]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/mediasite.py
debian/control: Mark my GitHub trees as the preferred repos.
[youtubedl] / youtube_dl / extractor / mediasite.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5 import json
6
7 from .common import InfoExtractor
8 from ..compat import (
9 compat_str,
10 compat_urlparse,
11 )
12 from ..utils import (
13 ExtractorError,
14 float_or_none,
15 mimetype2ext,
16 str_or_none,
17 try_get,
18 unescapeHTML,
19 unsmuggle_url,
20 url_or_none,
21 urljoin,
22 )
23
24
25 _ID_RE = r'(?:[0-9a-f]{32,34}|[0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{12,14})'
26
27
28 class MediasiteIE(InfoExtractor):
29 _VALID_URL = r'(?xi)https?://[^/]+/Mediasite/(?:Play|Showcase/(?:default|livebroadcast)/Presentation)/(?P<id>%s)(?P<query>\?[^#]+|)' % _ID_RE
30 _TESTS = [
31 {
32 'url': 'https://hitsmediaweb.h-its.org/mediasite/Play/2db6c271681e4f199af3c60d1f82869b1d',
33 'info_dict': {
34 'id': '2db6c271681e4f199af3c60d1f82869b1d',
35 'ext': 'mp4',
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',
40 },
41 },
42 {
43 'url': 'http://mediasite.uib.no/Mediasite/Play/90bb363295d945d6b548c867d01181361d?catalog=a452b7df-9ae1-46b7-a3ba-aceeb285f3eb',
44 'info_dict': {
45 'id': '90bb363295d945d6b548c867d01181361d',
46 'ext': 'mp4',
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,
50 },
51 },
52 {
53 'url': 'https://collegerama.tudelft.nl/Mediasite/Play/585a43626e544bdd97aeb71a0ec907a01d',
54 'md5': '481fda1c11f67588c0d9d8fbdced4e39',
55 'info_dict': {
56 'id': '585a43626e544bdd97aeb71a0ec907a01d',
57 'ext': 'mp4',
58 'title': 'Een nieuwe wereld: waarden, bewustzijn en techniek van de mensheid 2.0.',
59 'description': '',
60 'thumbnail': r're:^https?://.*\.jpg(?:\?.*)?$',
61 'duration': 7713.088,
62 'timestamp': 1413309600,
63 'upload_date': '20141014',
64 },
65 },
66 {
67 'url': 'https://collegerama.tudelft.nl/Mediasite/Play/86a9ea9f53e149079fbdb4202b521ed21d?catalog=fd32fd35-6c99-466c-89d4-cd3c431bc8a4',
68 'md5': 'ef1fdded95bdf19b12c5999949419c92',
69 'info_dict': {
70 'id': '86a9ea9f53e149079fbdb4202b521ed21d',
71 'ext': 'wmv',
72 'title': '64ste Vakantiecursus: Afvalwater',
73 'description': 'md5:7fd774865cc69d972f542b157c328305',
74 'thumbnail': r're:^https?://.*\.jpg(?:\?.*?)?$',
75 'duration': 10853,
76 'timestamp': 1326446400,
77 'upload_date': '20120113',
78 },
79 },
80 {
81 'url': 'http://digitalops.sandia.gov/Mediasite/Play/24aace4429fc450fb5b38cdbf424a66e1d',
82 'md5': '9422edc9b9a60151727e4b6d8bef393d',
83 'info_dict': {
84 'id': '24aace4429fc450fb5b38cdbf424a66e1d',
85 'ext': 'mp4',
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,
90 'duration': 7794,
91 }
92 },
93 {
94 'url': 'https://collegerama.tudelft.nl/Mediasite/Showcase/livebroadcast/Presentation/ada7020854f743c49fbb45c9ec7dbb351d',
95 'only_matching': True,
96 },
97 {
98 'url': 'https://mediasite.ntnu.no/Mediasite/Showcase/default/Presentation/7d8b913259334b688986e970fae6fcb31d',
99 'only_matching': True,
100 },
101 {
102 # dashed id
103 'url': 'https://hitsmediaweb.h-its.org/mediasite/Play/2db6c271-681e-4f19-9af3-c60d1f82869b1d',
104 'only_matching': True,
105 }
106 ]
107
108 # look in Mediasite.Core.js (Mediasite.ContentStreamType[*])
109 _STREAM_TYPES = {
110 0: 'video1', # the main video
111 2: 'slide',
112 3: 'presentation',
113 4: 'video2', # screencast?
114 5: 'video3',
115 }
116
117 @staticmethod
118 def _extract_urls(webpage):
119 return [
120 unescapeHTML(mobj.group('url'))
121 for mobj in re.finditer(
122 r'(?xi)<iframe\b[^>]+\bsrc=(["\'])(?P<url>(?:(?:https?:)?//[^/]+)?/Mediasite/Play/%s(?:\?.*?)?)\1' % _ID_RE,
123 webpage)]
124
125 def _real_extract(self, url):
126 url, data = unsmuggle_url(url, {})
127 mobj = re.match(self._VALID_URL, url)
128 resource_id = mobj.group('id')
129 query = mobj.group('query')
130
131 webpage, urlh = self._download_webpage_handle(url, resource_id) # XXX: add UrlReferrer?
132 redirect_url = compat_str(urlh.geturl())
133
134 # XXX: might have also extracted UrlReferrer and QueryString from the html
135 service_path = compat_urlparse.urljoin(redirect_url, self._html_search_regex(
136 r'<div[^>]+\bid=["\']ServicePath[^>]+>(.+?)</div>', webpage, resource_id,
137 default='/Mediasite/PlayerService/PlayerService.svc/json'))
138
139 player_options = self._download_json(
140 '%s/GetPlayerOptions' % service_path, resource_id,
141 headers={
142 'Content-type': 'application/json; charset=utf-8',
143 'X-Requested-With': 'XMLHttpRequest',
144 },
145 data=json.dumps({
146 'getPlayerOptionsRequest': {
147 'ResourceId': resource_id,
148 'QueryString': query,
149 'UrlReferrer': data.get('UrlReferrer', ''),
150 'UseScreenReader': False,
151 }
152 }).encode('utf-8'))['d']
153
154 presentation = player_options['Presentation']
155 title = presentation['Title']
156
157 if presentation is None:
158 raise ExtractorError(
159 'Mediasite says: %s' % player_options['PlayerPresentationStatusMessage'],
160 expected=True)
161
162 thumbnails = []
163 formats = []
164 for snum, Stream in enumerate(presentation['Streams']):
165 stream_type = Stream.get('StreamType')
166 if stream_type is None:
167 continue
168
169 video_urls = Stream.get('VideoUrls')
170 if not isinstance(video_urls, list):
171 video_urls = []
172
173 stream_id = self._STREAM_TYPES.get(
174 stream_type, 'type%u' % stream_type)
175
176 stream_formats = []
177 for unum, VideoUrl in enumerate(video_urls):
178 video_url = url_or_none(VideoUrl.get('Location'))
179 if not video_url:
180 continue
181 # XXX: if Stream.get('CanChangeScheme', False), switch scheme to HTTP/HTTPS
182
183 media_type = VideoUrl.get('MediaType')
184 if media_type == 'SS':
185 stream_formats.extend(self._extract_ism_formats(
186 video_url, resource_id,
187 ism_id='%s-%u.%u' % (stream_id, snum, unum),
188 fatal=False))
189 elif media_type == 'Dash':
190 stream_formats.extend(self._extract_mpd_formats(
191 video_url, resource_id,
192 mpd_id='%s-%u.%u' % (stream_id, snum, unum),
193 fatal=False))
194 else:
195 stream_formats.append({
196 'format_id': '%s-%u.%u' % (stream_id, snum, unum),
197 'url': video_url,
198 'ext': mimetype2ext(VideoUrl.get('MimeType')),
199 })
200
201 # TODO: if Stream['HasSlideContent']:
202 # synthesise an MJPEG video stream '%s-%u.slides' % (stream_type, snum)
203 # from Stream['Slides']
204 # this will require writing a custom downloader...
205
206 # disprefer 'secondary' streams
207 if stream_type != 0:
208 for fmt in stream_formats:
209 fmt['preference'] = -1
210
211 thumbnail_url = Stream.get('ThumbnailUrl')
212 if thumbnail_url:
213 thumbnails.append({
214 'id': '%s-%u' % (stream_id, snum),
215 'url': urljoin(redirect_url, thumbnail_url),
216 'preference': -1 if stream_type != 0 else 0,
217 })
218 formats.extend(stream_formats)
219
220 self._sort_formats(formats)
221
222 # XXX: Presentation['Presenters']
223 # XXX: Presentation['Transcript']
224
225 return {
226 'id': resource_id,
227 'title': title,
228 'description': presentation.get('Description'),
229 'duration': float_or_none(presentation.get('Duration'), 1000),
230 'timestamp': float_or_none(presentation.get('UnixTime'), 1000),
231 'formats': formats,
232 'thumbnails': thumbnails,
233 }
234
235
236 class MediasiteCatalogIE(InfoExtractor):
237 _VALID_URL = r'''(?xi)
238 (?P<url>https?://[^/]+/Mediasite)
239 /Catalog/Full/
240 (?P<catalog_id>{0})
241 (?:
242 /(?P<current_folder_id>{0})
243 /(?P<root_dynamic_folder_id>{0})
244 )?
245 '''.format(_ID_RE)
246 _TESTS = [{
247 'url': 'http://events7.mediasite.com/Mediasite/Catalog/Full/631f9e48530d454381549f955d08c75e21',
248 'info_dict': {
249 'id': '631f9e48530d454381549f955d08c75e21',
250 'title': 'WCET Summit: Adaptive Learning in Higher Ed: Improving Outcomes Dynamically',
251 },
252 'playlist_count': 6,
253 'expected_warnings': ['is not a supported codec'],
254 }, {
255 # with CurrentFolderId and RootDynamicFolderId
256 'url': 'https://medaudio.medicine.iu.edu/Mediasite/Catalog/Full/9518c4a6c5cf4993b21cbd53e828a92521/97a9db45f7ab47428c77cd2ed74bb98f14/9518c4a6c5cf4993b21cbd53e828a92521',
257 'info_dict': {
258 'id': '9518c4a6c5cf4993b21cbd53e828a92521',
259 'title': 'IUSM Family and Friends Sessions',
260 },
261 'playlist_count': 2,
262 }, {
263 'url': 'http://uipsyc.mediasite.com/mediasite/Catalog/Full/d5d79287c75243c58c50fef50174ec1b21',
264 'only_matching': True,
265 }, {
266 # no AntiForgeryToken
267 'url': 'https://live.libraries.psu.edu/Mediasite/Catalog/Full/8376d4b24dd1457ea3bfe4cf9163feda21',
268 'only_matching': True,
269 }, {
270 'url': 'https://medaudio.medicine.iu.edu/Mediasite/Catalog/Full/9518c4a6c5cf4993b21cbd53e828a92521/97a9db45f7ab47428c77cd2ed74bb98f14/9518c4a6c5cf4993b21cbd53e828a92521',
271 'only_matching': True,
272 }, {
273 # dashed id
274 'url': 'http://events7.mediasite.com/Mediasite/Catalog/Full/631f9e48-530d-4543-8154-9f955d08c75e',
275 'only_matching': True,
276 }]
277
278 def _real_extract(self, url):
279 mobj = re.match(self._VALID_URL, url)
280 mediasite_url = mobj.group('url')
281 catalog_id = mobj.group('catalog_id')
282 current_folder_id = mobj.group('current_folder_id') or catalog_id
283 root_dynamic_folder_id = mobj.group('root_dynamic_folder_id')
284
285 webpage = self._download_webpage(url, catalog_id)
286
287 # AntiForgeryToken is optional (e.g. [1])
288 # 1. https://live.libraries.psu.edu/Mediasite/Catalog/Full/8376d4b24dd1457ea3bfe4cf9163feda21
289 anti_forgery_token = self._search_regex(
290 r'AntiForgeryToken\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1',
291 webpage, 'anti forgery token', default=None, group='value')
292 if anti_forgery_token:
293 anti_forgery_header = self._search_regex(
294 r'AntiForgeryHeaderName\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1',
295 webpage, 'anti forgery header name',
296 default='X-SOFO-AntiForgeryHeader', group='value')
297
298 data = {
299 'IsViewPage': True,
300 'IsNewFolder': True,
301 'AuthTicket': None,
302 'CatalogId': catalog_id,
303 'CurrentFolderId': current_folder_id,
304 'RootDynamicFolderId': root_dynamic_folder_id,
305 'ItemsPerPage': 1000,
306 'PageIndex': 0,
307 'PermissionMask': 'Execute',
308 'CatalogSearchType': 'SearchInFolder',
309 'SortBy': 'Date',
310 'SortDirection': 'Descending',
311 'StartDate': None,
312 'EndDate': None,
313 'StatusFilterList': None,
314 'PreviewKey': None,
315 'Tags': [],
316 }
317
318 headers = {
319 'Content-Type': 'application/json; charset=UTF-8',
320 'Referer': url,
321 'X-Requested-With': 'XMLHttpRequest',
322 }
323 if anti_forgery_token:
324 headers[anti_forgery_header] = anti_forgery_token
325
326 catalog = self._download_json(
327 '%s/Catalog/Data/GetPresentationsForFolder' % mediasite_url,
328 catalog_id, data=json.dumps(data).encode(), headers=headers)
329
330 entries = []
331 for video in catalog['PresentationDetailsList']:
332 if not isinstance(video, dict):
333 continue
334 video_id = str_or_none(video.get('Id'))
335 if not video_id:
336 continue
337 entries.append(self.url_result(
338 '%s/Play/%s' % (mediasite_url, video_id),
339 ie=MediasiteIE.ie_key(), video_id=video_id))
340
341 title = try_get(
342 catalog, lambda x: x['CurrentFolder']['Name'], compat_str)
343
344 return self.playlist_result(entries, catalog_id, title,)
345
346
347 class MediasiteNamedCatalogIE(InfoExtractor):
348 _VALID_URL = r'(?xi)(?P<url>https?://[^/]+/Mediasite)/Catalog/catalogs/(?P<catalog_name>[^/?#&]+)'
349 _TESTS = [{
350 'url': 'https://msite.misis.ru/Mediasite/Catalog/catalogs/2016-industrial-management-skriabin-o-o',
351 'only_matching': True,
352 }]
353
354 def _real_extract(self, url):
355 mobj = re.match(self._VALID_URL, url)
356 mediasite_url = mobj.group('url')
357 catalog_name = mobj.group('catalog_name')
358
359 webpage = self._download_webpage(url, catalog_name)
360
361 catalog_id = self._search_regex(
362 r'CatalogId\s*:\s*["\'](%s)' % _ID_RE, webpage, 'catalog id')
363
364 return self.url_result(
365 '%s/Catalog/Full/%s' % (mediasite_url, catalog_id),
366 ie=MediasiteCatalogIE.ie_key(), video_id=catalog_id)