yt_dlp/extractor/mediasite.py

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