_ 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/(?P<id>[0-9]+)'
  17     _TEST = {
  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
  30     def _real_extract(self, url):
  31         video_id = self._match_id(url)
  32         webpage = self._download_webpage(url, video_id)
  33
  34         media_id = self._search_regex(
  35             (r'id=(["\'])media-player-(?P<id>.+?)\1',
  36              r'<video[^>]+id=(["\'])uuid-(?P<id>.+?)\1',
  37              r'<video[^>]+data-uuid=(["\'])(?P<id>.+?)\1'),
  38             webpage, 'media id', group='id')
  39
  40         data = self._parse_json(
  41             self._download_webpage(
  42                 'https://media.loc.gov/services/v1/media?id=%s&context=json' % media_id,
  43                 video_id),
  44             video_id)['mediaObject']
  45
  46         derivative = data['derivatives'][0]
  47         media_url = derivative['derivativeUrl']
  48
  49         # Following algorithm was extracted from setAVSource js function
  50         # found in webpage
  51         media_url = media_url.replace('rtmp', 'https')
  52
  53         is_video = data.get('mediaType', 'v').lower() == 'v'
  54         ext = determine_ext(media_url)
  55         if ext not in ('mp4', 'mp3'):
  56             media_url += '.mp4' if is_video else '.mp3'
  57
  58         if 'vod/mp4:' in media_url:
  59             formats = [{
  60                 'url': media_url.replace('vod/mp4:', 'hls-vod/media/') + '.m3u8',
  61                 'format_id': 'hls',
  62                 'ext': 'mp4',
  63                 'protocol': 'm3u8_native',
  64             }]
  65         elif 'vod/mp3:' in media_url:
  66             formats = [{
  67                 'url': media_url.replace('vod/mp3:', ''),
  68                 'vcodec': 'none',
  69             }]
  70
  71         self._sort_formats(formats)
  72
  73         title = derivative.get('shortName') or data.get('shortName') or self._og_search_title(webpage)
  74         duration = float_or_none(data.get('duration'))
  75         view_count = int_or_none(data.get('viewCount'))
  76
  77         return {
  78             'id': video_id,
  79             'title': title,
  80             'thumbnail': self._og_search_thumbnail(webpage),
  81             'duration': duration,
  82             'view_count': view_count,
  83             'formats': formats,
  84         }