youtube_dl/extractor/nhk.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import urljoin
   7
   8
   9 class NhkBaseIE(InfoExtractor):
  10     _API_URL_TEMPLATE = 'https://api.nhk.or.jp/nhkworld/%sod%slist/v7a/%s/%s/%s/all%s.json'
  11     _BASE_URL_REGEX = r'https?://www3\.nhk\.or\.jp/nhkworld/(?P<lang>[a-z]{2})/ondemand'
  12     _TYPE_REGEX = r'/(?P<type>video|audio)/'
  13
  14     def _call_api(self, m_id, lang, is_video, is_episode, is_clip):
  15         return self._download_json(
  16             self._API_URL_TEMPLATE % (
  17                 'v' if is_video else 'r',
  18                 'clip' if is_clip else 'esd',
  19                 'episode' if is_episode else 'program',
  20                 m_id, lang, '/all' if is_video else ''),
  21             m_id, query={'apikey': 'EJfK8jdS57GqlupFgAfAAwr573q01y6k'})['data']['episodes'] or []
  22
  23     def _extract_episode_info(self, url, episode=None):
  24         fetch_episode = episode is None
  25         lang, m_type, episode_id = re.match(NhkVodIE._VALID_URL, url).groups()
  26         if episode_id.isdigit():
  27             episode_id = episode_id[:4] + '-' + episode_id[4:]
  28
  29         is_video = m_type == 'video'
  30         if fetch_episode:
  31             episode = self._call_api(
  32                 episode_id, lang, is_video, True, episode_id[:4] == '9999')[0]
  33         title = episode.get('sub_title_clean') or episode['sub_title']
  34
  35         def get_clean_field(key):
  36             return episode.get(key + '_clean') or episode.get(key)
  37
  38         series = get_clean_field('title')
  39
  40         thumbnails = []
  41         for s, w, h in [('', 640, 360), ('_l', 1280, 720)]:
  42             img_path = episode.get('image' + s)
  43             if not img_path:
  44                 continue
  45             thumbnails.append({
  46                 'id': '%dp' % h,
  47                 'height': h,
  48                 'width': w,
  49                 'url': 'https://www3.nhk.or.jp' + img_path,
  50             })
  51
  52         info = {
  53             'id': episode_id + '-' + lang,
  54             'title': '%s - %s' % (series, title) if series and title else title,
  55             'description': get_clean_field('description'),
  56             'thumbnails': thumbnails,
  57             'series': series,
  58             'episode': title,
  59         }
  60         if is_video:
  61             vod_id = episode['vod_id']
  62             info.update({
  63                 '_type': 'url_transparent',
  64                 'ie_key': 'Piksel',
  65                 'url': 'https://player.piksel.com/v/refid/nhkworld/prefid/' + vod_id,
  66                 'id': vod_id,
  67             })
  68         else:
  69             if fetch_episode:
  70                 audio_path = episode['audio']['audio']
  71                 info['formats'] = self._extract_m3u8_formats(
  72                     'https://nhkworld-vh.akamaihd.net/i%s/master.m3u8' % audio_path,
  73                     episode_id, 'm4a', entry_protocol='m3u8_native',
  74                     m3u8_id='hls', fatal=False)
  75                 for f in info['formats']:
  76                     f['language'] = lang
  77             else:
  78                 info.update({
  79                     '_type': 'url_transparent',
  80                     'ie_key': NhkVodIE.ie_key(),
  81                     'url': url,
  82                 })
  83         return info
  84
  85
  86 class NhkVodIE(NhkBaseIE):
  87     _VALID_URL = r'%s%s(?P<id>\d{7}|[^/]+?-\d{8}-[0-9a-z]+)' % (NhkBaseIE._BASE_URL_REGEX, NhkBaseIE._TYPE_REGEX)
  88     # Content available only for a limited period of time. Visit
  89     # https://www3.nhk.or.jp/nhkworld/en/ondemand/ for working samples.
  90     _TESTS = [{
  91         # video clip
  92         'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/video/9999011/',
  93         'md5': '7a90abcfe610ec22a6bfe15bd46b30ca',
  94         'info_dict': {
  95             'id': 'a95j5iza',
  96             'ext': 'mp4',
  97             'title': "Dining with the Chef - Chef Saito's Family recipe: MENCHI-KATSU",
  98             'description': 'md5:5aee4a9f9d81c26281862382103b0ea5',
  99             'timestamp': 1565965194,
 100             'upload_date': '20190816',
 101         },
 102     }, {
 103         # audio clip
 104         'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/audio/r_inventions-20201104-1/',
 105         'info_dict': {
 106             'id': 'r_inventions-20201104-1-en',
 107             'ext': 'm4a',
 108             'title': "Japan's Top Inventions - Miniature Video Cameras",
 109             'description': 'md5:07ea722bdbbb4936fdd360b6a480c25b',
 110         },
 111         'params': {
 112             # m3u8 download
 113             'skip_download': True,
 114         },
 115     }, {
 116         'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/video/2015173/',
 117         'only_matching': True,
 118     }, {
 119         'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/audio/plugin-20190404-1/',
 120         'only_matching': True,
 121     }, {
 122         'url': 'https://www3.nhk.or.jp/nhkworld/fr/ondemand/audio/plugin-20190404-1/',
 123         'only_matching': True,
 124     }, {
 125         'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/audio/j_art-20150903-1/',
 126         'only_matching': True,
 127     }]
 128
 129     def _real_extract(self, url):
 130         return self._extract_episode_info(url)
 131
 132
 133 class NhkVodProgramIE(NhkBaseIE):
 134     _VALID_URL = r'%s/program%s(?P<id>[0-9a-z]+)(?:.+?\btype=(?P<episode_type>clip|(?:radio|tv)Episode))?' % (NhkBaseIE._BASE_URL_REGEX, NhkBaseIE._TYPE_REGEX)
 135     _TESTS = [{
 136         # video program episodes
 137         'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/program/video/japanrailway',
 138         'info_dict': {
 139             'id': 'japanrailway',
 140             'title': 'Japan Railway Journal',
 141         },
 142         'playlist_mincount': 1,
 143     }, {
 144         # video program clips
 145         'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/program/video/japanrailway/?type=clip',
 146         'info_dict': {
 147             'id': 'japanrailway',
 148             'title': 'Japan Railway Journal',
 149         },
 150         'playlist_mincount': 5,
 151     }, {
 152         'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/program/video/10yearshayaomiyazaki/',
 153         'only_matching': True,
 154     }, {
 155         # audio program
 156         'url': 'https://www3.nhk.or.jp/nhkworld/en/ondemand/program/audio/listener/',
 157         'only_matching': True,
 158     }]
 159
 160     def _real_extract(self, url):
 161         lang, m_type, program_id, episode_type = re.match(self._VALID_URL, url).groups()
 162
 163         episodes = self._call_api(
 164             program_id, lang, m_type == 'video', False, episode_type == 'clip')
 165
 166         entries = []
 167         for episode in episodes:
 168             episode_path = episode.get('url')
 169             if not episode_path:
 170                 continue
 171             entries.append(self._extract_episode_info(
 172                 urljoin(url, episode_path), episode))
 173
 174         program_title = None
 175         if entries:
 176             program_title = entries[0].get('series')
 177
 178         return self.playlist_result(entries, program_id, program_title)