yt_dlp/extractor/mediasite.py

   1 import re
   2 import json
   3
   4 from .common import InfoExtractor
   5 from ..compat import (
   6     compat_str,
   7     compat_urlparse,
   8 )
   9 from ..utils import (
  10     ExtractorError,
  11     float_or_none,
  12     mimetype2ext,
  13     str_or_none,
  14     try_call,
  15     try_get,
  16     smuggle_url,
  17     unsmuggle_url,
  18     url_or_none,
  19     urljoin,
  20 )
  21
  22
  23 _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})'
  24
  25
  26 class MediasiteIE(InfoExtractor):
  27     _VALID_URL = r'(?xi)https?://[^/]+/Mediasite/(?:Play|Showcase/[^/#?]+/Presentation)/(?P<id>%s)(?P<query>\?[^#]+|)' % _ID_RE
  28     _EMBED_REGEX = [r'(?xi)<iframe\b[^>]+\bsrc=(["\'])(?P<url>(?:(?:https?:)?//[^/]+)?/Mediasite/Play/%s(?:\?.*?)?)\1' % _ID_RE]
  29     _TESTS = [
  30         {
  31             'url': 'https://hitsmediaweb.h-its.org/mediasite/Play/2db6c271681e4f199af3c60d1f82869b1d',
  32             'info_dict': {
  33                 'id': '2db6c271681e4f199af3c60d1f82869b1d',
  34                 'ext': 'mp4',
  35                 'title': 'Lecture: Tuesday, September 20, 2016 - Sir Andrew Wiles',
  36                 '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.',
  37                 'timestamp': 1474268400.0,
  38                 'upload_date': '20160919',
  39             },
  40         },
  41         {
  42             'url': 'http://mediasite.uib.no/Mediasite/Play/90bb363295d945d6b548c867d01181361d?catalog=a452b7df-9ae1-46b7-a3ba-aceeb285f3eb',
  43             'info_dict': {
  44                 'id': '90bb363295d945d6b548c867d01181361d',
  45                 'ext': 'mp4',
  46                 'upload_date': '20150429',
  47                 'title': '5) IT-forum 2015-Dag 1  - Dungbeetle -  How and why Rain created a tiny bug tracker for Unity',
  48                 'timestamp': 1430311380.0,
  49             },
  50         },
  51         {
  52             'url': 'https://collegerama.tudelft.nl/Mediasite/Play/585a43626e544bdd97aeb71a0ec907a01d',
  53             'md5': '481fda1c11f67588c0d9d8fbdced4e39',
  54             'info_dict': {
  55                 'id': '585a43626e544bdd97aeb71a0ec907a01d',
  56                 'ext': 'mp4',
  57                 'title': 'Een nieuwe wereld: waarden, bewustzijn en techniek van de mensheid 2.0.',
  58                 'description': '',
  59                 'thumbnail': r're:^https?://.*\.jpg(?:\?.*)?$',
  60                 'duration': 7713.088,
  61                 'timestamp': 1413309600,
  62                 'upload_date': '20141014',
  63             },
  64         },
  65         {
  66             'url': 'https://collegerama.tudelft.nl/Mediasite/Play/86a9ea9f53e149079fbdb4202b521ed21d?catalog=fd32fd35-6c99-466c-89d4-cd3c431bc8a4',
  67             'md5': 'ef1fdded95bdf19b12c5999949419c92',
  68             'info_dict': {
  69                 'id': '86a9ea9f53e149079fbdb4202b521ed21d',
  70                 'ext': 'wmv',
  71                 'title': '64ste Vakantiecursus: Afvalwater',
  72                 'description': 'md5:7fd774865cc69d972f542b157c328305',
  73                 'thumbnail': r're:^https?://.*\.jpg(?:\?.*?)?$',
  74                 'duration': 10853,
  75                 'timestamp': 1326446400,
  76                 'upload_date': '20120113',
  77             },
  78         },
  79         {
  80             'url': 'http://digitalops.sandia.gov/Mediasite/Play/24aace4429fc450fb5b38cdbf424a66e1d',
  81             'md5': '9422edc9b9a60151727e4b6d8bef393d',
  82             'info_dict': {
  83                 'id': '24aace4429fc450fb5b38cdbf424a66e1d',
  84                 'ext': 'mp4',
  85                 'title': 'Xyce Software Training - Section 1',
  86                 'description': r're:(?s)SAND Number: SAND 2013-7800.{200,}',
  87                 'upload_date': '20120409',
  88                 'timestamp': 1333983600,
  89                 'duration': 7794,
  90             }
  91         },
  92         {
  93             'url': 'https://collegerama.tudelft.nl/Mediasite/Showcase/livebroadcast/Presentation/ada7020854f743c49fbb45c9ec7dbb351d',
  94             'only_matching': True,
  95         },
  96         {
  97             'url': 'https://mediasite.ntnu.no/Mediasite/Showcase/default/Presentation/7d8b913259334b688986e970fae6fcb31d',
  98             'only_matching': True,
  99         },
 100         {
 101             # dashed id
 102             'url': 'https://hitsmediaweb.h-its.org/mediasite/Play/2db6c271-681e-4f19-9af3-c60d1f82869b1d',
 103             'only_matching': True,
 104         }
 105     ]
 106
 107     # look in Mediasite.Core.js (Mediasite.ContentStreamType[*])
 108     _STREAM_TYPES = {
 109         0: 'video1',  # the main video
 110         2: 'slide',
 111         3: 'presentation',
 112         4: 'video2',  # screencast?
 113         5: 'video3',
 114     }
 115
 116     @classmethod
 117     def _extract_embed_urls(cls, url, webpage):
 118         for embed_url in super()._extract_embed_urls(url, webpage):
 119             yield smuggle_url(embed_url, {'UrlReferrer': url})
 120
 121     def __extract_slides(self, *, stream_id, snum, Stream, duration, images):
 122         slide_base_url = Stream['SlideBaseUrl']
 123
 124         fname_template = Stream['SlideImageFileNameTemplate']
 125         if fname_template != 'slide_{0:D4}.jpg':
 126             self.report_warning('Unusual slide file name template; report a bug if slide downloading fails')
 127         fname_template = re.sub(r'\{0:D([0-9]+)\}', r'{0:0\1}', fname_template)
 128
 129         fragments = []
 130         for i, slide in enumerate(Stream['Slides']):
 131             if i == 0:
 132                 if slide['Time'] > 0:
 133                     default_slide = images.get('DefaultSlide')
 134                     if default_slide is None:
 135                         default_slide = images.get('DefaultStreamImage')
 136                     if default_slide is not None:
 137                         default_slide = default_slide['ImageFilename']
 138                     if default_slide is not None:
 139                         fragments.append({
 140                             'path': default_slide,
 141                             'duration': slide['Time'] / 1000,
 142                         })
 143
 144             next_time = try_call(
 145                 lambda: Stream['Slides'][i + 1]['Time'],
 146                 lambda: duration,
 147                 lambda: slide['Time'],
 148                 expected_type=(int, float))
 149
 150             fragments.append({
 151                 'path': fname_template.format(slide.get('Number', i + 1)),
 152                 'duration': (next_time - slide['Time']) / 1000
 153             })
 154
 155         return {
 156             'format_id': '%s-%u.slides' % (stream_id, snum),
 157             'ext': 'mhtml',
 158             'url': slide_base_url,
 159             'protocol': 'mhtml',
 160             'acodec': 'none',
 161             'vcodec': 'none',
 162             'format_note': 'Slides',
 163             'fragments': fragments,
 164             'fragment_base_url': slide_base_url,
 165         }
 166
 167     def _real_extract(self, url):
 168         url, data = unsmuggle_url(url, {})
 169         mobj = self._match_valid_url(url)
 170         resource_id = mobj.group('id')
 171         query = mobj.group('query')
 172
 173         webpage, urlh = self._download_webpage_handle(url, resource_id)  # XXX: add UrlReferrer?
 174         redirect_url = urlh.url
 175
 176         # XXX: might have also extracted UrlReferrer and QueryString from the html
 177         service_path = compat_urlparse.urljoin(redirect_url, self._html_search_regex(
 178             r'<div[^>]+\bid=["\']ServicePath[^>]+>(.+?)</div>', webpage, resource_id,
 179             default='/Mediasite/PlayerService/PlayerService.svc/json'))
 180
 181         player_options = self._download_json(
 182             '%s/GetPlayerOptions' % service_path, resource_id,
 183             headers={
 184                 'Content-type': 'application/json; charset=utf-8',
 185                 'X-Requested-With': 'XMLHttpRequest',
 186             },
 187             data=json.dumps({
 188                 'getPlayerOptionsRequest': {
 189                     'ResourceId': resource_id,
 190                     'QueryString': query,
 191                     'UrlReferrer': data.get('UrlReferrer', ''),
 192                     'UseScreenReader': False,
 193                 }
 194             }).encode('utf-8'))['d']
 195
 196         presentation = player_options['Presentation']
 197         title = presentation['Title']
 198
 199         if presentation is None:
 200             raise ExtractorError(
 201                 'Mediasite says: %s' % player_options['PlayerPresentationStatusMessage'],
 202                 expected=True)
 203
 204         thumbnails = []
 205         formats = []
 206         for snum, Stream in enumerate(presentation['Streams']):
 207             stream_type = Stream.get('StreamType')
 208             if stream_type is None:
 209                 continue
 210
 211             video_urls = Stream.get('VideoUrls')
 212             if not isinstance(video_urls, list):
 213                 video_urls = []
 214
 215             stream_id = self._STREAM_TYPES.get(
 216                 stream_type, 'type%u' % stream_type)
 217
 218             stream_formats = []
 219             for unum, VideoUrl in enumerate(video_urls):
 220                 video_url = url_or_none(VideoUrl.get('Location'))
 221                 if not video_url:
 222                     continue
 223                 # XXX: if Stream.get('CanChangeScheme', False), switch scheme to HTTP/HTTPS
 224
 225                 media_type = VideoUrl.get('MediaType')
 226                 if media_type == 'SS':
 227                     stream_formats.extend(self._extract_ism_formats(
 228                         video_url, resource_id,
 229                         ism_id='%s-%u.%u' % (stream_id, snum, unum),
 230                         fatal=False))
 231                 elif media_type == 'Dash':
 232                     stream_formats.extend(self._extract_mpd_formats(
 233                         video_url, resource_id,
 234                         mpd_id='%s-%u.%u' % (stream_id, snum, unum),
 235                         fatal=False))
 236                 else:
 237                     stream_formats.append({
 238                         'format_id': '%s-%u.%u' % (stream_id, snum, unum),
 239                         'url': video_url,
 240                         'ext': mimetype2ext(VideoUrl.get('MimeType')),
 241                     })
 242
 243             if Stream.get('HasSlideContent', False):
 244                 images = player_options['PlayerLayoutOptions']['Images']
 245                 stream_formats.append(self.__extract_slides(
 246                     stream_id=stream_id,
 247                     snum=snum,
 248                     Stream=Stream,
 249                     duration=presentation.get('Duration'),
 250                     images=images,
 251                 ))
 252
 253             # disprefer 'secondary' streams
 254             if stream_type != 0:
 255                 for fmt in stream_formats:
 256                     fmt['quality'] = -10
 257
 258             thumbnail_url = Stream.get('ThumbnailUrl')
 259             if thumbnail_url:
 260                 thumbnails.append({
 261                     'id': '%s-%u' % (stream_id, snum),
 262                     'url': urljoin(redirect_url, thumbnail_url),
 263                     'preference': -1 if stream_type != 0 else 0,
 264                 })
 265             formats.extend(stream_formats)
 266
 267         # XXX: Presentation['Presenters']
 268         # XXX: Presentation['Transcript']
 269
 270         return {
 271             'id': resource_id,
 272             'title': title,
 273             'description': presentation.get('Description'),
 274             'duration': float_or_none(presentation.get('Duration'), 1000),
 275             'timestamp': float_or_none(presentation.get('UnixTime'), 1000),
 276             'formats': formats,
 277             'thumbnails': thumbnails,
 278         }
 279
 280
 281 class MediasiteCatalogIE(InfoExtractor):
 282     _VALID_URL = r'''(?xi)
 283                         (?P<url>https?://[^/]+/Mediasite)
 284                         /Catalog/Full/
 285                         (?P<catalog_id>{0})
 286                         (?:
 287                             /(?P<current_folder_id>{0})
 288                             /(?P<root_dynamic_folder_id>{0})
 289                         )?
 290                     '''.format(_ID_RE)
 291     _TESTS = [{
 292         'url': 'http://events7.mediasite.com/Mediasite/Catalog/Full/631f9e48530d454381549f955d08c75e21',
 293         'info_dict': {
 294             'id': '631f9e48530d454381549f955d08c75e21',
 295             'title': 'WCET Summit: Adaptive Learning in Higher Ed: Improving Outcomes Dynamically',
 296         },
 297         'playlist_count': 6,
 298         'expected_warnings': ['is not a supported codec'],
 299     }, {
 300         # with CurrentFolderId and RootDynamicFolderId
 301         'url': 'https://medaudio.medicine.iu.edu/Mediasite/Catalog/Full/9518c4a6c5cf4993b21cbd53e828a92521/97a9db45f7ab47428c77cd2ed74bb98f14/9518c4a6c5cf4993b21cbd53e828a92521',
 302         'info_dict': {
 303             'id': '9518c4a6c5cf4993b21cbd53e828a92521',
 304             'title': 'IUSM Family and Friends Sessions',
 305         },
 306         'playlist_count': 2,
 307     }, {
 308         'url': 'http://uipsyc.mediasite.com/mediasite/Catalog/Full/d5d79287c75243c58c50fef50174ec1b21',
 309         'only_matching': True,
 310     }, {
 311         # no AntiForgeryToken
 312         'url': 'https://live.libraries.psu.edu/Mediasite/Catalog/Full/8376d4b24dd1457ea3bfe4cf9163feda21',
 313         'only_matching': True,
 314     }, {
 315         'url': 'https://medaudio.medicine.iu.edu/Mediasite/Catalog/Full/9518c4a6c5cf4993b21cbd53e828a92521/97a9db45f7ab47428c77cd2ed74bb98f14/9518c4a6c5cf4993b21cbd53e828a92521',
 316         'only_matching': True,
 317     }, {
 318         # dashed id
 319         'url': 'http://events7.mediasite.com/Mediasite/Catalog/Full/631f9e48-530d-4543-8154-9f955d08c75e',
 320         'only_matching': True,
 321     }]
 322
 323     def _real_extract(self, url):
 324         mobj = self._match_valid_url(url)
 325         mediasite_url = mobj.group('url')
 326         catalog_id = mobj.group('catalog_id')
 327         current_folder_id = mobj.group('current_folder_id') or catalog_id
 328         root_dynamic_folder_id = mobj.group('root_dynamic_folder_id')
 329
 330         webpage = self._download_webpage(url, catalog_id)
 331
 332         # AntiForgeryToken is optional (e.g. [1])
 333         # 1. https://live.libraries.psu.edu/Mediasite/Catalog/Full/8376d4b24dd1457ea3bfe4cf9163feda21
 334         anti_forgery_token = self._search_regex(
 335             r'AntiForgeryToken\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1',
 336             webpage, 'anti forgery token', default=None, group='value')
 337         if anti_forgery_token:
 338             anti_forgery_header = self._search_regex(
 339                 r'AntiForgeryHeaderName\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1',
 340                 webpage, 'anti forgery header name',
 341                 default='X-SOFO-AntiForgeryHeader', group='value')
 342
 343         data = {
 344             'IsViewPage': True,
 345             'IsNewFolder': True,
 346             'AuthTicket': None,
 347             'CatalogId': catalog_id,
 348             'CurrentFolderId': current_folder_id,
 349             'RootDynamicFolderId': root_dynamic_folder_id,
 350             'ItemsPerPage': 1000,
 351             'PageIndex': 0,
 352             'PermissionMask': 'Execute',
 353             'CatalogSearchType': 'SearchInFolder',
 354             'SortBy': 'Date',
 355             'SortDirection': 'Descending',
 356             'StartDate': None,
 357             'EndDate': None,
 358             'StatusFilterList': None,
 359             'PreviewKey': None,
 360             'Tags': [],
 361         }
 362
 363         headers = {
 364             'Content-Type': 'application/json; charset=UTF-8',
 365             'Referer': url,
 366             'X-Requested-With': 'XMLHttpRequest',
 367         }
 368         if anti_forgery_token:
 369             headers[anti_forgery_header] = anti_forgery_token
 370
 371         catalog = self._download_json(
 372             '%s/Catalog/Data/GetPresentationsForFolder' % mediasite_url,
 373             catalog_id, data=json.dumps(data).encode(), headers=headers)
 374
 375         entries = []
 376         for video in catalog['PresentationDetailsList']:
 377             if not isinstance(video, dict):
 378                 continue
 379             video_id = str_or_none(video.get('Id'))
 380             if not video_id:
 381                 continue
 382             entries.append(self.url_result(
 383                 '%s/Play/%s' % (mediasite_url, video_id),
 384                 ie=MediasiteIE.ie_key(), video_id=video_id))
 385
 386         title = try_get(
 387             catalog, lambda x: x['CurrentFolder']['Name'], compat_str)
 388
 389         return self.playlist_result(entries, catalog_id, title,)
 390
 391
 392 class MediasiteNamedCatalogIE(InfoExtractor):
 393     _VALID_URL = r'(?xi)(?P<url>https?://[^/]+/Mediasite)/Catalog/catalogs/(?P<catalog_name>[^/?#&]+)'
 394     _TESTS = [{
 395         'url': 'https://msite.misis.ru/Mediasite/Catalog/catalogs/2016-industrial-management-skriabin-o-o',
 396         'only_matching': True,
 397     }]
 398
 399     def _real_extract(self, url):
 400         mobj = self._match_valid_url(url)
 401         mediasite_url = mobj.group('url')
 402         catalog_name = mobj.group('catalog_name')
 403
 404         webpage = self._download_webpage(url, catalog_name)
 405
 406         catalog_id = self._search_regex(
 407             r'CatalogId\s*:\s*["\'](%s)' % _ID_RE, webpage, 'catalog id')
 408
 409         return self.url_result(
 410             '%s/Catalog/Full/%s' % (mediasite_url, catalog_id),
 411             ie=MediasiteCatalogIE.ie_key(), video_id=catalog_id)