youtube_dl/extractor/npr.py

   1 from __future__ import unicode_literals
   2
   3 from .common import InfoExtractor
   4 from ..compat import compat_urllib_parse_urlencode
   5 from ..utils import (
   6     int_or_none,
   7     qualities,
   8 )
   9
  10
  11 class NprIE(InfoExtractor):
  12     _VALID_URL = r'https?://(?:www\.)?npr\.org/player/v2/mediaPlayer\.html\?.*\bid=(?P<id>\d+)'
  13     _TESTS = [{
  14         'url': 'http://www.npr.org/player/v2/mediaPlayer.html?id=449974205',
  15         'info_dict': {
  16             'id': '449974205',
  17             'title': 'New Music From Beach House, Chairlift, CMJ Discoveries And More'
  18         },
  19         'playlist_count': 7,
  20     }, {
  21         'url': 'http://www.npr.org/player/v2/mediaPlayer.html?action=1&t=1&islist=false&id=446928052&m=446929930&live=1',
  22         'info_dict': {
  23             'id': '446928052',
  24             'title': "Songs We Love: Tigran Hamasyan, 'Your Mercy is Boundless'"
  25         },
  26         'playlist': [{
  27             'md5': '12fa60cb2d3ed932f53609d4aeceabf1',
  28             'info_dict': {
  29                 'id': '446929930',
  30                 'ext': 'mp3',
  31                 'title': 'Your Mercy is Boundless (Bazum en Qo gtutyunqd)',
  32                 'duration': 402,
  33             },
  34         }],
  35     }]
  36
  37     def _real_extract(self, url):
  38         playlist_id = self._match_id(url)
  39
  40         config = self._download_json(
  41             'http://api.npr.org/query?%s' % compat_urllib_parse_urlencode({
  42                 'id': playlist_id,
  43                 'fields': 'titles,audio,show',
  44                 'format': 'json',
  45                 'apiKey': 'MDAzMzQ2MjAyMDEyMzk4MTU1MDg3ZmM3MQ010',
  46             }), playlist_id)
  47
  48         story = config['list']['story'][0]
  49
  50         KNOWN_FORMATS = ('threegp', 'mp4', 'mp3')
  51         quality = qualities(KNOWN_FORMATS)
  52
  53         entries = []
  54         for audio in story.get('audio', []):
  55             title = audio.get('title', {}).get('$text')
  56             duration = int_or_none(audio.get('duration', {}).get('$text'))
  57             formats = []
  58             for format_id, formats_entry in audio.get('format', {}).items():
  59                 if not formats_entry:
  60                     continue
  61                 if isinstance(formats_entry, list):
  62                     formats_entry = formats_entry[0]
  63                 format_url = formats_entry.get('$text')
  64                 if not format_url:
  65                     continue
  66                 if format_id in KNOWN_FORMATS:
  67                     formats.append({
  68                         'url': format_url,
  69                         'format_id': format_id,
  70                         'ext': formats_entry.get('type'),
  71                         'quality': quality(format_id),
  72                     })
  73             self._sort_formats(formats)
  74             entries.append({
  75                 'id': audio['id'],
  76                 'title': title,
  77                 'duration': duration,
  78                 'formats': formats,
  79             })
  80
  81         playlist_title = story.get('title', {}).get('$text')
  82         return self.playlist_result(entries, playlist_id, playlist_title)