_ Git - youtube-dl/blob - 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}'
  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
 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     @staticmethod
 113     def _extract_urls(webpage):
 114         return [
 115             unescapeHTML(mobj.group('url'))
 116             for mobj in re.finditer(
 117                 r'(?xi)<iframe\b[^>]+\bsrc=(["\'])(?P<url>(?:(?:https?:)?//[^/]+)?/Mediasite/Play/%s(?:\?.*?)?)\1' % _ID_RE,
 118                 webpage)]
 119
 120     def _real_extract(self, url):
 121         url, data = unsmuggle_url(url, {})
 122         mobj = re.match(self._VALID_URL, url)
 123         resource_id = mobj.group('id')
 124         query = mobj.group('query')
 125
 126         webpage, urlh = self._download_webpage_handle(url, resource_id)  # XXX: add UrlReferrer?
 127         redirect_url = compat_str(urlh.geturl())
 128
 129         # XXX: might have also extracted UrlReferrer and QueryString from the html
 130         service_path = compat_urlparse.urljoin(redirect_url, self._html_search_regex(
 131             r'<div[^>]+\bid=["\']ServicePath[^>]+>(.+?)</div>', webpage, resource_id,
 132             default='/Mediasite/PlayerService/PlayerService.svc/json'))
 133
 134         player_options = self._download_json(
 135             '%s/GetPlayerOptions' % service_path, resource_id,
 136             headers={
 137                 'Content-type': 'application/json; charset=utf-8',
 138                 'X-Requested-With': 'XMLHttpRequest',
 139             },
 140             data=json.dumps({
 141                 'getPlayerOptionsRequest': {
 142                     'ResourceId': resource_id,
 143                     'QueryString': query,
 144                     'UrlReferrer': data.get('UrlReferrer', ''),
 145                     'UseScreenReader': False,
 146                 }
 147             }).encode('utf-8'))['d']
 148
 149         presentation = player_options['Presentation']
 150         title = presentation['Title']
 151
 152         if presentation is None:
 153             raise ExtractorError(
 154                 'Mediasite says: %s' % player_options['PlayerPresentationStatusMessage'],
 155                 expected=True)
 156
 157         thumbnails = []
 158         formats = []
 159         for snum, Stream in enumerate(presentation['Streams']):
 160             stream_type = Stream.get('StreamType')
 161             if stream_type is None:
 162                 continue
 163
 164             video_urls = Stream.get('VideoUrls')
 165             if not isinstance(video_urls, list):
 166                 video_urls = []
 167
 168             stream_id = self._STREAM_TYPES.get(
 169                 stream_type, 'type%u' % stream_type)
 170
 171             stream_formats = []
 172             for unum, VideoUrl in enumerate(video_urls):
 173                 video_url = url_or_none(VideoUrl.get('Location'))
 174                 if not video_url:
 175                     continue
 176                 # XXX: if Stream.get('CanChangeScheme', False), switch scheme to HTTP/HTTPS
 177
 178                 media_type = VideoUrl.get('MediaType')
 179                 if media_type == 'SS':
 180                     stream_formats.extend(self._extract_ism_formats(
 181                         video_url, resource_id,
 182                         ism_id='%s-%u.%u' % (stream_id, snum, unum),
 183                         fatal=False))
 184                 elif media_type == 'Dash':
 185                     stream_formats.extend(self._extract_mpd_formats(
 186                         video_url, resource_id,
 187                         mpd_id='%s-%u.%u' % (stream_id, snum, unum),
 188                         fatal=False))
 189                 else:
 190                     stream_formats.append({
 191                         'format_id': '%s-%u.%u' % (stream_id, snum, unum),
 192                         'url': video_url,
 193                         'ext': mimetype2ext(VideoUrl.get('MimeType')),
 194                     })
 195
 196             # TODO: if Stream['HasSlideContent']:
 197             # synthesise an MJPEG video stream '%s-%u.slides' % (stream_type, snum)
 198             # from Stream['Slides']
 199             # this will require writing a custom downloader...
 200
 201             # disprefer 'secondary' streams
 202             if stream_type != 0:
 203                 for fmt in stream_formats:
 204                     fmt['preference'] = -1
 205
 206             thumbnail_url = Stream.get('ThumbnailUrl')
 207             if thumbnail_url:
 208                 thumbnails.append({
 209                     'id': '%s-%u' % (stream_id, snum),
 210                     'url': urljoin(redirect_url, thumbnail_url),
 211                     'preference': -1 if stream_type != 0 else 0,
 212                 })
 213             formats.extend(stream_formats)
 214
 215         self._sort_formats(formats)
 216
 217         # XXX: Presentation['Presenters']
 218         # XXX: Presentation['Transcript']
 219
 220         return {
 221             'id': resource_id,
 222             'title': title,
 223             'description': presentation.get('Description'),
 224             'duration': float_or_none(presentation.get('Duration'), 1000),
 225             'timestamp': float_or_none(presentation.get('UnixTime'), 1000),
 226             'formats': formats,
 227             'thumbnails': thumbnails,
 228         }
 229
 230
 231 class MediasiteCatalogIE(InfoExtractor):
 232     _VALID_URL = r'''(?xi)
 233                         (?P<url>https?://[^/]+/Mediasite)
 234                         /Catalog/Full/
 235                         (?P<catalog_id>{0})
 236                         (?:
 237                             /(?P<current_folder_id>{0})
 238                             /(?P<root_dynamic_folder_id>{0})
 239                         )?
 240                     '''.format(_ID_RE)
 241     _TESTS = [{
 242         'url': 'http://events7.mediasite.com/Mediasite/Catalog/Full/631f9e48530d454381549f955d08c75e21',
 243         'info_dict': {
 244             'id': '631f9e48530d454381549f955d08c75e21',
 245             'title': 'WCET Summit: Adaptive Learning in Higher Ed: Improving Outcomes Dynamically',
 246         },
 247         'playlist_count': 6,
 248         'expected_warnings': ['is not a supported codec'],
 249     }, {
 250         # with CurrentFolderId and RootDynamicFolderId
 251         'url': 'https://medaudio.medicine.iu.edu/Mediasite/Catalog/Full/9518c4a6c5cf4993b21cbd53e828a92521/97a9db45f7ab47428c77cd2ed74bb98f14/9518c4a6c5cf4993b21cbd53e828a92521',
 252         'info_dict': {
 253             'id': '9518c4a6c5cf4993b21cbd53e828a92521',
 254             'title': 'IUSM Family and Friends Sessions',
 255         },
 256         'playlist_count': 2,
 257     }, {
 258         'url': 'http://uipsyc.mediasite.com/mediasite/Catalog/Full/d5d79287c75243c58c50fef50174ec1b21',
 259         'only_matching': True,
 260     }, {
 261         # no AntiForgeryToken
 262         'url': 'https://live.libraries.psu.edu/Mediasite/Catalog/Full/8376d4b24dd1457ea3bfe4cf9163feda21',
 263         'only_matching': True,
 264     }, {
 265         'url': 'https://medaudio.medicine.iu.edu/Mediasite/Catalog/Full/9518c4a6c5cf4993b21cbd53e828a92521/97a9db45f7ab47428c77cd2ed74bb98f14/9518c4a6c5cf4993b21cbd53e828a92521',
 266         'only_matching': True,
 267     }]
 268
 269     def _real_extract(self, url):
 270         mobj = re.match(self._VALID_URL, url)
 271         mediasite_url = mobj.group('url')
 272         catalog_id = mobj.group('catalog_id')
 273         current_folder_id = mobj.group('current_folder_id') or catalog_id
 274         root_dynamic_folder_id = mobj.group('root_dynamic_folder_id')
 275
 276         webpage = self._download_webpage(url, catalog_id)
 277
 278         # AntiForgeryToken is optional (e.g. [1])
 279         # 1. https://live.libraries.psu.edu/Mediasite/Catalog/Full/8376d4b24dd1457ea3bfe4cf9163feda21
 280         anti_forgery_token = self._search_regex(
 281             r'AntiForgeryToken\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1',
 282             webpage, 'anti forgery token', default=None, group='value')
 283         if anti_forgery_token:
 284             anti_forgery_header = self._search_regex(
 285                 r'AntiForgeryHeaderName\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1',
 286                 webpage, 'anti forgery header name',
 287                 default='X-SOFO-AntiForgeryHeader', group='value')
 288
 289         data = {
 290             'IsViewPage': True,
 291             'IsNewFolder': True,
 292             'AuthTicket': None,
 293             'CatalogId': catalog_id,
 294             'CurrentFolderId': current_folder_id,
 295             'RootDynamicFolderId': root_dynamic_folder_id,
 296             'ItemsPerPage': 1000,
 297             'PageIndex': 0,
 298             'PermissionMask': 'Execute',
 299             'CatalogSearchType': 'SearchInFolder',
 300             'SortBy': 'Date',
 301             'SortDirection': 'Descending',
 302             'StartDate': None,
 303             'EndDate': None,
 304             'StatusFilterList': None,
 305             'PreviewKey': None,
 306             'Tags': [],
 307         }
 308
 309         headers = {
 310             'Content-Type': 'application/json; charset=UTF-8',
 311             'Referer': url,
 312             'X-Requested-With': 'XMLHttpRequest',
 313         }
 314         if anti_forgery_token:
 315             headers[anti_forgery_header] = anti_forgery_token
 316
 317         catalog = self._download_json(
 318             '%s/Catalog/Data/GetPresentationsForFolder' % mediasite_url,
 319             catalog_id, data=json.dumps(data).encode(), headers=headers)
 320
 321         entries = []
 322         for video in catalog['PresentationDetailsList']:
 323             if not isinstance(video, dict):
 324                 continue
 325             video_id = str_or_none(video.get('Id'))
 326             if not video_id:
 327                 continue
 328             entries.append(self.url_result(
 329                 '%s/Play/%s' % (mediasite_url, video_id),
 330                 ie=MediasiteIE.ie_key(), video_id=video_id))
 331
 332         title = try_get(
 333             catalog, lambda x: x['CurrentFolder']['Name'], compat_str)
 334
 335         return self.playlist_result(entries, catalog_id, title,)