yt_dlp/extractor/joqrag.py

   1 import datetime as dt
   2 import urllib.parse
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     clean_html,
   7     datetime_from_str,
   8     unified_timestamp,
   9     urljoin,
  10 )
  11
  12
  13 class JoqrAgIE(InfoExtractor):
  14     IE_DESC = '超!A&G+ 文化放送 (f.k.a. AGQR) Nippon Cultural Broadcasting, Inc. (JOQR)'
  15     _VALID_URL = [r'https?://www\.uniqueradio\.jp/agplayer5/(?:player|inc-player-hls)\.php',
  16                   r'https?://(?:www\.)?joqr\.co\.jp/ag/',
  17                   r'https?://(?:www\.)?joqr\.co\.jp/qr/ag(?:daily|regular)program/?(?:$|[#?])']
  18     _TESTS = [{
  19         'url': 'https://www.uniqueradio.jp/agplayer5/player.php',
  20         'info_dict': {
  21             'id': 'live',
  22             'title': str,
  23             'channel': '超!A&G+',
  24             'description': str,
  25             'live_status': 'is_live',
  26             'release_timestamp': int,
  27         },
  28         'params': {
  29             'skip_download': True,
  30             'ignore_no_formats_error': True,
  31         },
  32     }, {
  33         'url': 'https://www.uniqueradio.jp/agplayer5/inc-player-hls.php',
  34         'only_matching': True,
  35     }, {
  36         'url': 'https://www.joqr.co.jp/ag/article/103760/',
  37         'only_matching': True,
  38     }, {
  39         'url': 'http://www.joqr.co.jp/qr/agdailyprogram/',
  40         'only_matching': True,
  41     }, {
  42         'url': 'http://www.joqr.co.jp/qr/agregularprogram/',
  43         'only_matching': True,
  44     }]
  45
  46     def _extract_metadata(self, variable, html):
  47         return clean_html(urllib.parse.unquote_plus(self._search_regex(
  48             rf'var\s+{variable}\s*=\s*(["\'])(?P<value>(?:(?!\1).)+)\1',
  49             html, 'metadata', group='value', default=''))) or None
  50
  51     def _extract_start_timestamp(self, video_id, is_live):
  52         def extract_start_time_from(date_str):
  53             dt_ = datetime_from_str(date_str) + dt.timedelta(hours=9)
  54             date = dt_.strftime('%Y%m%d')
  55             start_time = self._search_regex(
  56                 r'<h3[^>]+\bclass="dailyProgram-itemHeaderTime"[^>]*>[\s\d:]+–\s*(\d{1,2}:\d{1,2})',
  57                 self._download_webpage(
  58                     f'https://www.joqr.co.jp/qr/agdailyprogram/?date={date}', video_id,
  59                     note=f'Downloading program list of {date}', fatal=False,
  60                     errnote=f'Failed to download program list of {date}') or '',
  61                 'start time', default=None)
  62             if start_time:
  63                 return unified_timestamp(f'{dt_.strftime("%Y/%m/%d")} {start_time} +09:00')
  64             return None
  65
  66         start_timestamp = extract_start_time_from('today')
  67         if not start_timestamp:
  68             return None
  69
  70         if not is_live or start_timestamp < datetime_from_str('now').timestamp():
  71             return start_timestamp
  72         else:
  73             return extract_start_time_from('yesterday')
  74
  75     def _real_extract(self, url):
  76         video_id = 'live'
  77
  78         metadata = self._download_webpage(
  79             'https://www.uniqueradio.jp/aandg', video_id,
  80             note='Downloading metadata', errnote='Failed to download metadata')
  81         title = self._extract_metadata('Program_name', metadata)
  82
  83         if not title or title == '放送休止':
  84             formats = []
  85             live_status = 'is_upcoming'
  86             release_timestamp = self._extract_start_timestamp(video_id, False)
  87             msg = 'This stream is not currently live'
  88             if release_timestamp:
  89                 msg += (' and will start at '
  90                         + dt.datetime.fromtimestamp(release_timestamp).strftime('%Y-%m-%d %H:%M:%S'))
  91             self.raise_no_formats(msg, expected=True)
  92         else:
  93             m3u8_path = self._search_regex(
  94                 r'<source\s[^>]*\bsrc="([^"]+)"',
  95                 self._download_webpage(
  96                     'https://www.uniqueradio.jp/agplayer5/inc-player-hls.php', video_id,
  97                     note='Downloading player data', errnote='Failed to download player data'),
  98                 'm3u8 url')
  99             formats = self._extract_m3u8_formats(
 100                 urljoin('https://www.uniqueradio.jp/', m3u8_path), video_id)
 101             live_status = 'is_live'
 102             release_timestamp = self._extract_start_timestamp(video_id, True)
 103
 104         return {
 105             'id': video_id,
 106             'title': title,
 107             'channel': '超!A&G+',
 108             'description': self._extract_metadata('Program_text', metadata),
 109             'formats': formats,
 110             'live_status': live_status,
 111             'release_timestamp': release_timestamp,
 112         }