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     unescapeHTML,
  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     _TESTS = [
  29         {
  30             'url': 'https://hitsmediaweb.h-its.org/mediasite/Play/2db6c271681e4f199af3c60d1f82869b1d',
  31             'info_dict': {
  32                 'id': '2db6c271681e4f199af3c60d1f82869b1d',
  33                 'ext': 'mp4',
  34                 'title': 'Lecture: Tuesday, September 20, 2016 - Sir Andrew Wiles',
  35                 '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.',
  36                 'timestamp': 1474268400.0,
  37                 'upload_date': '20160919',
  38             },
  39         },
  40         {
  41             'url': 'http://mediasite.uib.no/Mediasite/Play/90bb363295d945d6b548c867d01181361d?catalog=a452b7df-9ae1-46b7-a3ba-aceeb285f3eb',
  42             'info_dict': {
  43                 'id': '90bb363295d945d6b548c867d01181361d',
  44                 'ext': 'mp4',
  45                 'upload_date': '20150429',
  46                 'title': '5) IT-forum 2015-Dag 1  - Dungbeetle -  How and why Rain created a tiny bug tracker for Unity',
  47                 'timestamp': 1430311380.0,
  48             },
  49         },
  50         {
  51             'url': 'https://collegerama.tudelft.nl/Mediasite/Play/585a43626e544bdd97aeb71a0ec907a01d',
  52             'md5': '481fda1c11f67588c0d9d8fbdced4e39',
  53             'info_dict': {
  54                 'id': '585a43626e544bdd97aeb71a0ec907a01d',
  55                 'ext': 'mp4',
  56                 'title': 'Een nieuwe wereld: waarden, bewustzijn en techniek van de mensheid 2.0.',
  57                 'description': '',
  58                 'thumbnail': r're:^https?://.*\.jpg(?:\?.*)?$',
  59                 'duration': 7713.088,
  60                 'timestamp': 1413309600,
  61                 'upload_date': '20141014',
  62             },
  63         },
  64         {
  65             'url': 'https://collegerama.tudelft.nl/Mediasite/Play/86a9ea9f53e149079fbdb4202b521ed21d?catalog=fd32fd35-6c99-466c-89d4-cd3c431bc8a4',
  66             'md5': 'ef1fdded95bdf19b12c5999949419c92',
  67             'info_dict': {
  68                 'id': '86a9ea9f53e149079fbdb4202b521ed21d',
  69                 'ext': 'wmv',
  70                 'title': '64ste Vakantiecursus: Afvalwater',
  71                 'description': 'md5:7fd774865cc69d972f542b157c328305',
  72                 'thumbnail': r're:^https?://.*\.jpg(?:\?.*?)?$',
  73                 'duration': 10853,
  74                 'timestamp': 1326446400,
  75                 'upload_date': '20120113',
  76             },
  77         },
  78         {
  79             'url': 'http://digitalops.sandia.gov/Mediasite/Play/24aace4429fc450fb5b38cdbf424a66e1d',
  80             'md5': '9422edc9b9a60151727e4b6d8bef393d',
  81             'info_dict': {
  82                 'id': '24aace4429fc450fb5b38cdbf424a66e1d',
  83                 'ext': 'mp4',
  84                 'title': 'Xyce Software Training - Section 1',
  85                 'description': r're:(?s)SAND Number: SAND 2013-7800.{200,}',
  86                 'upload_date': '20120409',
  87                 'timestamp': 1333983600,
  88                 'duration': 7794,
  89             }
  90         },
  91         {
  92             'url': 'https://collegerama.tudelft.nl/Mediasite/Showcase/livebroadcast/Presentation/ada7020854f743c49fbb45c9ec7dbb351d',
  93             'only_matching': True,
  94         },
  95         {
  96             'url': 'https://mediasite.ntnu.no/Mediasite/Showcase/default/Presentation/7d8b913259334b688986e970fae6fcb31d',
  97             'only_matching': True,
  98         },
  99         {
 100             # dashed id
 101             'url': 'https://hitsmediaweb.h-its.org/mediasite/Play/2db6c271-681e-4f19-9af3-c60d1f82869b1d',
 102             'only_matching': True,
 103         }
 104     ]
 105
 106     # look in Mediasite.Core.js (Mediasite.ContentStreamType[*])
 107     _STREAM_TYPES = {
 108         0: 'video1',  # the main video
 109         2: 'slide',
 110         3: 'presentation',
 111         4: 'video2',  # screencast?
 112         5: 'video3',
 113     }
 114
 115     @staticmethod
 116     def _extract_urls(webpage):
 117         return [
 118             unescapeHTML(mobj.group('url'))
 119             for mobj in re.finditer(
 120                 r'(?xi)<iframe\b[^>]+\bsrc=(["\'])(?P<url>(?:(?:https?:)?//[^/]+)?/Mediasite/Play/%s(?:\?.*?)?)\1' % _ID_RE,
 121                 webpage)]
 122
 123     def __extract_slides(self, *, stream_id, snum, Stream, duration, images):
 124         slide_base_url = Stream['SlideBaseUrl']
 125
 126         fname_template = Stream['SlideImageFileNameTemplate']
 127         if fname_template != 'slide_{0:D4}.jpg':
 128             self.report_warning('Unusual slide file name template; report a bug if slide downloading fails')
 129         fname_template = re.sub(r'\{0:D([0-9]+)\}', r'{0:0\1}', fname_template)
 130
 131         fragments = []
 132         for i, slide in enumerate(Stream['Slides']):
 133             if i == 0:
 134                 if slide['Time'] > 0:
 135                     default_slide = images.get('DefaultSlide')
 136                     if default_slide is None:
 137                         default_slide = images.get('DefaultStreamImage')
 138                     if default_slide is not None:
 139                         default_slide = default_slide['ImageFilename']
 140                     if default_slide is not None:
 141                         fragments.append({
 142                             'path': default_slide,
 143                             'duration': slide['Time'] / 1000,
 144                         })
 145
 146             next_time = try_call(
 147                 lambda: Stream['Slides'][i + 1]['Time'],
 148                 lambda: duration,
 149                 lambda: slide['Time'],
 150                 expected_type=(int, float))
 151
 152             fragments.append({
 153                 'path': fname_template.format(slide.get('Number', i + 1)),
 154                 'duration': (next_time - slide['Time']) / 1000
 155             })
 156
 157         return {
 158             'format_id': '%s-%u.slides' % (stream_id, snum),
 159             'ext': 'mhtml',
 160             'url': slide_base_url,
 161             'protocol': 'mhtml',
 162             'acodec': 'none',
 163             'vcodec': 'none',
 164             'format_note': 'Slides',
 165             'fragments': fragments,
 166             'fragment_base_url': slide_base_url,
 167         }
 168
 169     def _real_extract(self, url):
 170         url, data = unsmuggle_url(url, {})
 171         mobj = self._match_valid_url(url)
 172         resource_id = mobj.group('id')
 173         query = mobj.group('query')
 174
 175         webpage, urlh = self._download_webpage_handle(url, resource_id)  # XXX: add UrlReferrer?
 176         redirect_url = urlh.geturl()
 177
 178         # XXX: might have also extracted UrlReferrer and QueryString from the html
 179         service_path = compat_urlparse.urljoin(redirect_url, self._html_search_regex(
 180             r'<div[^>]+\bid=["\']ServicePath[^>]+>(.+?)</div>', webpage, resource_id,
 181             default='/Mediasite/PlayerService/PlayerService.svc/json'))
 182
 183         player_options = self._download_json(
 184             '%s/GetPlayerOptions' % service_path, resource_id,
 185             headers={
 186                 'Content-type': 'application/json; charset=utf-8',
 187                 'X-Requested-With': 'XMLHttpRequest',
 188             },
 189             data=json.dumps({
 190                 'getPlayerOptionsRequest': {
 191                     'ResourceId': resource_id,
 192                     'QueryString': query,
 193                     'UrlReferrer': data.get('UrlReferrer', ''),
 194                     'UseScreenReader': False,
 195                 }
 196             }).encode('utf-8'))['d']
 197
 198         presentation = player_options['Presentation']
 199         title = presentation['Title']
 200
 201         if presentation is None:
 202             raise ExtractorError(
 203                 'Mediasite says: %s' % player_options['PlayerPresentationStatusMessage'],
 204                 expected=True)
 205
 206         thumbnails = []
 207         formats = []
 208         for snum, Stream in enumerate(presentation['Streams']):
 209             stream_type = Stream.get('StreamType')
 210             if stream_type is None:
 211                 continue
 212
 213             video_urls = Stream.get('VideoUrls')
 214             if not isinstance(video_urls, list):
 215                 video_urls = []
 216
 217             stream_id = self._STREAM_TYPES.get(
 218                 stream_type, 'type%u' % stream_type)
 219
 220             stream_formats = []
 221             for unum, VideoUrl in enumerate(video_urls):
 222                 video_url = url_or_none(VideoUrl.get('Location'))
 223                 if not video_url:
 224                     continue
 225                 # XXX: if Stream.get('CanChangeScheme', False), switch scheme to HTTP/HTTPS
 226
 227                 media_type = VideoUrl.get('MediaType')
 228                 if media_type == 'SS':
 229                     stream_formats.extend(self._extract_ism_formats(
 230                         video_url, resource_id,
 231                         ism_id='%s-%u.%u' % (stream_id, snum, unum),
 232                         fatal=False))
 233                 elif media_type == 'Dash':
 234                     stream_formats.extend(self._extract_mpd_formats(
 235                         video_url, resource_id,
 236                         mpd_id='%s-%u.%u' % (stream_id, snum, unum),
 237                         fatal=False))
 238                 else:
 239                     stream_formats.append({
 240                         'format_id': '%s-%u.%u' % (stream_id, snum, unum),
 241                         'url': video_url,
 242                         'ext': mimetype2ext(VideoUrl.get('MimeType')),
 243                     })
 244
 245             if Stream.get('HasSlideContent', False):
 246                 images = player_options['PlayerLayoutOptions']['Images']
 247                 stream_formats.append(self.__extract_slides(
 248                     stream_id=stream_id,
 249                     snum=snum,
 250                     Stream=Stream,
 251                     duration=presentation.get('Duration'),
 252                     images=images,
 253                 ))
 254
 255             # disprefer 'secondary' streams
 256             if stream_type != 0:
 257                 for fmt in stream_formats:
 258                     fmt['quality'] = -10
 259
 260             thumbnail_url = Stream.get('ThumbnailUrl')
 261             if thumbnail_url:
 262                 thumbnails.append({
 263                     'id': '%s-%u' % (stream_id, snum),
 264                     'url': urljoin(redirect_url, thumbnail_url),
 265                     'preference': -1 if stream_type != 0 else 0,
 266                 })
 267             formats.extend(stream_formats)
 268
 269         self._sort_formats(formats)
 270
 271         # XXX: Presentation['Presenters']
 272         # XXX: Presentation['Transcript']
 273
 274         return {
 275             'id': resource_id,
 276             'title': title,
 277             'description': presentation.get('Description'),
 278             'duration': float_or_none(presentation.get('Duration'), 1000),
 279             'timestamp': float_or_none(presentation.get('UnixTime'), 1000),
 280             'formats': formats,
 281             'thumbnails': thumbnails,
 282         }
 283
 284
 285 class MediasiteCatalogIE(InfoExtractor):
 286     _VALID_URL = r'''(?xi)
 287                         (?P<url>https?://[^/]+/Mediasite)
 288                         /Catalog/Full/
 289                         (?P<catalog_id>{0})
 290                         (?:
 291                             /(?P<current_folder_id>{0})
 292                             /(?P<root_dynamic_folder_id>{0})
 293                         )?
 294                     '''.format(_ID_RE)
 295     _TESTS = [{
 296         'url': 'http://events7.mediasite.com/Mediasite/Catalog/Full/631f9e48530d454381549f955d08c75e21',
 297         'info_dict': {
 298             'id': '631f9e48530d454381549f955d08c75e21',
 299             'title': 'WCET Summit: Adaptive Learning in Higher Ed: Improving Outcomes Dynamically',
 300         },
 301         'playlist_count': 6,
 302         'expected_warnings': ['is not a supported codec'],
 303     }, {
 304         # with CurrentFolderId and RootDynamicFolderId
 305         'url': 'https://medaudio.medicine.iu.edu/Mediasite/Catalog/Full/9518c4a6c5cf4993b21cbd53e828a92521/97a9db45f7ab47428c77cd2ed74bb98f14/9518c4a6c5cf4993b21cbd53e828a92521',
 306         'info_dict': {
 307             'id': '9518c4a6c5cf4993b21cbd53e828a92521',
 308             'title': 'IUSM Family and Friends Sessions',
 309         },
 310         'playlist_count': 2,
 311     }, {
 312         'url': 'http://uipsyc.mediasite.com/mediasite/Catalog/Full/d5d79287c75243c58c50fef50174ec1b21',
 313         'only_matching': True,
 314     }, {
 315         # no AntiForgeryToken
 316         'url': 'https://live.libraries.psu.edu/Mediasite/Catalog/Full/8376d4b24dd1457ea3bfe4cf9163feda21',
 317         'only_matching': True,
 318     }, {
 319         'url': 'https://medaudio.medicine.iu.edu/Mediasite/Catalog/Full/9518c4a6c5cf4993b21cbd53e828a92521/97a9db45f7ab47428c77cd2ed74bb98f14/9518c4a6c5cf4993b21cbd53e828a92521',
 320         'only_matching': True,
 321     }, {
 322         # dashed id
 323         'url': 'http://events7.mediasite.com/Mediasite/Catalog/Full/631f9e48-530d-4543-8154-9f955d08c75e',
 324         'only_matching': True,
 325     }]
 326
 327     def _real_extract(self, url):
 328         mobj = self._match_valid_url(url)
 329         mediasite_url = mobj.group('url')
 330         catalog_id = mobj.group('catalog_id')
 331         current_folder_id = mobj.group('current_folder_id') or catalog_id
 332         root_dynamic_folder_id = mobj.group('root_dynamic_folder_id')
 333
 334         webpage = self._download_webpage(url, catalog_id)
 335
 336         # AntiForgeryToken is optional (e.g. [1])
 337         # 1. https://live.libraries.psu.edu/Mediasite/Catalog/Full/8376d4b24dd1457ea3bfe4cf9163feda21
 338         anti_forgery_token = self._search_regex(
 339             r'AntiForgeryToken\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1',
 340             webpage, 'anti forgery token', default=None, group='value')
 341         if anti_forgery_token:
 342             anti_forgery_header = self._search_regex(
 343                 r'AntiForgeryHeaderName\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1',
 344                 webpage, 'anti forgery header name',
 345                 default='X-SOFO-AntiForgeryHeader', group='value')
 346
 347         data = {
 348             'IsViewPage': True,
 349             'IsNewFolder': True,
 350             'AuthTicket': None,
 351             'CatalogId': catalog_id,
 352             'CurrentFolderId': current_folder_id,
 353             'RootDynamicFolderId': root_dynamic_folder_id,
 354             'ItemsPerPage': 1000,
 355             'PageIndex': 0,
 356             'PermissionMask': 'Execute',
 357             'CatalogSearchType': 'SearchInFolder',
 358             'SortBy': 'Date',
 359             'SortDirection': 'Descending',
 360             'StartDate': None,
 361             'EndDate': None,
 362             'StatusFilterList': None,
 363             'PreviewKey': None,
 364             'Tags': [],
 365         }
 366
 367         headers = {
 368             'Content-Type': 'application/json; charset=UTF-8',
 369             'Referer': url,
 370             'X-Requested-With': 'XMLHttpRequest',
 371         }
 372         if anti_forgery_token:
 373             headers[anti_forgery_header] = anti_forgery_token
 374
 375         catalog = self._download_json(
 376             '%s/Catalog/Data/GetPresentationsForFolder' % mediasite_url,
 377             catalog_id, data=json.dumps(data).encode(), headers=headers)
 378
 379         entries = []
 380         for video in catalog['PresentationDetailsList']:
 381             if not isinstance(video, dict):
 382                 continue
 383             video_id = str_or_none(video.get('Id'))
 384             if not video_id:
 385                 continue
 386             entries.append(self.url_result(
 387                 '%s/Play/%s' % (mediasite_url, video_id),
 388                 ie=MediasiteIE.ie_key(), video_id=video_id))
 389
 390         title = try_get(
 391             catalog, lambda x: x['CurrentFolder']['Name'], compat_str)
 392
 393         return self.playlist_result(entries, catalog_id, title,)
 394
 395
 396 class MediasiteNamedCatalogIE(InfoExtractor):
 397     _VALID_URL = r'(?xi)(?P<url>https?://[^/]+/Mediasite)/Catalog/catalogs/(?P<catalog_name>[^/?#&]+)'
 398     _TESTS = [{
 399         'url': 'https://msite.misis.ru/Mediasite/Catalog/catalogs/2016-industrial-management-skriabin-o-o',
 400         'only_matching': True,
 401     }]
 402
 403     def _real_extract(self, url):
 404         mobj = self._match_valid_url(url)
 405         mediasite_url = mobj.group('url')
 406         catalog_name = mobj.group('catalog_name')
 407
 408         webpage = self._download_webpage(url, catalog_name)
 409
 410         catalog_id = self._search_regex(
 411             r'CatalogId\s*:\s*["\'](%s)' % _ID_RE, webpage, 'catalog id')
 412
 413         return self.url_result(
 414             '%s/Catalog/Full/%s' % (mediasite_url, catalog_id),
 415             ie=MediasiteCatalogIE.ie_key(), video_id=catalog_id)