youtube_dl/extractor/npr.py

   1 from __future__ import unicode_literals
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     int_or_none,
   6     qualities,
   7 )
   8
   9
  10 class NprIE(InfoExtractor):
  11     _VALID_URL = r'https?://(?:www\.)?npr\.org/(?:sections/[^/]+/)?\d{4}/\d{2}/\d{2}/(?P<id>\d+)'
  12     _TESTS = [{
  13         'url': 'https://www.npr.org/sections/allsongs/2015/10/21/449974205/new-music-from-beach-house-chairlift-cmj-discoveries-and-more',
  14         'info_dict': {
  15             'id': '449974205',
  16             'title': 'New Music From Beach House, Chairlift, CMJ Discoveries And More'
  17         },
  18         'playlist_count': 7,
  19     }, {
  20         'url': 'https://www.npr.org/sections/deceptivecadence/2015/10/09/446928052/music-from-the-shadows-ancient-armenian-hymns-and-piano-jazz',
  21         'info_dict': {
  22             'id': '446928052',
  23             'title': "Songs We Love: Tigran Hamasyan, 'Your Mercy is Boundless'"
  24         },
  25         'playlist': [{
  26             'md5': '12fa60cb2d3ed932f53609d4aeceabf1',
  27             'info_dict': {
  28                 'id': '446929930',
  29                 'ext': 'mp3',
  30                 'title': 'Your Mercy is Boundless (Bazum en Qo gtutyunqd)',
  31                 'duration': 402,
  32             },
  33         }],
  34     }, {
  35         # mutlimedia, not media title
  36         'url': 'https://www.npr.org/2017/06/19/533198237/tigers-jaw-tiny-desk-concert',
  37         'info_dict': {
  38             'id': '533198237',
  39             'title': 'Tigers Jaw: Tiny Desk Concert',
  40         },
  41         'playlist': [{
  42             'md5': '12fa60cb2d3ed932f53609d4aeceabf1',
  43             'info_dict': {
  44                 'id': '533201718',
  45                 'ext': 'mp4',
  46                 'title': 'Tigers Jaw: Tiny Desk Concert',
  47                 'duration': 402,
  48             },
  49         }],
  50         'expected_warnings': ['Failed to download m3u8 information'],
  51     }]
  52
  53     def _real_extract(self, url):
  54         playlist_id = self._match_id(url)
  55
  56         story = self._download_json(
  57             'http://api.npr.org/query', playlist_id, query={
  58                 'id': playlist_id,
  59                 'fields': 'audio,multimedia,title',
  60                 'format': 'json',
  61                 'apiKey': 'MDAzMzQ2MjAyMDEyMzk4MTU1MDg3ZmM3MQ010',
  62             })['list']['story'][0]
  63         playlist_title = story.get('title', {}).get('$text')
  64
  65         KNOWN_FORMATS = ('threegp', 'm3u8', 'smil', 'mp4', 'mp3')
  66         quality = qualities(KNOWN_FORMATS)
  67
  68         entries = []
  69         for media in story.get('audio', []) + story.get('multimedia', []):
  70             media_id = media['id']
  71
  72             formats = []
  73             for format_id, formats_entry in media.get('format', {}).items():
  74                 if not formats_entry:
  75                     continue
  76                 if isinstance(formats_entry, list):
  77                     formats_entry = formats_entry[0]
  78                 format_url = formats_entry.get('$text')
  79                 if not format_url:
  80                     continue
  81                 if format_id in KNOWN_FORMATS:
  82                     if format_id == 'm3u8':
  83                         formats.extend(self._extract_m3u8_formats(
  84                             format_url, media_id, 'mp4', 'm3u8_native',
  85                             m3u8_id='hls', fatal=False))
  86                     elif format_id == 'smil':
  87                         smil_formats = self._extract_smil_formats(
  88                             format_url, media_id, transform_source=lambda s: s.replace(
  89                                 'rtmp://flash.npr.org/ondemand/', 'https://ondemand.npr.org/'))
  90                         self._check_formats(smil_formats, media_id)
  91                         formats.extend(smil_formats)
  92                     else:
  93                         formats.append({
  94                             'url': format_url,
  95                             'format_id': format_id,
  96                             'quality': quality(format_id),
  97                         })
  98             self._sort_formats(formats)
  99
 100             entries.append({
 101                 'id': media_id,
 102                 'title': media.get('title', {}).get('$text') or playlist_title,
 103                 'thumbnail': media.get('altImageUrl', {}).get('$text'),
 104                 'duration': int_or_none(media.get('duration', {}).get('$text')),
 105                 'formats': formats,
 106             })
 107
 108         return self.playlist_result(entries, playlist_id, playlist_title)