[ie/nuum] Add extractors (#8868)

Authored by: DmitryScaletta, seproDev

Co-authored-by: sepro <4618135+seproDev@users.noreply.github.com>
This commit is contained in:
DmitryScaletta 2024-02-05 05:17:39 +03:00 committed by GitHub
parent 07256b9fee
commit acaf806c15
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
3 changed files with 204 additions and 164 deletions

View File

@ -1354,6 +1354,11 @@
NYTimesCookingIE,
NYTimesCookingRecipeIE,
)
from .nuum import (
NuumLiveIE,
NuumTabIE,
NuumMediaIE,
)
from .nuvid import NuvidIE
from .nzherald import NZHeraldIE
from .nzonscreen import NZOnScreenIE
@ -2315,11 +2320,6 @@
WashingtonPostIE,
WashingtonPostArticleIE,
)
from .wasdtv import (
WASDTVStreamIE,
WASDTVRecordIE,
WASDTVClipIE,
)
from .wat import WatIE
from .wdr import (
WDRIE,

199
yt_dlp/extractor/nuum.py Normal file
View File

@ -0,0 +1,199 @@
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'))

View File

@ -1,159 +0,0 @@
from .common import InfoExtractor
from ..utils import (
ExtractorError,
int_or_none,
parse_iso8601,
traverse_obj,
try_get,
)
class WASDTVBaseIE(InfoExtractor):
def _fetch(self, path, video_id, description, query={}):
response = self._download_json(
f'https://wasd.tv/api/{path}', video_id, query=query,
note=f'Downloading {description} metadata',
errnote=f'Unable to download {description} metadata')
error = response.get('error')
if error:
raise ExtractorError(f'{self.IE_NAME} returned error: {error}', expected=True)
return response.get('result')
def _extract_thumbnails(self, thumbnails_dict):
return [{
'url': url,
'preference': index,
} for index, url in enumerate(
traverse_obj(thumbnails_dict, (('small', 'medium', 'large'),))) if url]
def _real_extract(self, url):
container = self._get_container(url)
stream = traverse_obj(container, ('media_container_streams', 0))
media = try_get(stream, lambda x: x['stream_media'][0])
if not media:
raise ExtractorError('Can not extract media data.', expected=True)
media_meta = media.get('media_meta')
media_url, is_live = self._get_media_url(media_meta)
video_id = media.get('media_id') or container.get('media_container_id')
formats, subtitles = self._extract_m3u8_formats_and_subtitles(media_url, video_id, 'mp4')
return {
'id': str(video_id),
'title': container.get('media_container_name') or self._og_search_title(self._download_webpage(url, video_id)),
'description': container.get('media_container_description'),
'thumbnails': self._extract_thumbnails(media_meta.get('media_preview_images')),
'timestamp': parse_iso8601(container.get('created_at')),
'view_count': int_or_none(stream.get('stream_current_viewers' if is_live else 'stream_total_viewers')),
'is_live': is_live,
'formats': formats,
'subtitles': subtitles,
}
def _get_container(self, url):
raise NotImplementedError('Subclass for get media container')
def _get_media_url(self, media_meta):
raise NotImplementedError('Subclass for get media url')
class WASDTVStreamIE(WASDTVBaseIE):
IE_NAME = 'wasdtv:stream'
_VALID_URL = r'https?://wasd\.tv/(?P<id>[^/#?]+)$'
_TESTS = [{
'url': 'https://wasd.tv/24_7',
'info_dict': {
'id': '559738',
'ext': 'mp4',
'title': 'Live 24/7 Music',
'description': '24&#x2F;7 Music',
'timestamp': int,
'upload_date': r're:^\d{8}$',
'is_live': True,
'view_count': int,
},
}]
def _get_container(self, url):
nickname = self._match_id(url)
channel = self._fetch(f'channels/nicknames/{nickname}', video_id=nickname, description='channel')
channel_id = channel.get('channel_id')
containers = self._fetch(
'v2/media-containers', channel_id, 'running media containers',
query={
'channel_id': channel_id,
'media_container_type': 'SINGLE',
'media_container_status': 'RUNNING',
})
if not containers:
raise ExtractorError(f'{nickname} is offline', expected=True)
return containers[0]
def _get_media_url(self, media_meta):
return media_meta['media_url'], True
class WASDTVRecordIE(WASDTVBaseIE):
IE_NAME = 'wasdtv:record'
_VALID_URL = r'https?://wasd\.tv/[^/#?]+(?:/videos)?\?record=(?P<id>\d+)$'
_TESTS = [{
'url': 'https://wasd.tv/spacemita/videos?record=907755',
'md5': 'c9899dd85be4cc997816ff9f9ca516ce',
'info_dict': {
'id': '906825',
'ext': 'mp4',
'title': 'Музыкальный',
'description': 'md5:f510388d929ff60ae61d4c3cab3137cc',
'timestamp': 1645812079,
'upload_date': '20220225',
'thumbnail': r're:^https?://.+\.jpg',
'is_live': False,
'view_count': int,
},
}, {
'url': 'https://wasd.tv/spacemita?record=907755',
'only_matching': True,
}]
def _get_container(self, url):
container_id = self._match_id(url)
return self._fetch(
f'v2/media-containers/{container_id}', container_id, 'media container')
def _get_media_url(self, media_meta):
media_archive_url = media_meta.get('media_archive_url')
if media_archive_url:
return media_archive_url, False
return media_meta['media_url'], True
class WASDTVClipIE(WASDTVBaseIE):
IE_NAME = 'wasdtv:clip'
_VALID_URL = r'https?://wasd\.tv/[^/#?]+/clips\?clip=(?P<id>\d+)$'
_TESTS = [{
'url': 'https://wasd.tv/spacemita/clips?clip=26804',
'md5': '818885e720143d7a4e776ff66fcff148',
'info_dict': {
'id': '26804',
'ext': 'mp4',
'title': 'Пуш флексит на голове стримера',
'timestamp': 1646682908,
'upload_date': '20220307',
'thumbnail': r're:^https?://.+\.jpg',
'view_count': int,
},
}]
def _real_extract(self, url):
clip_id = self._match_id(url)
clip = self._fetch(f'v2/clips/{clip_id}', video_id=clip_id, description='clip')
clip_data = clip.get('clip_data')
formats, subtitles = self._extract_m3u8_formats_and_subtitles(clip_data.get('url'), video_id=clip_id, ext='mp4')
return {
'id': clip_id,
'title': clip.get('clip_title') or self._og_search_title(self._download_webpage(url, clip_id, fatal=False)),
'thumbnails': self._extract_thumbnails(clip_data.get('preview')),
'timestamp': parse_iso8601(clip.get('created_at')),
'view_count': int_or_none(clip.get('clip_views_count')),
'formats': formats,
'subtitles': subtitles,
}