yt_dlp/extractor/streetvoice.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     int_or_none,
   4     parse_iso8601,
   5     str_or_none,
   6     strip_or_none,
   7     try_get,
   8     urljoin,
   9 )
  10
  11
  12 class StreetVoiceIE(InfoExtractor):
  13     _VALID_URL = r'https?://(?:.+?\.)?streetvoice\.com/[^/]+/songs/(?P<id>[0-9]+)'
  14     _TESTS = [{
  15         'url': 'https://streetvoice.com/skippylu/songs/123688/',
  16         'md5': '0eb535970629a5195685355f3ed60bfd',
  17         'info_dict': {
  18             'id': '123688',
  19             'ext': 'mp3',
  20             'title': '流浪',
  21             'description': 'md5:8eb0bfcc9dcd8aa82bd6efca66e3fea6',
  22             'thumbnail': r're:^https?://.*\.jpg',
  23             'duration': 270,
  24             'upload_date': '20100923',
  25             'uploader': 'Crispy脆樂團',
  26             'uploader_id': '627810',
  27             'uploader_url': 're:^https?://streetvoice.com/skippylu/',
  28             'timestamp': 1285261661,
  29             'view_count': int,
  30             'like_count': int,
  31             'comment_count': int,
  32             'repost_count': int,
  33             'track': '流浪',
  34             'track_id': '123688',
  35             'album': '2010',
  36         },
  37     }, {
  38         'url': 'http://tw.streetvoice.com/skippylu/songs/94440/',
  39         'only_matching': True,
  40     }]
  41
  42     def _real_extract(self, url):
  43         song_id = self._match_id(url)
  44         base_url = f'https://streetvoice.com/api/v4/song/{song_id}/'
  45         song = self._download_json(base_url, song_id, query={
  46             'fields': 'album,comments_count,created_at,id,image,length,likes_count,name,nickname,plays_count,profile,share_count,synopsis,user,username',
  47         })
  48         title = song['name']
  49
  50         formats = []
  51         for suffix, format_id in [('hls/file', 'hls'), ('file', 'http'), ('file/original', 'original')]:
  52             f_url = (self._download_json(
  53                 base_url + suffix + '/', song_id,
  54                 f'Downloading {format_id} format URL',
  55                 data=b'', fatal=False) or {}).get('file')
  56             if not f_url:
  57                 continue
  58             f = {
  59                 'ext': 'mp3',
  60                 'format_id': format_id,
  61                 'url': f_url,
  62                 'vcodec': 'none',
  63             }
  64             if format_id == 'hls':
  65                 f['protocol'] = 'm3u8_native'
  66             abr = self._search_regex(r'\.mp3\.(\d+)k', f_url, 'bitrate', default=None)
  67             if abr:
  68                 abr = int(abr)
  69                 f.update({
  70                     'abr': abr,
  71                     'tbr': abr,
  72                 })
  73             formats.append(f)
  74
  75         user = song.get('user') or {}
  76         username = user.get('username')
  77         get_count = lambda x: int_or_none(song.get(x + '_count'))
  78
  79         return {
  80             'id': song_id,
  81             'formats': formats,
  82             'title': title,
  83             'description': strip_or_none(song.get('synopsis')),
  84             'thumbnail': song.get('image'),
  85             'duration': int_or_none(song.get('length')),
  86             'timestamp': parse_iso8601(song.get('created_at')),
  87             'uploader': try_get(user, lambda x: x['profile']['nickname']),
  88             'uploader_id': str_or_none(user.get('id')),
  89             'uploader_url': urljoin(url, f'/{username}/') if username else None,
  90             'view_count': get_count('plays'),
  91             'like_count': get_count('likes'),
  92             'comment_count': get_count('comments'),
  93             'repost_count': get_count('share'),
  94             'track': title,
  95             'track_id': song_id,
  96             'album': try_get(song, lambda x: x['album']['name']),
  97         }