_ Git - youtube-dl/blob - youtube_dl/extractor/libraryofcongress.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5
   6 from ..utils import (
   7     determine_ext,
   8     float_or_none,
   9     int_or_none,
  10 )
  11
  12
  13 class LibraryOfCongressIE(InfoExtractor):
  14     IE_NAME = 'loc'
  15     IE_DESC = 'Library of Congress'
  16     _VALID_URL = r'https?://(?:www\.)?loc\.gov/(?:item/|today/cyberlc/feature_wdesc\.php\?.*\brec=)(?P<id>[0-9]+)'
  17     _TESTS = [{
  18         'url': 'http://loc.gov/item/90716351/',
  19         'md5': '353917ff7f0255aa6d4b80a034833de8',
  20         'info_dict': {
  21             'id': '90716351',
  22             'ext': 'mp4',
  23             'title': "Pa's trip to Mars",
  24             'thumbnail': 're:^https?://.*\.jpg$',
  25             'duration': 0,
  26             'view_count': int,
  27         },
  28     }, {
  29         'url': 'https://www.loc.gov/today/cyberlc/feature_wdesc.php?rec=5578',
  30         'only_matching': True,
  31     }]
  32
  33     def _real_extract(self, url):
  34         video_id = self._match_id(url)
  35         webpage = self._download_webpage(url, video_id)
  36
  37         media_id = self._search_regex(
  38             (r'id=(["\'])media-player-(?P<id>.+?)\1',
  39              r'<video[^>]+id=(["\'])uuid-(?P<id>.+?)\1',
  40              r'<video[^>]+data-uuid=(["\'])(?P<id>.+?)\1',
  41              r'mediaObjectId\s*:\s*(["\'])(?P<id>.+?)\1'),
  42             webpage, 'media id', group='id')
  43
  44         data = self._download_json(
  45             'https://media.loc.gov/services/v1/media?id=%s&context=json' % media_id,
  46             video_id)['mediaObject']
  47
  48         derivative = data['derivatives'][0]
  49         media_url = derivative['derivativeUrl']
  50
  51         # Following algorithm was extracted from setAVSource js function
  52         # found in webpage
  53         media_url = media_url.replace('rtmp', 'https')
  54
  55         is_video = data.get('mediaType', 'v').lower() == 'v'
  56         ext = determine_ext(media_url)
  57         if ext not in ('mp4', 'mp3'):
  58             media_url += '.mp4' if is_video else '.mp3'
  59
  60         if 'vod/mp4:' in media_url:
  61             formats = [{
  62                 'url': media_url.replace('vod/mp4:', 'hls-vod/media/') + '.m3u8',
  63                 'format_id': 'hls',
  64                 'ext': 'mp4',
  65                 'protocol': 'm3u8_native',
  66             }]
  67         elif 'vod/mp3:' in media_url:
  68             formats = [{
  69                 'url': media_url.replace('vod/mp3:', ''),
  70                 'vcodec': 'none',
  71             }]
  72
  73         self._sort_formats(formats)
  74
  75         title = derivative.get('shortName') or data.get('shortName') or self._og_search_title(webpage)
  76         duration = float_or_none(data.get('duration'))
  77         view_count = int_or_none(data.get('viewCount'))
  78
  79         return {
  80             'id': video_id,
  81             'title': title,
  82             'thumbnail': self._og_search_thumbnail(webpage, default=None),
  83             'duration': duration,
  84             'view_count': view_count,
  85             'formats': formats,
  86         }