yt_dlp/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/[^/#?]+/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 __extract_slides(self, *, stream_id, snum, Stream, duration, images):
 126         slide_base_url = Stream['SlideBaseUrl']
 127
 128         fname_template = Stream['SlideImageFileNameTemplate']
 129         if fname_template != 'slide_{0:D4}.jpg':
 130             self.report_warning('Unusual slide file name template; report a bug if slide downloading fails')
 131         fname_template = re.sub(r'\{0:D([0-9]+)\}', r'{0:0\1}', fname_template)
 132
 133         fragments = []
 134         for i, slide in enumerate(Stream['Slides']):
 135             if i == 0:
 136                 if slide['Time'] > 0:
 137                     default_slide = images.get('DefaultSlide')
 138                     if default_slide is None:
 139                         default_slide = images.get('DefaultStreamImage')
 140                     if default_slide is not None:
 141                         default_slide = default_slide['ImageFilename']
 142                     if default_slide is not None:
 143                         fragments.append({
 144                             'path': default_slide,
 145                             'duration': slide['Time'] / 1000,
 146                         })
 147
 148             next_time = try_get(None, [
 149                 lambda _: Stream['Slides'][i + 1]['Time'],
 150                 lambda _: duration,
 151                 lambda _: slide['Time'],
 152             ], expected_type=(int, float))
 153
 154             fragments.append({
 155                 'path': fname_template.format(slide.get('Number', i + 1)),
 156                 'duration': (next_time - slide['Time']) / 1000
 157             })
 158
 159         return {
 160             'format_id': '%s-%u.slides' % (stream_id, snum),
 161             'ext': 'mhtml',
 162             'url': slide_base_url,
 163             'protocol': 'mhtml',
 164             'acodec': 'none',
 165             'vcodec': 'none',
 166             'format_note': 'Slides',
 167             'fragments': fragments,
 168             'fragment_base_url': slide_base_url,
 169         }
 170
 171     def _real_extract(self, url):
 172         url, data = unsmuggle_url(url, {})
 173         mobj = self._match_valid_url(url)
 174         resource_id = mobj.group('id')
 175         query = mobj.group('query')
 176
 177         webpage, urlh = self._download_webpage_handle(url, resource_id)  # XXX: add UrlReferrer?
 178         redirect_url = urlh.geturl()
 179
 180         # XXX: might have also extracted UrlReferrer and QueryString from the html
 181         service_path = compat_urlparse.urljoin(redirect_url, self._html_search_regex(
 182             r'<div[^>]+\bid=["\']ServicePath[^>]+>(.+?)</div>', webpage, resource_id,
 183             default='/Mediasite/PlayerService/PlayerService.svc/json'))
 184
 185         player_options = self._download_json(
 186             '%s/GetPlayerOptions' % service_path, resource_id,
 187             headers={
 188                 'Content-type': 'application/json; charset=utf-8',
 189                 'X-Requested-With': 'XMLHttpRequest',
 190             },
 191             data=json.dumps({
 192                 'getPlayerOptionsRequest': {
 193                     'ResourceId': resource_id,
 194                     'QueryString': query,
 195                     'UrlReferrer': data.get('UrlReferrer', ''),
 196                     'UseScreenReader': False,
 197                 }
 198             }).encode('utf-8'))['d']
 199
 200         presentation = player_options['Presentation']
 201         title = presentation['Title']
 202
 203         if presentation is None:
 204             raise ExtractorError(
 205                 'Mediasite says: %s' % player_options['PlayerPresentationStatusMessage'],
 206                 expected=True)
 207
 208         thumbnails = []
 209         formats = []
 210         for snum, Stream in enumerate(presentation['Streams']):
 211             stream_type = Stream.get('StreamType')
 212             if stream_type is None:
 213                 continue
 214
 215             video_urls = Stream.get('VideoUrls')
 216             if not isinstance(video_urls, list):
 217                 video_urls = []
 218
 219             stream_id = self._STREAM_TYPES.get(
 220                 stream_type, 'type%u' % stream_type)
 221
 222             stream_formats = []
 223             for unum, VideoUrl in enumerate(video_urls):
 224                 video_url = url_or_none(VideoUrl.get('Location'))
 225                 if not video_url:
 226                     continue
 227                 # XXX: if Stream.get('CanChangeScheme', False), switch scheme to HTTP/HTTPS
 228
 229                 media_type = VideoUrl.get('MediaType')
 230                 if media_type == 'SS':
 231                     stream_formats.extend(self._extract_ism_formats(
 232                         video_url, resource_id,
 233                         ism_id='%s-%u.%u' % (stream_id, snum, unum),
 234                         fatal=False))
 235                 elif media_type == 'Dash':
 236                     stream_formats.extend(self._extract_mpd_formats(
 237                         video_url, resource_id,
 238                         mpd_id='%s-%u.%u' % (stream_id, snum, unum),
 239                         fatal=False))
 240                 else:
 241                     stream_formats.append({
 242                         'format_id': '%s-%u.%u' % (stream_id, snum, unum),
 243                         'url': video_url,
 244                         'ext': mimetype2ext(VideoUrl.get('MimeType')),
 245                     })
 246
 247             if Stream.get('HasSlideContent', False):
 248                 images = player_options['PlayerLayoutOptions']['Images']
 249                 stream_formats.append(self.__extract_slides(
 250                     stream_id=stream_id,
 251                     snum=snum,
 252                     Stream=Stream,
 253                     duration=presentation.get('Duration'),
 254                     images=images,
 255                 ))
 256
 257             # disprefer 'secondary' streams
 258             if stream_type != 0:
 259                 for fmt in stream_formats:
 260                     fmt['quality'] = -10
 261
 262             thumbnail_url = Stream.get('ThumbnailUrl')
 263             if thumbnail_url:
 264                 thumbnails.append({
 265                     'id': '%s-%u' % (stream_id, snum),
 266                     'url': urljoin(redirect_url, thumbnail_url),
 267                     'preference': -1 if stream_type != 0 else 0,
 268                 })
 269             formats.extend(stream_formats)
 270
 271         self._sort_formats(formats)
 272
 273         # XXX: Presentation['Presenters']
 274         # XXX: Presentation['Transcript']
 275
 276         return {
 277             'id': resource_id,
 278             'title': title,
 279             'description': presentation.get('Description'),
 280             'duration': float_or_none(presentation.get('Duration'), 1000),
 281             'timestamp': float_or_none(presentation.get('UnixTime'), 1000),
 282             'formats': formats,
 283             'thumbnails': thumbnails,
 284         }
 285
 286
 287 class MediasiteCatalogIE(InfoExtractor):
 288     _VALID_URL = r'''(?xi)
 289                         (?P<url>https?://[^/]+/Mediasite)
 290                         /Catalog/Full/
 291                         (?P<catalog_id>{0})
 292                         (?:
 293                             /(?P<current_folder_id>{0})
 294                             /(?P<root_dynamic_folder_id>{0})
 295                         )?
 296                     '''.format(_ID_RE)
 297     _TESTS = [{
 298         'url': 'http://events7.mediasite.com/Mediasite/Catalog/Full/631f9e48530d454381549f955d08c75e21',
 299         'info_dict': {
 300             'id': '631f9e48530d454381549f955d08c75e21',
 301             'title': 'WCET Summit: Adaptive Learning in Higher Ed: Improving Outcomes Dynamically',
 302         },
 303         'playlist_count': 6,
 304         'expected_warnings': ['is not a supported codec'],
 305     }, {
 306         # with CurrentFolderId and RootDynamicFolderId
 307         'url': 'https://medaudio.medicine.iu.edu/Mediasite/Catalog/Full/9518c4a6c5cf4993b21cbd53e828a92521/97a9db45f7ab47428c77cd2ed74bb98f14/9518c4a6c5cf4993b21cbd53e828a92521',
 308         'info_dict': {
 309             'id': '9518c4a6c5cf4993b21cbd53e828a92521',
 310             'title': 'IUSM Family and Friends Sessions',
 311         },
 312         'playlist_count': 2,
 313     }, {
 314         'url': 'http://uipsyc.mediasite.com/mediasite/Catalog/Full/d5d79287c75243c58c50fef50174ec1b21',
 315         'only_matching': True,
 316     }, {
 317         # no AntiForgeryToken
 318         'url': 'https://live.libraries.psu.edu/Mediasite/Catalog/Full/8376d4b24dd1457ea3bfe4cf9163feda21',
 319         'only_matching': True,
 320     }, {
 321         'url': 'https://medaudio.medicine.iu.edu/Mediasite/Catalog/Full/9518c4a6c5cf4993b21cbd53e828a92521/97a9db45f7ab47428c77cd2ed74bb98f14/9518c4a6c5cf4993b21cbd53e828a92521',
 322         'only_matching': True,
 323     }, {
 324         # dashed id
 325         'url': 'http://events7.mediasite.com/Mediasite/Catalog/Full/631f9e48-530d-4543-8154-9f955d08c75e',
 326         'only_matching': True,
 327     }]
 328
 329     def _real_extract(self, url):
 330         mobj = self._match_valid_url(url)
 331         mediasite_url = mobj.group('url')
 332         catalog_id = mobj.group('catalog_id')
 333         current_folder_id = mobj.group('current_folder_id') or catalog_id
 334         root_dynamic_folder_id = mobj.group('root_dynamic_folder_id')
 335
 336         webpage = self._download_webpage(url, catalog_id)
 337
 338         # AntiForgeryToken is optional (e.g. [1])
 339         # 1. https://live.libraries.psu.edu/Mediasite/Catalog/Full/8376d4b24dd1457ea3bfe4cf9163feda21
 340         anti_forgery_token = self._search_regex(
 341             r'AntiForgeryToken\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1',
 342             webpage, 'anti forgery token', default=None, group='value')
 343         if anti_forgery_token:
 344             anti_forgery_header = self._search_regex(
 345                 r'AntiForgeryHeaderName\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1',
 346                 webpage, 'anti forgery header name',
 347                 default='X-SOFO-AntiForgeryHeader', group='value')
 348
 349         data = {
 350             'IsViewPage': True,
 351             'IsNewFolder': True,
 352             'AuthTicket': None,
 353             'CatalogId': catalog_id,
 354             'CurrentFolderId': current_folder_id,
 355             'RootDynamicFolderId': root_dynamic_folder_id,
 356             'ItemsPerPage': 1000,
 357             'PageIndex': 0,
 358             'PermissionMask': 'Execute',
 359             'CatalogSearchType': 'SearchInFolder',
 360             'SortBy': 'Date',
 361             'SortDirection': 'Descending',
 362             'StartDate': None,
 363             'EndDate': None,
 364             'StatusFilterList': None,
 365             'PreviewKey': None,
 366             'Tags': [],
 367         }
 368
 369         headers = {
 370             'Content-Type': 'application/json; charset=UTF-8',
 371             'Referer': url,
 372             'X-Requested-With': 'XMLHttpRequest',
 373         }
 374         if anti_forgery_token:
 375             headers[anti_forgery_header] = anti_forgery_token
 376
 377         catalog = self._download_json(
 378             '%s/Catalog/Data/GetPresentationsForFolder' % mediasite_url,
 379             catalog_id, data=json.dumps(data).encode(), headers=headers)
 380
 381         entries = []
 382         for video in catalog['PresentationDetailsList']:
 383             if not isinstance(video, dict):
 384                 continue
 385             video_id = str_or_none(video.get('Id'))
 386             if not video_id:
 387                 continue
 388             entries.append(self.url_result(
 389                 '%s/Play/%s' % (mediasite_url, video_id),
 390                 ie=MediasiteIE.ie_key(), video_id=video_id))
 391
 392         title = try_get(
 393             catalog, lambda x: x['CurrentFolder']['Name'], compat_str)
 394
 395         return self.playlist_result(entries, catalog_id, title,)
 396
 397
 398 class MediasiteNamedCatalogIE(InfoExtractor):
 399     _VALID_URL = r'(?xi)(?P<url>https?://[^/]+/Mediasite)/Catalog/catalogs/(?P<catalog_name>[^/?#&]+)'
 400     _TESTS = [{
 401         'url': 'https://msite.misis.ru/Mediasite/Catalog/catalogs/2016-industrial-management-skriabin-o-o',
 402         'only_matching': True,
 403     }]
 404
 405     def _real_extract(self, url):
 406         mobj = self._match_valid_url(url)
 407         mediasite_url = mobj.group('url')
 408         catalog_name = mobj.group('catalog_name')
 409
 410         webpage = self._download_webpage(url, catalog_name)
 411
 412         catalog_id = self._search_regex(
 413             r'CatalogId\s*:\s*["\'](%s)' % _ID_RE, webpage, 'catalog id')
 414
 415         return self.url_result(
 416             '%s/Catalog/Full/%s' % (mediasite_url, catalog_id),
 417             ie=MediasiteCatalogIE.ie_key(), video_id=catalog_id)