url
= video_data
.get(key
)
101 ext
= determine_ext(video_url
)
104 bitrate
= int_or_none(self
._search
_regex
(
105 r
'(\d+)\.%s' % ext
, video_url
, 'bitrate', default
=None))
108 'format_id': format_id
,
113 smil_url
= video_data
.get('smil_url')
115 formats
.extend(self
._extract
_smil
_formats
(smil_url
, video_id
, fatal
=False))
117 m3u8_url
= video_data
.get('m3u8_url')
119 formats
.extend(self
._extract
_m
3u8_formats
(
120 m3u8_url
, video_id
, 'mp4', 'm3u8_native',
121 m3u8_id
='hls', fatal
=False))
123 f4m_url
= video_data
.get('f4m_url')
125 formats
.extend(self
._extract
_f
4m
_formats
(
126 f4m_url
, video_id
, f4m_id
='hds', fatal
=False))
127 self
._sort
_formats
(formats
)
130 'author_id': comment
.get('author_id'),
131 'author': comment
.get('author', {}).get('full_name'),
132 'id': comment
.get('id'),
133 'text': comment
['text'],
134 'timestamp': parse_iso8601(comment
.get('created_at')),
135 } for comment
in video_data
.get('comments', {}).get('data', [])]
140 'title': video_data
['caption'],
141 'description': video_data
.get('description'),
142 'thumbnail': video_data
.get('thumbnail_url'),
143 'duration': float_or_none(video_data
.get('duration'), 1000),
144 'timestamp': parse_iso8601(video_data
.get('publish_at')),
145 'like_count': video_data
.get('likes', {}).get('total'),
146 'comment_count': video_data
.get('comments', {}).get('total'),
147 'view_count': video_data
.get('views'),
148 'comments': comments
,
151 def _extract_stream_info(self
, stream_info
):
152 broadcast_id
= compat_str(stream_info
['broadcast_id'])
153 is_live
= stream_info
.get('is_live')
156 smil_url
= stream_info
.get('play_url')
158 formats
.extend(self
._extract
_smil
_formats
(smil_url
, broadcast_id
))
160 m3u8_url
= stream_info
.get('m3u8_url')
162 formats
.extend(self
._extract
_m
3u8_formats
(
163 m3u8_url
, broadcast_id
, 'mp4', 'm3u8_native',
164 m3u8_id
='hls', fatal
=False))
166 rtsp_url
= stream_info
.get('rtsp_url')
172 self
._sort
_formats
(formats
)
177 'title': stream_info
['stream_title'],
178 'thumbnail': stream_info
.get('thumbnail_url'),
182 def _extract_event(self
, event_data
):
183 event_id
= compat_str(event_data
['id'])
184 account_id
= compat_str(event_data
['owner_account_id'])
185 feed_root_url
= self
._API
_URL
_TEMPLATE
% (account_id
, event_id
) + '/feed.json'
187 stream_info
= event_data
.get('stream_info')
189 return self
._extract
_stream
_info
(stream_info
)
193 for i
in itertools
.count(1):
194 if last_video
is None:
195 info_url
= feed_root_url
197 info_url
= '{root}?&id={id}&newer=-1&type=video'.format(
198 root
=feed_root_url
, id=last_video
)
199 videos_info
= self
._download
_json
(
200 info_url
, event_id
, 'Downloading page {0}'.format(i
))['data']
201 videos_info
= [v
['data'] for v
in videos_info
if v
['type'] == 'video']
204 for v
in videos_info
:
205 v_id
= compat_str(v
['id'])
206 entries
.append(self
.url_result(
207 'http://livestream.com/accounts/%s/events/%s/videos/%s' % (account_id
, event_id
, v_id
),
208 'Livestream', v_id
, v
.get('caption')))
209 last_video
= videos_info
[-1]['id']
210 return self
.playlist_result(entries
, event_id
, event_data
['full_name'])
212 def _real_extract(self
, url
):
213 mobj
= self
._match
_valid
_url
(url
)
214 video_id
= mobj
.group('id')
215 event
= mobj
.group('event_id') or mobj
.group('event_name')
216 account
= mobj
.group('account_id') or mobj
.group('account_name')
217 api_url
= self
._API
_URL
_TEMPLATE
% (account
, event
)
219 video_data
= self
._download
_json
(
220 api_url
+ '/videos/%s' % video_id
, video_id
)
221 return self
._extract
_video
_info
(video_data
)
223 event_data
= self
._download
_json
(api_url
, video_id
)
224 return self
._extract
_event
(event_data
)
227 # The original version of Livestream uses a different system
228 class LivestreamOriginalIE(InfoExtractor
):
229 IE_NAME
= 'livestream:original'
230 _VALID_URL
= r
'''(?x)https?://original\.livestream\.com/
231 (?P<user>[^/\?#]+)(?:/(?P<type>video|folder)
232 (?:(?:\?.*?Id=|/)(?P<id>.*?)(&|$))?)?
235 'url': 'http://original.livestream.com/dealbook/video?clipId=pla_8aa4a3f1-ba15-46a4-893b-902210e138fb',
237 'id': 'pla_8aa4a3f1-ba15-46a4-893b-902210e138fb',
239 'title': 'Spark 1 (BitCoin) with Cameron Winklevoss & Tyler Winklevoss of Winklevoss Capital',
244 'url': 'https://original.livestream.com/newplay/folder?dirId=a07bf706-d0e4-4e75-a747-b021d84f2fd3',
246 'id': 'a07bf706-d0e4-4e75-a747-b021d84f2fd3',
248 'playlist_mincount': 4,
251 'url': 'http://original.livestream.com/znsbahamas',
252 'only_matching': True,
255 def _extract_video_info(self
, user
, video_id
):
256 api_url
= 'http://x%sx.api.channel.livestream.com/2.0/clipdetails?extendedInfo=true&id=%s' % (user
, video_id
)
257 info
= self
._download
_xml
(api_url
, video_id
)
259 item
= info
.find('channel').find('item')
260 title
= xpath_text(item
, 'title')
261 media_ns
= {'media': 'http://search.yahoo.com/mrss'}
262 thumbnail_url
= xpath_attr(
263 item
, xpath_with_ns('media:thumbnail', media_ns
), 'url')
264 duration
= float_or_none(xpath_attr(
265 item
, xpath_with_ns('media:content', media_ns
), 'duration'))
266 ls_ns
= {'ls': 'http://api.channel.livestream.com/2.0'}
267 view_count
= int_or_none(xpath_text(
268 item
, xpath_with_ns('ls:viewsCount', ls_ns
)))
273 'thumbnail': thumbnail_url
,
274 'duration': duration
,
275 'view_count': view_count
,
278 def _extract_video_formats(self
, video_data
, video_id
):
281 progressive_url
= video_data
.get('progressiveUrl')
284 'url': progressive_url
,
288 m3u8_url
= video_data
.get('httpUrl')
290 formats
.extend(self
._extract
_m
3u8_formats
(
291 m3u8_url
, video_id
, 'mp4', 'm3u8_native',
292 m3u8_id
='hls', fatal
=False))
294 rtsp_url
= video_data
.get('rtspUrl')
301 self
._sort
_formats
(formats
)
304 def _extract_folder(self
, url
, folder_id
):
305 webpage
= self
._download
_webpage
(url
, folder_id
)
306 paths
= orderedSet(re
.findall(
308 <li\s+class="folder">\s*<a\s+href="|
309 <a\s+href="(?=https?://livestre\.am/)
310 )([^"]+)"''', webpage
))
314 'url': compat_urlparse
.urljoin(url
, p
),
317 return self
.playlist_result(entries
, folder_id
)
319 def _real_extract(self
, url
):
320 mobj
= self
._match
_valid
_url
(url
)
321 user
= mobj
.group('user')
322 url_type
= mobj
.group('type')
323 content_id
= mobj
.group('id')
324 if url_type
== 'folder':
325 return self
._extract
_folder
(url
, content_id
)
327 # this url is used on mobile devices
328 stream_url
= 'http://x%sx.api.channel.livestream.com/3.0/getstream.json' % user
331 stream_url
+= '?id=%s' % content_id
332 info
= self
._extract
_video
_info
(user
, content_id
)
335 webpage
= self
._download
_webpage
(url
, content_id
)
337 'title': self
._og
_search
_title
(webpage
),
338 'description': self
._og
_search
_description
(webpage
),
339 'thumbnail': self
._search
_regex
(r
'channelLogo\.src\s*=\s*"([^"]+)"', webpage
, 'thumbnail', None),
341 video_data
= self
._download
_json
(stream_url
, content_id
)
342 is_live
= video_data
.get('isLive')
345 'title': info
['title'],
346 'formats': self
._extract
_video
_formats
(video_data
, content_id
),
352 # The server doesn't support HEAD request, the generic extractor can't detect
354 class LivestreamShortenerIE(InfoExtractor
):
355 IE_NAME
= 'livestream:shortener'
356 IE_DESC
= False # Do not list
357 _VALID_URL
= r
'https?://livestre\.am/(?P<id>.+)'
359 def _real_extract(self
, url
):
360 mobj
= self
._match
_valid
_url
(url
)
361 id = mobj
.group('id')
362 webpage
= self
._download
_webpage
(url
, id)
364 return self
.url_result(self
._og
_search
_url
(webpage
))