yt_dlp/extractor/voicy.py

   1 import itertools
   2
   3 from .common import InfoExtractor
   4 from ..compat import compat_str
   5 from ..utils import (
   6     ExtractorError,
   7     smuggle_url,
   8     str_or_none,
   9     traverse_obj,
  10     unified_strdate,
  11     unsmuggle_url,
  12 )
  13
  14
  15 class VoicyBaseIE(InfoExtractor):
  16     def _extract_from_playlist_data(self, value):
  17         voice_id = compat_str(value.get('PlaylistId'))
  18         upload_date = unified_strdate(value.get('Published'), False)
  19         items = [self._extract_single_article(voice_data) for voice_data in value['VoiceData']]
  20         return {
  21             '_type': 'multi_video',
  22             'entries': items,
  23             'id': voice_id,
  24             'title': compat_str(value.get('PlaylistName')),
  25             'uploader': value.get('SpeakerName'),
  26             'uploader_id': str_or_none(value.get('SpeakerId')),
  27             'channel': value.get('ChannelName'),
  28             'channel_id': str_or_none(value.get('ChannelId')),
  29             'upload_date': upload_date,
  30         }
  31
  32     def _extract_single_article(self, entry):
  33         formats = [{
  34             'url': entry['VoiceHlsFile'],
  35             'format_id': 'hls',
  36             'ext': 'm4a',
  37             'acodec': 'aac',
  38             'vcodec': 'none',
  39             'protocol': 'm3u8_native',
  40         }, {
  41             'url': entry['VoiceFile'],
  42             'format_id': 'mp3',
  43             'ext': 'mp3',
  44             'acodec': 'mp3',
  45             'vcodec': 'none',
  46         }]
  47         return {
  48             'id': compat_str(entry.get('ArticleId')),
  49             'title': entry.get('ArticleTitle'),
  50             'description': entry.get('MediaName'),
  51             'formats': formats,
  52         }
  53
  54     def _call_api(self, url, video_id, **kwargs):
  55         response = self._download_json(url, video_id, **kwargs)
  56         if response.get('Status') != 0:
  57             message = traverse_obj(response, ('Value', 'Error', 'Message'), expected_type=compat_str)
  58             if not message:
  59                 message = 'There was a error in the response: %d' % response.get('Status')
  60             raise ExtractorError(message, expected=False)
  61         return response.get('Value')
  62
  63
  64 class VoicyIE(VoicyBaseIE):
  65     _WORKING = False
  66     IE_NAME = 'voicy'
  67     _VALID_URL = r'https?://voicy\.jp/channel/(?P<channel_id>\d+)/(?P<id>\d+)'
  68     ARTICLE_LIST_API_URL = 'https://vmw.api.voicy.jp/articles_list?channel_id=%s&pid=%s'
  69     _TESTS = [{
  70         'url': 'https://voicy.jp/channel/1253/122754',
  71         'info_dict': {
  72             'id': '122754',
  73             'title': '1/21(木)声日記：ついに原稿終わった！！',
  74             'uploader': 'ちょまど@ ITエンジニアなオタク',
  75             'uploader_id': '7339',
  76         },
  77         'playlist_mincount': 9,
  78     }]
  79
  80     def _real_extract(self, url):
  81         mobj = self._match_valid_url(url)
  82         assert mobj
  83         voice_id = mobj.group('id')
  84         channel_id = mobj.group('channel_id')
  85         url, article_list = unsmuggle_url(url)
  86         if not article_list:
  87             article_list = self._call_api(self.ARTICLE_LIST_API_URL % (channel_id, voice_id), voice_id)
  88         return self._extract_from_playlist_data(article_list)
  89
  90
  91 class VoicyChannelIE(VoicyBaseIE):
  92     _WORKING = False
  93     IE_NAME = 'voicy:channel'
  94     _VALID_URL = r'https?://voicy\.jp/channel/(?P<id>\d+)'
  95     PROGRAM_LIST_API_URL = 'https://vmw.api.voicy.jp/program_list/all?channel_id=%s&limit=20&public_type=3%s'
  96     _TESTS = [{
  97         'url': 'https://voicy.jp/channel/1253/',
  98         'info_dict': {
  99             'id': '7339',
 100             'title': 'ゆるふわ日常ラジオ #ちょまラジ',
 101             'uploader': 'ちょまど@ ITエンジニアなオタク',
 102             'uploader_id': '7339',
 103         },
 104         'playlist_mincount': 54,
 105     }]
 106
 107     @classmethod
 108     def suitable(cls, url):
 109         return not VoicyIE.suitable(url) and super().suitable(url)
 110
 111     def _entries(self, channel_id):
 112         pager = ''
 113         for count in itertools.count(1):
 114             article_list = self._call_api(self.PROGRAM_LIST_API_URL % (channel_id, pager), channel_id, note='Paging #%d' % count)
 115             playlist_data = article_list.get('PlaylistData')
 116             if not playlist_data:
 117                 break
 118             yield from playlist_data
 119             last = playlist_data[-1]
 120             pager = '&pid=%d&p_date=%s&play_count=%s' % (last['PlaylistId'], last['Published'], last['PlayCount'])
 121
 122     def _real_extract(self, url):
 123         channel_id = self._match_id(url)
 124         articles = self._entries(channel_id)
 125
 126         first_article = next(articles, None)
 127         title = traverse_obj(first_article, ('ChannelName', ), expected_type=compat_str)
 128         speaker_name = traverse_obj(first_article, ('SpeakerName', ), expected_type=compat_str)
 129         if not title and speaker_name:
 130             title = 'Uploads from %s' % speaker_name
 131         if not title:
 132             title = 'Uploads from channel ID %s' % channel_id
 133
 134         articles = itertools.chain([first_article], articles) if first_article else articles
 135
 136         playlist = (
 137             self.url_result(smuggle_url('https://voicy.jp/channel/%s/%d' % (channel_id, value['PlaylistId']), value), VoicyIE.ie_key())
 138             for value in articles)
 139         return {
 140             '_type': 'playlist',
 141             'entries': playlist,
 142             'id': channel_id,
 143             'title': title,
 144             'channel': speaker_name,
 145             'channel_id': channel_id,
 146         }