yt_dlp/extractor/line.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4
   5 from .common import InfoExtractor
   6 from ..compat import compat_str
   7 from ..utils import (
   8     format_field,
   9     int_or_none,
  10     str_or_none,
  11 )
  12
  13
  14 class LineLiveBaseIE(InfoExtractor):
  15     _API_BASE_URL = 'https://live-api.line-apps.com/web/v4.0/channel/'
  16
  17     def _parse_broadcast_item(self, item):
  18         broadcast_id = compat_str(item['id'])
  19         title = item['title']
  20         is_live = item.get('isBroadcastingNow')
  21
  22         thumbnails = []
  23         for thumbnail_id, thumbnail_url in (item.get('thumbnailURLs') or {}).items():
  24             if not thumbnail_url:
  25                 continue
  26             thumbnails.append({
  27                 'id': thumbnail_id,
  28                 'url': thumbnail_url,
  29             })
  30
  31         channel = item.get('channel') or {}
  32         channel_id = str_or_none(channel.get('id'))
  33
  34         return {
  35             'id': broadcast_id,
  36             'title': title,
  37             'thumbnails': thumbnails,
  38             'timestamp': int_or_none(item.get('createdAt')),
  39             'channel': channel.get('name'),
  40             'channel_id': channel_id,
  41             'channel_url': format_field(channel_id, template='https://live.line.me/channels/%s'),
  42             'duration': int_or_none(item.get('archiveDuration')),
  43             'view_count': int_or_none(item.get('viewerCount')),
  44             'comment_count': int_or_none(item.get('chatCount')),
  45             'is_live': is_live,
  46         }
  47
  48
  49 class LineLiveIE(LineLiveBaseIE):
  50     _VALID_URL = r'https?://live\.line\.me/channels/(?P<channel_id>\d+)/broadcast/(?P<id>\d+)'
  51     _TESTS = [{
  52         'url': 'https://live.line.me/channels/5833718/broadcast/18373277',
  53         'md5': '2c15843b8cb3acd55009ddcb2db91f7c',
  54         'info_dict': {
  55             'id': '18373277',
  56             'title': '2021/12/05 （15分犬）定例譲渡会🐶',
  57             'ext': 'mp4',
  58             'timestamp': 1638674925,
  59             'upload_date': '20211205',
  60             'thumbnail': 'md5:e1f5817e60f4a72b7e43377cf308d7ef',
  61             'channel_url': 'https://live.line.me/channels/5833718',
  62             'channel': 'Yahooニュース掲載🗞プロフ見てね🐕🐕',
  63             'channel_id': '5833718',
  64             'duration': 937,
  65             'view_count': int,
  66             'comment_count': int,
  67             'is_live': False,
  68         }
  69     }, {
  70         # archiveStatus == 'DELETED'
  71         'url': 'https://live.line.me/channels/4778159/broadcast/16378488',
  72         'only_matching': True,
  73     }]
  74
  75     def _real_extract(self, url):
  76         channel_id, broadcast_id = self._match_valid_url(url).groups()
  77         broadcast = self._download_json(
  78             self._API_BASE_URL + '%s/broadcast/%s' % (channel_id, broadcast_id),
  79             broadcast_id)
  80         item = broadcast['item']
  81         info = self._parse_broadcast_item(item)
  82         protocol = 'm3u8' if info['is_live'] else 'm3u8_native'
  83         formats = []
  84         for k, v in (broadcast.get(('live' if info['is_live'] else 'archived') + 'HLSURLs') or {}).items():
  85             if not v:
  86                 continue
  87             if k == 'abr':
  88                 formats.extend(self._extract_m3u8_formats(
  89                     v, broadcast_id, 'mp4', protocol,
  90                     m3u8_id='hls', fatal=False))
  91                 continue
  92             f = {
  93                 'ext': 'mp4',
  94                 'format_id': 'hls-' + k,
  95                 'protocol': protocol,
  96                 'url': v,
  97             }
  98             if not k.isdigit():
  99                 f['vcodec'] = 'none'
 100             formats.append(f)
 101         if not formats:
 102             archive_status = item.get('archiveStatus')
 103             if archive_status != 'ARCHIVED':
 104                 self.raise_no_formats('this video has been ' + archive_status.lower(), expected=True)
 105         self._sort_formats(formats)
 106         info['formats'] = formats
 107         return info
 108
 109
 110 class LineLiveChannelIE(LineLiveBaseIE):
 111     _VALID_URL = r'https?://live\.line\.me/channels/(?P<id>\d+)(?!/broadcast/\d+)(?:[/?&#]|$)'
 112     _TEST = {
 113         'url': 'https://live.line.me/channels/5893542',
 114         'info_dict': {
 115             'id': '5893542',
 116             'title': 'いくらちゃんだよぉ🦒',
 117             'description': 'md5:4d418087973ad081ceb1b3481f0b1816',
 118         },
 119         'playlist_mincount': 29
 120     }
 121
 122     def _archived_broadcasts_entries(self, archived_broadcasts, channel_id):
 123         while True:
 124             for row in (archived_broadcasts.get('rows') or []):
 125                 share_url = str_or_none(row.get('shareURL'))
 126                 if not share_url:
 127                     continue
 128                 info = self._parse_broadcast_item(row)
 129                 info.update({
 130                     '_type': 'url',
 131                     'url': share_url,
 132                     'ie_key': LineLiveIE.ie_key(),
 133                 })
 134                 yield info
 135             if not archived_broadcasts.get('hasNextPage'):
 136                 return
 137             archived_broadcasts = self._download_json(
 138                 self._API_BASE_URL + channel_id + '/archived_broadcasts',
 139                 channel_id, query={
 140                     'lastId': info['id'],
 141                 })
 142
 143     def _real_extract(self, url):
 144         channel_id = self._match_id(url)
 145         channel = self._download_json(self._API_BASE_URL + channel_id, channel_id)
 146         return self.playlist_result(
 147             self._archived_broadcasts_entries(channel.get('archivedBroadcasts') or {}, channel_id),
 148             channel_id, channel.get('title'), channel.get('information'))