2 from __future__ import unicode_literals
7 from .common import InfoExtractor
25 _ID_RE = r'[0-9a-f]{32,34}'
28 class MediasiteIE(InfoExtractor):
29 _VALID_URL = r'(?xi)https?://[^/]+/Mediasite/(?:Play|Showcase/(?:default|livebroadcast)/Presentation)/(?P<id>%s)(?P<query>\?[^#]+|)' % _ID_RE
32 'url': 'https://hitsmediaweb.h-its.org/mediasite/Play/2db6c271681e4f199af3c60d1f82869b1d',
34 'id': '2db6c271681e4f199af3c60d1f82869b1d',
36 'title': 'Lecture: Tuesday, September 20, 2016 - Sir Andrew Wiles',
37 'description': 'Sir Andrew Wiles: “Equations in arithmetic”\\n\\nI will describe some of the interactions between modern number theory and the problem of solving equations in rational numbers or integers\\u0027.',
38 'timestamp': 1474268400.0,
39 'upload_date': '20160919',
43 'url': 'http://mediasite.uib.no/Mediasite/Play/90bb363295d945d6b548c867d01181361d?catalog=a452b7df-9ae1-46b7-a3ba-aceeb285f3eb',
45 'id': '90bb363295d945d6b548c867d01181361d',
47 'upload_date': '20150429',
48 'title': '5) IT-forum 2015-Dag 1 - Dungbeetle - How and why Rain created a tiny bug tracker for Unity',
49 'timestamp': 1430311380.0,
53 'url': 'https://collegerama.tudelft.nl/Mediasite/Play/585a43626e544bdd97aeb71a0ec907a01d',
54 'md5': '481fda1c11f67588c0d9d8fbdced4e39',
56 'id': '585a43626e544bdd97aeb71a0ec907a01d',
58 'title': 'Een nieuwe wereld: waarden, bewustzijn en techniek van de mensheid 2.0.',
60 'thumbnail': r're:^https?://.*\.jpg(?:\?.*)?$',
62 'timestamp': 1413309600,
63 'upload_date': '20141014',
67 'url': 'https://collegerama.tudelft.nl/Mediasite/Play/86a9ea9f53e149079fbdb4202b521ed21d?catalog=fd32fd35-6c99-466c-89d4-cd3c431bc8a4',
68 'md5': 'ef1fdded95bdf19b12c5999949419c92',
70 'id': '86a9ea9f53e149079fbdb4202b521ed21d',
72 'title': '64ste Vakantiecursus: Afvalwater',
73 'description': 'md5:7fd774865cc69d972f542b157c328305',
74 'thumbnail': r're:^https?://.*\.jpg(?:\?.*?)?$',
76 'timestamp': 1326446400,
77 'upload_date': '20120113',
81 'url': 'http://digitalops.sandia.gov/Mediasite/Play/24aace4429fc450fb5b38cdbf424a66e1d',
82 'md5': '9422edc9b9a60151727e4b6d8bef393d',
84 'id': '24aace4429fc450fb5b38cdbf424a66e1d',
86 'title': 'Xyce Software Training - Section 1',
87 'description': r're:(?s)SAND Number: SAND 2013-7800.{200,}',
88 'upload_date': '20120409',
89 'timestamp': 1333983600,
94 'url': 'https://collegerama.tudelft.nl/Mediasite/Showcase/livebroadcast/Presentation/ada7020854f743c49fbb45c9ec7dbb351d',
95 'only_matching': True,
98 'url': 'https://mediasite.ntnu.no/Mediasite/Showcase/default/Presentation/7d8b913259334b688986e970fae6fcb31d',
99 'only_matching': True,
103 # look in Mediasite.Core.js (Mediasite.ContentStreamType[*])
105 0: 'video1', # the main video
108 4: 'video2', # screencast?
113 def _extract_urls(webpage):
115 unescapeHTML(mobj.group('url'))
116 for mobj in re.finditer(
117 r'(?xi)<iframe\b[^>]+\bsrc=(["\'])(?P<url>(?:(?:https?:)?//[^/]+)?/Mediasite/Play/%s(?:\?.*?)?)\1' % _ID_RE,
120 def _real_extract(self, url):
121 url, data = unsmuggle_url(url, {})
122 mobj = re.match(self._VALID_URL, url)
123 resource_id = mobj.group('id')
124 query = mobj.group('query')
126 webpage, urlh = self._download_webpage_handle(url, resource_id) # XXX: add UrlReferrer?
127 redirect_url = compat_str(urlh.geturl())
129 # XXX: might have also extracted UrlReferrer and QueryString from the html
130 service_path = compat_urlparse.urljoin(redirect_url, self._html_search_regex(
131 r'<div[^>]+\bid=["\']ServicePath[^>]+>(.+?)</div>', webpage, resource_id,
132 default='/Mediasite/PlayerService/PlayerService.svc/json'))
134 player_options = self._download_json(
135 '%s/GetPlayerOptions' % service_path, resource_id,
137 'Content-type': 'application/json; charset=utf-8',
138 'X-Requested-With': 'XMLHttpRequest',
141 'getPlayerOptionsRequest': {
142 'ResourceId': resource_id,
143 'QueryString': query,
144 'UrlReferrer': data.get('UrlReferrer', ''),
145 'UseScreenReader': False,
147 }).encode('utf-8'))['d']
149 presentation = player_options['Presentation']
150 title = presentation['Title']
152 if presentation is None:
153 raise ExtractorError(
154 'Mediasite says: %s' % player_options['PlayerPresentationStatusMessage'],
159 for snum, Stream in enumerate(presentation['Streams']):
160 stream_type = Stream.get('StreamType')
161 if stream_type is None:
164 video_urls = Stream.get('VideoUrls')
165 if not isinstance(video_urls, list):
168 stream_id = self._STREAM_TYPES.get(
169 stream_type, 'type%u' % stream_type)
172 for unum, VideoUrl in enumerate(video_urls):
173 video_url = url_or_none(VideoUrl.get('Location'))
176 # XXX: if Stream.get('CanChangeScheme', False), switch scheme to HTTP/HTTPS
178 media_type = VideoUrl.get('MediaType')
179 if media_type == 'SS':
180 stream_formats.extend(self._extract_ism_formats(
181 video_url, resource_id,
182 ism_id='%s-%u.%u' % (stream_id, snum, unum),
184 elif media_type == 'Dash':
185 stream_formats.extend(self._extract_mpd_formats(
186 video_url, resource_id,
187 mpd_id='%s-%u.%u' % (stream_id, snum, unum),
190 stream_formats.append({
191 'format_id': '%s-%u.%u' % (stream_id, snum, unum),
193 'ext': mimetype2ext(VideoUrl.get('MimeType')),
196 # TODO: if Stream['HasSlideContent']:
197 # synthesise an MJPEG video stream '%s-%u.slides' % (stream_type, snum)
198 # from Stream['Slides']
199 # this will require writing a custom downloader...
201 # disprefer 'secondary' streams
203 for fmt in stream_formats:
204 fmt['preference'] = -1
206 thumbnail_url = Stream.get('ThumbnailUrl')
209 'id': '%s-%u' % (stream_id, snum),
210 'url': urljoin(redirect_url, thumbnail_url),
211 'preference': -1 if stream_type != 0 else 0,
213 formats.extend(stream_formats)
215 self._sort_formats(formats)
217 # XXX: Presentation['Presenters']
218 # XXX: Presentation['Transcript']
223 'description': presentation.get('Description'),
224 'duration': float_or_none(presentation.get('Duration'), 1000),
225 'timestamp': float_or_none(presentation.get('UnixTime'), 1000),
227 'thumbnails': thumbnails,
231 class MediasiteCatalogIE(InfoExtractor):
232 _VALID_URL = r'''(?xi)
233 (?P<url>https?://[^/]+/Mediasite)
237 /(?P<current_folder_id>{0})
238 /(?P<root_dynamic_folder_id>{0})
242 'url': 'http://events7.mediasite.com/Mediasite/Catalog/Full/631f9e48530d454381549f955d08c75e21',
244 'id': '631f9e48530d454381549f955d08c75e21',
245 'title': 'WCET Summit: Adaptive Learning in Higher Ed: Improving Outcomes Dynamically',
248 'expected_warnings': ['is not a supported codec'],
250 # with CurrentFolderId and RootDynamicFolderId
251 'url': 'https://medaudio.medicine.iu.edu/Mediasite/Catalog/Full/9518c4a6c5cf4993b21cbd53e828a92521/97a9db45f7ab47428c77cd2ed74bb98f14/9518c4a6c5cf4993b21cbd53e828a92521',
253 'id': '9518c4a6c5cf4993b21cbd53e828a92521',
254 'title': 'IUSM Family and Friends Sessions',
258 'url': 'http://uipsyc.mediasite.com/mediasite/Catalog/Full/d5d79287c75243c58c50fef50174ec1b21',
259 'only_matching': True,
261 # no AntiForgeryToken
262 'url': 'https://live.libraries.psu.edu/Mediasite/Catalog/Full/8376d4b24dd1457ea3bfe4cf9163feda21',
263 'only_matching': True,
265 'url': 'https://medaudio.medicine.iu.edu/Mediasite/Catalog/Full/9518c4a6c5cf4993b21cbd53e828a92521/97a9db45f7ab47428c77cd2ed74bb98f14/9518c4a6c5cf4993b21cbd53e828a92521',
266 'only_matching': True,
269 def _real_extract(self, url):
270 mobj = re.match(self._VALID_URL, url)
271 mediasite_url = mobj.group('url')
272 catalog_id = mobj.group('catalog_id')
273 current_folder_id = mobj.group('current_folder_id') or catalog_id
274 root_dynamic_folder_id = mobj.group('root_dynamic_folder_id')
276 webpage = self._download_webpage(url, catalog_id)
278 # AntiForgeryToken is optional (e.g. [1])
279 # 1. https://live.libraries.psu.edu/Mediasite/Catalog/Full/8376d4b24dd1457ea3bfe4cf9163feda21
280 anti_forgery_token = self._search_regex(
281 r'AntiForgeryToken\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1',
282 webpage, 'anti forgery token', default=None, group='value')
283 if anti_forgery_token:
284 anti_forgery_header = self._search_regex(
285 r'AntiForgeryHeaderName\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1',
286 webpage, 'anti forgery header name',
287 default='X-SOFO-AntiForgeryHeader', group='value')
293 'CatalogId': catalog_id,
294 'CurrentFolderId': current_folder_id,
295 'RootDynamicFolderId': root_dynamic_folder_id,
296 'ItemsPerPage': 1000,
298 'PermissionMask': 'Execute',
299 'CatalogSearchType': 'SearchInFolder',
301 'SortDirection': 'Descending',
304 'StatusFilterList': None,
310 'Content-Type': 'application/json; charset=UTF-8',
312 'X-Requested-With': 'XMLHttpRequest',
314 if anti_forgery_token:
315 headers[anti_forgery_header] = anti_forgery_token
317 catalog = self._download_json(
318 '%s/Catalog/Data/GetPresentationsForFolder' % mediasite_url,
319 catalog_id, data=json.dumps(data).encode(), headers=headers)
322 for video in catalog['PresentationDetailsList']:
323 if not isinstance(video, dict):
325 video_id = str_or_none(video.get('Id'))
328 entries.append(self.url_result(
329 '%s/Play/%s' % (mediasite_url, video_id),
330 ie=MediasiteIE.ie_key(), video_id=video_id))
333 catalog, lambda x: x['CurrentFolder']['Name'], compat_str)
335 return self.playlist_result(entries, catalog_id, title,)