[yt-dlp.git] / yt_dlp / extractor / nuum.py

import functools

from .common import InfoExtractor
from ..utils import (
    ExtractorError,
    OnDemandPagedList,
    UserNotLive,
    filter_dict,
    int_or_none,
    parse_iso8601,
    str_or_none,
    url_or_none,
)
from ..utils.traversal import traverse_obj


class NuumBaseIE(InfoExtractor):
    def _call_api(self, path, video_id, description, query={}):
        response = self._download_json(
            f'https://nuum.ru/api/v2/{path}', video_id, query=query,
            note=f'Downloading {description} metadata',
            errnote=f'Unable to download {description} metadata')
        if error := response.get('error'):
            raise ExtractorError(f'API returned error: {error!r}')
        return response['result']

    def _get_channel_info(self, channel_name):
        return self._call_api(
            'broadcasts/public', video_id=channel_name, description='channel',
            query={
                'with_extra': 'true',
                'channel_name': channel_name,
                'with_deleted': 'true',
            })

    def _parse_video_data(self, container, extract_formats=True):
        stream = traverse_obj(container, ('media_container_streams', 0, {dict})) or {}
        media = traverse_obj(stream, ('stream_media', 0, {dict})) or {}
        media_url = traverse_obj(media, (
            'media_meta', ('media_archive_url', 'media_url'), {url_or_none}), get_all=False)

        video_id = str(container['media_container_id'])
        is_live = media.get('media_status') == 'RUNNING'

        formats, subtitles = None, None
        if extract_formats:
            formats, subtitles = self._extract_m3u8_formats_and_subtitles(
                media_url, video_id, 'mp4', live=is_live)

        return filter_dict({
            'id': video_id,
            'is_live': is_live,
            'formats': formats,
            'subtitles': subtitles,
            **traverse_obj(container, {
                'title': ('media_container_name', {str}),
                'description': ('media_container_description', {str}),
                'timestamp': ('created_at', {parse_iso8601}),
                'channel': ('media_container_channel', 'channel_name', {str}),
                'channel_id': ('media_container_channel', 'channel_id', {str_or_none}),
            }),
            **traverse_obj(stream, {
                'view_count': ('stream_total_viewers', {int_or_none}),
                'concurrent_view_count': ('stream_current_viewers', {int_or_none}),
            }),
            **traverse_obj(media, {
                'duration': ('media_duration', {int_or_none}),
                'thumbnail': ('media_meta', ('media_preview_archive_url', 'media_preview_url'), {url_or_none}),
            }, get_all=False),
        })


class NuumMediaIE(NuumBaseIE):
    IE_NAME = 'nuum:media'
    _VALID_URL = r'https?://nuum\.ru/(?:streams|videos|clips)/(?P<id>[\d]+)'
    _TESTS = [{
        'url': 'https://nuum.ru/streams/1592713-7-days-to-die',
        'only_matching': True,
    }, {
        'url': 'https://nuum.ru/videos/1567547-toxi-hurtz',
        'md5': 'f1d9118a30403e32b702a204eb03aca3',
        'info_dict': {
            'id': '1567547',
            'ext': 'mp4',
            'title': 'Toxi$ - Hurtz',
            'description': '',
            'timestamp': 1702631651,
            'upload_date': '20231215',
            'thumbnail': r're:^https?://.+\.jpg',
            'view_count': int,
            'concurrent_view_count': int,
            'channel_id': '6911',
            'channel': 'toxis',
            'duration': 116,
        },
    }, {
        'url': 'https://nuum.ru/clips/1552564-pro-misu',
        'md5': 'b248ae1565b1e55433188f11beeb0ca1',
        'info_dict': {
            'id': '1552564',
            'ext': 'mp4',
            'title': 'Про Мису 🙃',
            'timestamp': 1701971828,
            'upload_date': '20231207',
            'thumbnail': r're:^https?://.+\.jpg',
            'view_count': int,
            'concurrent_view_count': int,
            'channel_id': '3320',
            'channel': 'Misalelik',
            'duration': 41,
        },
    }]

    def _real_extract(self, url):
        video_id = self._match_id(url)
        video_data = self._call_api(f'media-containers/{video_id}', video_id, 'media')

        return self._parse_video_data(video_data)


class NuumLiveIE(NuumBaseIE):
    IE_NAME = 'nuum:live'
    _VALID_URL = r'https?://nuum\.ru/channel/(?P<id>[^/#?]+)/?(?:$|[#?])'
    _TESTS = [{
        'url': 'https://nuum.ru/channel/mts_live',
        'only_matching': True,
    }]

    def _real_extract(self, url):
        channel = self._match_id(url)
        channel_info = self._get_channel_info(channel)
        if traverse_obj(channel_info, ('channel', 'channel_is_live')) is False:
            raise UserNotLive(video_id=channel)

        info = self._parse_video_data(channel_info['media_container'])
        return {
            'webpage_url': f'https://nuum.ru/streams/{info["id"]}',
            'extractor_key': NuumMediaIE.ie_key(),
            'extractor': NuumMediaIE.IE_NAME,
            **info,
        }


class NuumTabIE(NuumBaseIE):
    IE_NAME = 'nuum:tab'
    _VALID_URL = r'https?://nuum\.ru/channel/(?P<id>[^/#?]+)/(?P<type>streams|videos|clips)'
    _TESTS = [{
        'url': 'https://nuum.ru/channel/dankon_/clips',
        'info_dict': {
            'id': 'dankon__clips',
            'title': 'Dankon_',
        },
        'playlist_mincount': 29,
    }, {
        'url': 'https://nuum.ru/channel/dankon_/videos',
        'info_dict': {
            'id': 'dankon__videos',
            'title': 'Dankon_',
        },
        'playlist_mincount': 2,
    }, {
        'url': 'https://nuum.ru/channel/dankon_/streams',
        'info_dict': {
            'id': 'dankon__streams',
            'title': 'Dankon_',
        },
        'playlist_mincount': 1,
    }]

    _PAGE_SIZE = 50

    def _fetch_page(self, channel_id, tab_type, tab_id, page):
        CONTAINER_TYPES = {
            'clips': ['SHORT_VIDEO', 'REVIEW_VIDEO'],
            'videos': ['LONG_VIDEO'],
            'streams': ['SINGLE'],
        }

        media_containers = self._call_api(
            'media-containers', video_id=tab_id, description=f'{tab_type} tab page {page + 1}',
            query={
                'limit': self._PAGE_SIZE,
                'offset': page * self._PAGE_SIZE,
                'channel_id': channel_id,
                'media_container_status': 'STOPPED',
                'media_container_type': CONTAINER_TYPES[tab_type],
            })
        for container in traverse_obj(media_containers, (..., {dict})):
            metadata = self._parse_video_data(container, extract_formats=False)
            yield self.url_result(f'https://nuum.ru/videos/{metadata["id"]}', NuumMediaIE, **metadata)

    def _real_extract(self, url):
        channel_name, tab_type = self._match_valid_url(url).group('id', 'type')
        tab_id = f'{channel_name}_{tab_type}'
        channel_data = self._get_channel_info(channel_name)['channel']

        return self.playlist_result(OnDemandPagedList(functools.partial(
            self._fetch_page, channel_data['channel_id'], tab_type, tab_id), self._PAGE_SIZE),
            playlist_id=tab_id, playlist_title=channel_data.get('channel_name'))
Commit	Line	Data
acaf806c D	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	if extract_formats:
	47	formats, subtitles = self._extract_m3u8_formats_and_subtitles(
	48	media_url, video_id, 'mp4', live=is_live)
	49
	50	return filter_dict({
	51	'id': video_id,
	52	'is_live': is_live,
	53	'formats': formats,
	54	'subtitles': subtitles,
	55	**traverse_obj(container, {
	56	'title': ('media_container_name', {str}),
	57	'description': ('media_container_description', {str}),
	58	'timestamp': ('created_at', {parse_iso8601}),
	59	'channel': ('media_container_channel', 'channel_name', {str}),
	60	'channel_id': ('media_container_channel', 'channel_id', {str_or_none}),
	61	}),
	62	**traverse_obj(stream, {
	63	'view_count': ('stream_total_viewers', {int_or_none}),
	64	'concurrent_view_count': ('stream_current_viewers', {int_or_none}),
65	}),
66	**traverse_obj(media, {
67	'duration': ('media_duration', {int_or_none}),
68	'thumbnail': ('media_meta', ('media_preview_archive_url', 'media_preview_url'), {url_or_none}),
69	}, get_all=False),
70	})
71
72
73	class NuumMediaIE(NuumBaseIE):
74	IE_NAME = 'nuum:media'
75	_VALID_URL = r'https?://nuum\.ru/(?:streams\|videos\|clips)/(?P<id>[\d]+)'
76	_TESTS = [{
77	'url': 'https://nuum.ru/streams/1592713-7-days-to-die',
78	'only_matching': True,
79	}, {
80	'url': 'https://nuum.ru/videos/1567547-toxi-hurtz',
81	'md5': 'f1d9118a30403e32b702a204eb03aca3',
82	'info_dict': {
83	'id': '1567547',
84	'ext': 'mp4',
85	'title': 'Toxi$ - Hurtz',
86	'description': '',
87	'timestamp': 1702631651,
88	'upload_date': '20231215',
89	'thumbnail': r're:^https?://.+\.jpg',
90	'view_count': int,
91	'concurrent_view_count': int,
92	'channel_id': '6911',
93	'channel': 'toxis',
94	'duration': 116,
95	},
96	}, {
97	'url': 'https://nuum.ru/clips/1552564-pro-misu',
98	'md5': 'b248ae1565b1e55433188f11beeb0ca1',
99	'info_dict': {
100	'id': '1552564',
101	'ext': 'mp4',
102	'title': 'Про Мису 🙃',
103	'timestamp': 1701971828,
104	'upload_date': '20231207',
105	'thumbnail': r're:^https?://.+\.jpg',
106	'view_count': int,
107	'concurrent_view_count': int,
108	'channel_id': '3320',
109	'channel': 'Misalelik',
110	'duration': 41,
111	},
112	}]
113
114	def _real_extract(self, url):
115	video_id = self._match_id(url)
116	video_data = self._call_api(f'media-containers/{video_id}', video_id, 'media')
117
118	return self._parse_video_data(video_data)
119
120
121	class NuumLiveIE(NuumBaseIE):
122	IE_NAME = 'nuum:live'
123	_VALID_URL = r'https?://nuum\.ru/channel/(?P<id>[^/#?]+)/?(?:$\|[#?])'
124	_TESTS = [{
125	'url': 'https://nuum.ru/channel/mts_live',
126	'only_matching': True,
127	}]
128
129	def _real_extract(self, url):
130	channel = self._match_id(url)
131	channel_info = self._get_channel_info(channel)
132	if traverse_obj(channel_info, ('channel', 'channel_is_live')) is False:
133	raise UserNotLive(video_id=channel)
134
135	info = self._parse_video_data(channel_info['media_container'])
136	return {
137	'webpage_url': f'https://nuum.ru/streams/{info["id"]}',
138	'extractor_key': NuumMediaIE.ie_key(),
139	'extractor': NuumMediaIE.IE_NAME,
140	**info,
141	}
142
143
144	class NuumTabIE(NuumBaseIE):
145	IE_NAME = 'nuum:tab'
146	_VALID_URL = r'https?://nuum\.ru/channel/(?P<id>[^/#?]+)/(?P<type>streams\|videos\|clips)'
147	_TESTS = [{
148	'url': 'https://nuum.ru/channel/dankon_/clips',
149	'info_dict': {
150	'id': 'dankon__clips',
151	'title': 'Dankon_',
152	},
153	'playlist_mincount': 29,
154	}, {
155	'url': 'https://nuum.ru/channel/dankon_/videos',
156	'info_dict': {
157	'id': 'dankon__videos',
158	'title': 'Dankon_',
159	},
160	'playlist_mincount': 2,
161	}, {
162	'url': 'https://nuum.ru/channel/dankon_/streams',
163	'info_dict': {
164	'id': 'dankon__streams',
165	'title': 'Dankon_',
166	},
167	'playlist_mincount': 1,
168	}]
169
170	_PAGE_SIZE = 50
171
172	def _fetch_page(self, channel_id, tab_type, tab_id, page):
173	CONTAINER_TYPES = {
174	'clips': ['SHORT_VIDEO', 'REVIEW_VIDEO'],
175	'videos': ['LONG_VIDEO'],
176	'streams': ['SINGLE'],
177	}
178
179	media_containers = self._call_api(
180	'media-containers', video_id=tab_id, description=f'{tab_type} tab page {page + 1}',
181	query={
182	'limit': self._PAGE_SIZE,
183	'offset': page * self._PAGE_SIZE,
184	'channel_id': channel_id,
185	'media_container_status': 'STOPPED',
186	'media_container_type': CONTAINER_TYPES[tab_type],
187	})
188	for container in traverse_obj(media_containers, (..., {dict})):
189	metadata = self._parse_video_data(container, extract_formats=False)
190	yield self.url_result(f'https://nuum.ru/videos/{metadata["id"]}', NuumMediaIE, **metadata)
191
192	def _real_extract(self, url):
193	channel_name, tab_type = self._match_valid_url(url).group('id', 'type')
194	tab_id = f'{channel_name}_{tab_type}'
195	channel_data = self._get_channel_info(channel_name)['channel']
196
197	return self.playlist_result(OnDemandPagedList(functools.partial(
198	self._fetch_page, channel_data['channel_id'], tab_type, tab_id), self._PAGE_SIZE),
199	playlist_id=tab_id, playlist_title=channel_data.get('channel_name'))