Raphaël G. Git Repositories - youtubedl/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     unescapeHTML,
  17     unsmuggle_url,
  18     url_or_none,
  19     urljoin,
  20 )
  21
  22
  23 class MediasiteIE(InfoExtractor):
  24     _VALID_URL = r'(?xi)https?://[^/]+/Mediasite/(?:Play|Showcase/(?:default|livebroadcast)/Presentation)/(?P<id>[0-9a-f]{32,34})(?P<query>\?[^#]+|)'
  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
  98     # look in Mediasite.Core.js (Mediasite.ContentStreamType[*])
  99     _STREAM_TYPES = {
 100         0: 'video1',  # the main video
 101         2: 'slide',
 102         3: 'presentation',
 103         4: 'video2',  # screencast?
 104         5: 'video3',
 105     }
 106
 107     @staticmethod
 108     def _extract_urls(webpage):
 109         return [
 110             unescapeHTML(mobj.group('url'))
 111             for mobj in re.finditer(
 112                 r'(?xi)<iframe\b[^>]+\bsrc=(["\'])(?P<url>(?:(?:https?:)?//[^/]+)?/Mediasite/Play/[0-9a-f]{32,34}(?:\?.*?)?)\1',
 113                 webpage)]
 114
 115     def _real_extract(self, url):
 116         url, data = unsmuggle_url(url, {})
 117         mobj = re.match(self._VALID_URL, url)
 118         resource_id = mobj.group('id')
 119         query = mobj.group('query')
 120
 121         webpage, urlh = self._download_webpage_handle(url, resource_id)  # XXX: add UrlReferrer?
 122         redirect_url = compat_str(urlh.geturl())
 123
 124         # XXX: might have also extracted UrlReferrer and QueryString from the html
 125         service_path = compat_urlparse.urljoin(redirect_url, self._html_search_regex(
 126             r'<div[^>]+\bid=["\']ServicePath[^>]+>(.+?)</div>', webpage, resource_id,
 127             default='/Mediasite/PlayerService/PlayerService.svc/json'))
 128
 129         player_options = self._download_json(
 130             '%s/GetPlayerOptions' % service_path, resource_id,
 131             headers={
 132                 'Content-type': 'application/json; charset=utf-8',
 133                 'X-Requested-With': 'XMLHttpRequest',
 134             },
 135             data=json.dumps({
 136                 'getPlayerOptionsRequest': {
 137                     'ResourceId': resource_id,
 138                     'QueryString': query,
 139                     'UrlReferrer': data.get('UrlReferrer', ''),
 140                     'UseScreenReader': False,
 141                 }
 142             }).encode('utf-8'))['d']
 143
 144         presentation = player_options['Presentation']
 145         title = presentation['Title']
 146
 147         if presentation is None:
 148             raise ExtractorError(
 149                 'Mediasite says: %s' % player_options['PlayerPresentationStatusMessage'],
 150                 expected=True)
 151
 152         thumbnails = []
 153         formats = []
 154         for snum, Stream in enumerate(presentation['Streams']):
 155             stream_type = Stream.get('StreamType')
 156             if stream_type is None:
 157                 continue
 158
 159             video_urls = Stream.get('VideoUrls')
 160             if not isinstance(video_urls, list):
 161                 video_urls = []
 162
 163             stream_id = self._STREAM_TYPES.get(
 164                 stream_type, 'type%u' % stream_type)
 165
 166             stream_formats = []
 167             for unum, VideoUrl in enumerate(video_urls):
 168                 video_url = url_or_none(VideoUrl.get('Location'))
 169                 if not video_url:
 170                     continue
 171                 # XXX: if Stream.get('CanChangeScheme', False), switch scheme to HTTP/HTTPS
 172
 173                 media_type = VideoUrl.get('MediaType')
 174                 if media_type == 'SS':
 175                     stream_formats.extend(self._extract_ism_formats(
 176                         video_url, resource_id,
 177                         ism_id='%s-%u.%u' % (stream_id, snum, unum),
 178                         fatal=False))
 179                 elif media_type == 'Dash':
 180                     stream_formats.extend(self._extract_mpd_formats(
 181                         video_url, resource_id,
 182                         mpd_id='%s-%u.%u' % (stream_id, snum, unum),
 183                         fatal=False))
 184                 else:
 185                     stream_formats.append({
 186                         'format_id': '%s-%u.%u' % (stream_id, snum, unum),
 187                         'url': video_url,
 188                         'ext': mimetype2ext(VideoUrl.get('MimeType')),
 189                     })
 190
 191             # TODO: if Stream['HasSlideContent']:
 192             # synthesise an MJPEG video stream '%s-%u.slides' % (stream_type, snum)
 193             # from Stream['Slides']
 194             # this will require writing a custom downloader...
 195
 196             # disprefer 'secondary' streams
 197             if stream_type != 0:
 198                 for fmt in stream_formats:
 199                     fmt['preference'] = -1
 200
 201             thumbnail_url = Stream.get('ThumbnailUrl')
 202             if thumbnail_url:
 203                 thumbnails.append({
 204                     'id': '%s-%u' % (stream_id, snum),
 205                     'url': urljoin(redirect_url, thumbnail_url),
 206                     'preference': -1 if stream_type != 0 else 0,
 207                 })
 208             formats.extend(stream_formats)
 209
 210         self._sort_formats(formats)
 211
 212         # XXX: Presentation['Presenters']
 213         # XXX: Presentation['Transcript']
 214
 215         return {
 216             'id': resource_id,
 217             'title': title,
 218             'description': presentation.get('Description'),
 219             'duration': float_or_none(presentation.get('Duration'), 1000),
 220             'timestamp': float_or_none(presentation.get('UnixTime'), 1000),
 221             'formats': formats,
 222             'thumbnails': thumbnails,
 223         }