yt_dlp/extractor/rice.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..compat import compat_parse_qs
   5 from ..utils import (
   6     xpath_text,
   7     xpath_element,
   8     int_or_none,
   9     parse_iso8601,
  10     ExtractorError,
  11 )
  12
  13
  14 class RICEIE(InfoExtractor):
  15     _VALID_URL = r'https?://mediahub\.rice\.edu/app/[Pp]ortal/video\.aspx\?(?P<query>.+)'
  16     _TEST = {
  17         'url': 'https://mediahub.rice.edu/app/Portal/video.aspx?PortalID=25ffd62c-3d01-4b29-8c70-7c94270efb3e&DestinationID=66bc9434-03bd-4725-b47e-c659d8d809db&ContentID=YEWIvbhb40aqdjMD1ALSqw',
  18         'md5': '9b83b4a2eead4912dc3b7fac7c449b6a',
  19         'info_dict': {
  20             'id': 'YEWIvbhb40aqdjMD1ALSqw',
  21             'ext': 'mp4',
  22             'title': 'Active Learning in Archeology',
  23             'upload_date': '20140616',
  24             'timestamp': 1402926346,
  25         }
  26     }
  27     _NS = 'http://schemas.datacontract.org/2004/07/ensembleVideo.Data.Service.Contracts.Models.Player.Config'
  28
  29     def _real_extract(self, url):
  30         qs = compat_parse_qs(self._match_valid_url(url).group('query'))
  31         if not qs.get('PortalID') or not qs.get('DestinationID') or not qs.get('ContentID'):
  32             raise ExtractorError('Invalid URL', expected=True)
  33
  34         portal_id = qs['PortalID'][0]
  35         playlist_id = qs['DestinationID'][0]
  36         content_id = qs['ContentID'][0]
  37
  38         content_data = self._download_xml('https://mediahub.rice.edu/api/portal/GetContentTitle', content_id, query={
  39             'portalId': portal_id,
  40             'playlistId': playlist_id,
  41             'contentId': content_id
  42         })
  43         metadata = xpath_element(content_data, './/metaData', fatal=True)
  44         title = xpath_text(metadata, 'primaryTitle', fatal=True)
  45         encodings = xpath_element(content_data, './/encodings', fatal=True)
  46         player_data = self._download_xml('https://mediahub.rice.edu/api/player/GetPlayerConfig', content_id, query={
  47             'temporaryLinkId': xpath_text(encodings, 'temporaryLinkId', fatal=True),
  48             'contentId': content_id,
  49         })
  50
  51         common_fmt = {}
  52         dimensions = xpath_text(encodings, 'dimensions')
  53         if dimensions:
  54             wh = dimensions.split('x')
  55             if len(wh) == 2:
  56                 common_fmt.update({
  57                     'width': int_or_none(wh[0]),
  58                     'height': int_or_none(wh[1]),
  59                 })
  60
  61         formats = []
  62         rtsp_path = xpath_text(player_data, self._xpath_ns('RtspPath', self._NS))
  63         if rtsp_path:
  64             fmt = {
  65                 'url': rtsp_path,
  66                 'format_id': 'rtsp',
  67             }
  68             fmt.update(common_fmt)
  69             formats.append(fmt)
  70         for source in player_data.findall(self._xpath_ns('.//Source', self._NS)):
  71             video_url = xpath_text(source, self._xpath_ns('File', self._NS))
  72             if not video_url:
  73                 continue
  74             if '.m3u8' in video_url:
  75                 formats.extend(self._extract_m3u8_formats(video_url, content_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
  76             else:
  77                 fmt = {
  78                     'url': video_url,
  79                     'format_id': video_url.split(':')[0],
  80                 }
  81                 fmt.update(common_fmt)
  82                 rtmp = re.search(r'^(?P<url>rtmp://[^/]+/(?P<app>.+))/(?P<playpath>mp4:.+)$', video_url)
  83                 if rtmp:
  84                     fmt.update({
  85                         'url': rtmp.group('url'),
  86                         'play_path': rtmp.group('playpath'),
  87                         'app': rtmp.group('app'),
  88                         'ext': 'flv',
  89                     })
  90                 formats.append(fmt)
  91
  92         thumbnails = []
  93         for content_asset in content_data.findall('.//contentAssets'):
  94             asset_type = xpath_text(content_asset, 'type')
  95             if asset_type == 'image':
  96                 image_url = xpath_text(content_asset, 'httpPath')
  97                 if not image_url:
  98                     continue
  99                 thumbnails.append({
 100                     'id': xpath_text(content_asset, 'ID'),
 101                     'url': image_url,
 102                 })
 103
 104         return {
 105             'id': content_id,
 106             'title': title,
 107             'description': xpath_text(metadata, 'abstract'),
 108             'duration': int_or_none(xpath_text(metadata, 'duration')),
 109             'timestamp': parse_iso8601(xpath_text(metadata, 'dateUpdated')),
 110             'thumbnails': thumbnails,
 111             'formats': formats,
 112         }