yt_dlp/extractor/nuum.py

   1 import functools
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     ExtractorError,
   6     OnDemandPagedList,
   7     UserNotLive,
   8     filter_dict,
   9     int_or_none,
  10     parse_iso8601,
  11     str_or_none,
  12     url_or_none,
  13 )
  14 from ..utils.traversal import traverse_obj
  15
  16
  17 class NuumBaseIE(InfoExtractor):
  18     def _call_api(self, path, video_id, description, query={}):
  19         response = self._download_json(
  20             f'https://nuum.ru/api/v2/{path}', video_id, query=query,
  21             note=f'Downloading {description} metadata',
  22             errnote=f'Unable to download {description} metadata')
  23         if error := response.get('error'):
  24             raise ExtractorError(f'API returned error: {error!r}')
  25         return response['result']
  26
  27     def _get_channel_info(self, channel_name):
  28         return self._call_api(
  29             'broadcasts/public', video_id=channel_name, description='channel',
  30             query={
  31                 'with_extra': 'true',
  32                 'channel_name': channel_name,
  33                 'with_deleted': 'true',
  34             })
  35
  36     def _parse_video_data(self, container, extract_formats=True):
  37         stream = traverse_obj(container, ('media_container_streams', 0, {dict})) or {}
  38         media = traverse_obj(stream, ('stream_media', 0, {dict})) or {}
  39         media_url = traverse_obj(media, (
  40             'media_meta', ('media_archive_url', 'media_url'), {url_or_none}), get_all=False)
  41
  42         video_id = str(container['media_container_id'])
  43         is_live = media.get('media_status') == 'RUNNING'
  44
  45         formats, subtitles = None, None
  46         headers = {'Referer': 'https://nuum.ru/'}
  47         if extract_formats:
  48             formats, subtitles = self._extract_m3u8_formats_and_subtitles(
  49                 media_url, video_id, 'mp4', live=is_live, headers=headers)
  50
  51         return filter_dict({
  52             'id': video_id,
  53             'is_live': is_live,
  54             'formats': formats,
  55             'subtitles': subtitles,
  56             'http_headers': headers,
  57             **traverse_obj(container, {
  58                 'title': ('media_container_name', {str}),
  59                 'description': ('media_container_description', {str}),
  60                 'timestamp': ('created_at', {parse_iso8601}),
  61                 'channel': ('media_container_channel', 'channel_name', {str}),
  62                 'channel_id': ('media_container_channel', 'channel_id', {str_or_none}),
  63             }),
  64             **traverse_obj(stream, {
  65                 'view_count': ('stream_total_viewers', {int_or_none}),
  66                 'concurrent_view_count': ('stream_current_viewers', {int_or_none}),
  67             }),
  68             **traverse_obj(media, {
  69                 'duration': ('media_duration', {int_or_none}),
  70                 'thumbnail': ('media_meta', ('media_preview_archive_url', 'media_preview_url'), {url_or_none}),
  71             }, get_all=False),
  72         })
  73
  74
  75 class NuumMediaIE(NuumBaseIE):
  76     IE_NAME = 'nuum:media'
  77     _VALID_URL = r'https?://nuum\.ru/(?:streams|videos|clips)/(?P<id>[\d]+)'
  78     _TESTS = [{
  79         'url': 'https://nuum.ru/streams/1592713-7-days-to-die',
  80         'only_matching': True,
  81     }, {
  82         'url': 'https://nuum.ru/videos/1567547-toxi-hurtz',
  83         'md5': 'ce28837a5bbffe6952d7bfd3d39811b0',
  84         'info_dict': {
  85             'id': '1567547',
  86             'ext': 'mp4',
  87             'title': 'Toxi$ - Hurtz',
  88             'description': '',
  89             'timestamp': 1702631651,
  90             'upload_date': '20231215',
  91             'thumbnail': r're:^https?://.+\.jpg',
  92             'view_count': int,
  93             'concurrent_view_count': int,
  94             'channel_id': '6911',
  95             'channel': 'toxis',
  96             'duration': 116,
  97         },
  98     }, {
  99         'url': 'https://nuum.ru/clips/1552564-pro-misu',
 100         'md5': 'b248ae1565b1e55433188f11beeb0ca1',
 101         'info_dict': {
 102             'id': '1552564',
 103             'ext': 'mp4',
 104             'title': 'Про Мису 🙃',
 105             'timestamp': 1701971828,
 106             'upload_date': '20231207',
 107             'thumbnail': r're:^https?://.+\.jpg',
 108             'view_count': int,
 109             'concurrent_view_count': int,
 110             'channel_id': '3320',
 111             'channel': 'Misalelik',
 112             'duration': 41,
 113         },
 114     }]
 115
 116     def _real_extract(self, url):
 117         video_id = self._match_id(url)
 118         video_data = self._call_api(f'media-containers/{video_id}', video_id, 'media')
 119
 120         return self._parse_video_data(video_data)
 121
 122
 123 class NuumLiveIE(NuumBaseIE):
 124     IE_NAME = 'nuum:live'
 125     _VALID_URL = r'https?://nuum\.ru/channel/(?P<id>[^/#?]+)/?(?:$|[#?])'
 126     _TESTS = [{
 127         'url': 'https://nuum.ru/channel/mts_live',
 128         'only_matching': True,
 129     }]
 130
 131     def _real_extract(self, url):
 132         channel = self._match_id(url)
 133         channel_info = self._get_channel_info(channel)
 134         if traverse_obj(channel_info, ('channel', 'channel_is_live')) is False:
 135             raise UserNotLive(video_id=channel)
 136
 137         info = self._parse_video_data(channel_info['media_container'])
 138         return {
 139             'webpage_url': f'https://nuum.ru/streams/{info["id"]}',
 140             'extractor_key': NuumMediaIE.ie_key(),
 141             'extractor': NuumMediaIE.IE_NAME,
 142             **info,
 143         }
 144
 145
 146 class NuumTabIE(NuumBaseIE):
 147     IE_NAME = 'nuum:tab'
 148     _VALID_URL = r'https?://nuum\.ru/channel/(?P<id>[^/#?]+)/(?P<type>streams|videos|clips)'
 149     _TESTS = [{
 150         'url': 'https://nuum.ru/channel/dankon_/clips',
 151         'info_dict': {
 152             'id': 'dankon__clips',
 153             'title': 'Dankon_',
 154         },
 155         'playlist_mincount': 29,
 156     }, {
 157         'url': 'https://nuum.ru/channel/dankon_/videos',
 158         'info_dict': {
 159             'id': 'dankon__videos',
 160             'title': 'Dankon_',
 161         },
 162         'playlist_mincount': 2,
 163     }, {
 164         'url': 'https://nuum.ru/channel/dankon_/streams',
 165         'info_dict': {
 166             'id': 'dankon__streams',
 167             'title': 'Dankon_',
 168         },
 169         'playlist_mincount': 1,
 170     }]
 171
 172     _PAGE_SIZE = 50
 173
 174     def _fetch_page(self, channel_id, tab_type, tab_id, page):
 175         CONTAINER_TYPES = {
 176             'clips': ['SHORT_VIDEO', 'REVIEW_VIDEO'],
 177             'videos': ['LONG_VIDEO'],
 178             'streams': ['SINGLE'],
 179         }
 180
 181         media_containers = self._call_api(
 182             'media-containers', video_id=tab_id, description=f'{tab_type} tab page {page + 1}',
 183             query={
 184                 'limit': self._PAGE_SIZE,
 185                 'offset': page * self._PAGE_SIZE,
 186                 'channel_id': channel_id,
 187                 'media_container_status': 'STOPPED',
 188                 'media_container_type': CONTAINER_TYPES[tab_type],
 189             })
 190         for container in traverse_obj(media_containers, (..., {dict})):
 191             metadata = self._parse_video_data(container, extract_formats=False)
 192             yield self.url_result(f'https://nuum.ru/videos/{metadata["id"]}', NuumMediaIE, **metadata)
 193
 194     def _real_extract(self, url):
 195         channel_name, tab_type = self._match_valid_url(url).group('id', 'type')
 196         tab_id = f'{channel_name}_{tab_type}'
 197         channel_data = self._get_channel_info(channel_name)['channel']
 198
 199         return self.playlist_result(OnDemandPagedList(functools.partial(
 200             self._fetch_page, channel_data['channel_id'], tab_type, tab_id), self._PAGE_SIZE),
 201             playlist_id=tab_id, playlist_title=channel_data.get('channel_name'))