[extractor/vlive] Replace with `VLiveWebArchiveIE` (#6196)

vlive has shut down: https://web.archive.org/web/20221031171019/https://www.vlive.tv/notice/4749

Authored by: seproDev
This commit is contained in:
sepro 2023-02-12 05:47:03 +01:00 committed by GitHub
parent a31d0fa6c3
commit b3eaab7ca2
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
4 changed files with 255 additions and 394 deletions

View File

@ -122,6 +122,7 @@ from .applepodcasts import ApplePodcastsIE
from .archiveorg import (
ArchiveOrgIE,
YoutubeWebArchiveIE,
VLiveWebArchiveIE,
)
from .arcpublishing import ArcPublishingIE
from .arkena import ArkenaIE
@ -2183,11 +2184,6 @@ from .vk import (
VKUserVideosIE,
VKWallPostIE,
)
from .vlive import (
VLiveIE,
VLivePostIE,
VLiveChannelIE,
)
from .vodlocker import VodlockerIE
from .vodpl import VODPlIE
from .vodplatform import VODPlatformIE

View File

@ -1,8 +1,10 @@
import json
import re
import urllib.error
import urllib.parse
from .common import InfoExtractor
from .naver import NaverBaseIE
from .youtube import YoutubeBaseInfoExtractor, YoutubeIE
from ..compat import compat_HTTPError, compat_urllib_parse_unquote
from ..utils import (
@ -945,3 +947,237 @@ class YoutubeWebArchiveIE(InfoExtractor):
if not info.get('title'):
info['title'] = video_id
return info
class VLiveWebArchiveIE(InfoExtractor):
IE_NAME = 'web.archive:vlive'
IE_DESC = 'web.archive.org saved vlive videos'
_VALID_URL = r'''(?x)
(?:https?://)?web\.archive\.org/
(?:web/)?(?:(?P<date>[0-9]{14})?[0-9A-Za-z_*]*/)? # /web and the version index is optional
(?:https?(?::|%3[Aa])//)?(?:
(?:(?:www|m)\.)?vlive\.tv(?::(?:80|443))?/(?:video|embed)/(?P<id>[0-9]+) # VLive URL
)
'''
_TESTS = [{
'url': 'https://web.archive.org/web/20221221144331/http://www.vlive.tv/video/1326',
'md5': 'cc7314812855ce56de70a06a27314983',
'info_dict': {
'id': '1326',
'ext': 'mp4',
'title': "Girl's Day's Broadcast",
'creator': "Girl's Day",
'view_count': int,
'uploader_id': 'muploader_a',
'uploader_url': None,
'uploader': None,
'upload_date': '20150817',
'thumbnail': r're:^https?://.*\.(?:jpg|png)$',
'timestamp': 1439816449,
'like_count': int,
'channel': 'Girl\'s Day',
'channel_id': 'FDF27',
'comment_count': int,
'release_timestamp': 1439818140,
'release_date': '20150817',
'duration': 1014,
},
'params': {
'skip_download': True,
},
}, {
'url': 'https://web.archive.org/web/20221221182103/http://www.vlive.tv/video/16937',
'info_dict': {
'id': '16937',
'ext': 'mp4',
'title': '첸백시 걍방',
'creator': 'EXO',
'view_count': int,
'subtitles': 'mincount:12',
'uploader_id': 'muploader_j',
'uploader_url': 'http://vlive.tv',
'uploader': None,
'upload_date': '20161112',
'thumbnail': r're:^https?://.*\.(?:jpg|png)$',
'timestamp': 1478923074,
'like_count': int,
'channel': 'EXO',
'channel_id': 'F94BD',
'comment_count': int,
'release_timestamp': 1478924280,
'release_date': '20161112',
'duration': 906,
},
'params': {
'skip_download': True,
},
}, {
'url': 'https://web.archive.org/web/20221127190050/http://www.vlive.tv/video/101870',
'info_dict': {
'id': '101870',
'ext': 'mp4',
'title': '[ⓓ xV] “레벨이들 매력에 반해? 안 반해?” 움직이는 HD 포토 (레드벨벳:Red Velvet)',
'creator': 'Dispatch',
'view_count': int,
'subtitles': 'mincount:6',
'uploader_id': 'V__FRA08071',
'uploader_url': 'http://vlive.tv',
'uploader': None,
'upload_date': '20181130',
'thumbnail': r're:^https?://.*\.(?:jpg|png)$',
'timestamp': 1543601327,
'like_count': int,
'channel': 'Dispatch',
'channel_id': 'C796F3',
'comment_count': int,
'release_timestamp': 1543601040,
'release_date': '20181130',
'duration': 279,
},
'params': {
'skip_download': True,
},
}]
# The wayback machine has special timestamp and "mode" values:
# timestamp:
# 1 = the first capture
# 2 = the last capture
# mode:
# id_ = Identity - perform no alterations of the original resource, return it as it was archived.
_WAYBACK_BASE_URL = 'https://web.archive.org/web/2id_/'
def _download_archived_page(self, url, video_id, *, timestamp='2', **kwargs):
for retry in self.RetryManager():
try:
return self._download_webpage(f'https://web.archive.org/web/{timestamp}id_/{url}', video_id, **kwargs)
except ExtractorError as e:
if isinstance(e.cause, urllib.error.HTTPError) and e.cause.code == 404:
raise ExtractorError('Page was not archived', expected=True)
retry.error = e
continue
def _download_archived_json(self, url, video_id, **kwargs):
page = self._download_archived_page(url, video_id, **kwargs)
if not page:
raise ExtractorError('Page was not archived', expected=True)
else:
return self._parse_json(page, video_id)
def _extract_formats_from_m3u8(self, m3u8_url, params, video_id):
m3u8_doc = self._download_archived_page(m3u8_url, video_id, note='Downloading m3u8', query=params, fatal=False)
if not m3u8_doc:
return
# M3U8 document should be changed to archive domain
m3u8_doc = m3u8_doc.splitlines()
url_base = m3u8_url.rsplit('/', 1)[0]
first_segment = None
for i, line in enumerate(m3u8_doc):
if not line.startswith('#'):
m3u8_doc[i] = f'{self._WAYBACK_BASE_URL}{url_base}/{line}?{urllib.parse.urlencode(params)}'
first_segment = first_segment or m3u8_doc[i]
# Segments may not have been archived. See https://web.archive.org/web/20221127190050/http://www.vlive.tv/video/101870
urlh = self._request_webpage(HEADRequest(first_segment), video_id, errnote=False,
fatal=False, note='Check first segment availablity')
if urlh:
formats, subtitles = self._parse_m3u8_formats_and_subtitles('\n'.join(m3u8_doc), ext='mp4', video_id=video_id)
if subtitles:
self._report_ignoring_subs('m3u8')
return formats
# Closely follows the logic of the ArchiveTeam grab script
# See: https://github.com/ArchiveTeam/vlive-grab/blob/master/vlive.lua
def _real_extract(self, url):
video_id, url_date = self._match_valid_url(url).group('id', 'date')
webpage = self._download_archived_page(f'https://www.vlive.tv/video/{video_id}', video_id, timestamp=url_date)
player_info = self._search_json(r'__PRELOADED_STATE__\s*=', webpage, 'player info', video_id)
user_country = traverse_obj(player_info, ('common', 'userCountry'))
main_script_url = self._search_regex(r'<script\s+src="([^"]+/js/main\.[^"]+\.js)"', webpage, 'main script url')
main_script = self._download_archived_page(main_script_url, video_id, note='Downloading main script')
app_id = self._search_regex(r'appId\s*=\s*"([^"]+)"', main_script, 'app id')
inkey = self._download_archived_json(
f'https://www.vlive.tv/globalv-web/vam-web/video/v1.0/vod/{video_id}/inkey', video_id, note='Fetching inkey', query={
'appId': app_id,
'platformType': 'PC',
'gcc': user_country,
'locale': 'en_US',
}, fatal=False)
vod_id = traverse_obj(player_info, ('postDetail', 'post', 'officialVideo', 'vodId'))
vod_data = self._download_archived_json(
f'https://apis.naver.com/rmcnmv/rmcnmv/vod/play/v2.0/{vod_id}', video_id, note='Fetching vod data', query={
'key': inkey.get('inkey'),
'pid': 'rmcPlayer_16692457559726800', # partially unix time and partially random. Fixed value used by archiveteam project
'sid': '2024',
'ver': '2.0',
'devt': 'html5_pc',
'doct': 'json',
'ptc': 'https',
'sptc': 'https',
'cpt': 'vtt',
'ctls': '%7B%22visible%22%3A%7B%22fullscreen%22%3Atrue%2C%22logo%22%3Afalse%2C%22playbackRate%22%3Afalse%2C%22scrap%22%3Afalse%2C%22playCount%22%3Atrue%2C%22commentCount%22%3Atrue%2C%22title%22%3Atrue%2C%22writer%22%3Atrue%2C%22expand%22%3Afalse%2C%22subtitles%22%3Atrue%2C%22thumbnails%22%3Atrue%2C%22quality%22%3Atrue%2C%22setting%22%3Atrue%2C%22script%22%3Afalse%2C%22logoDimmed%22%3Atrue%2C%22badge%22%3Atrue%2C%22seekingTime%22%3Atrue%2C%22muted%22%3Atrue%2C%22muteButton%22%3Afalse%2C%22viewerNotice%22%3Afalse%2C%22linkCount%22%3Afalse%2C%22createTime%22%3Afalse%2C%22thumbnail%22%3Atrue%7D%2C%22clicked%22%3A%7B%22expand%22%3Afalse%2C%22subtitles%22%3Afalse%7D%7D',
'pv': '4.26.9',
'dr': '1920x1080',
'cpl': 'en_US',
'lc': 'en_US',
'adi': '%5B%7B%22type%22%3A%22pre%22%2C%22exposure%22%3Afalse%2C%22replayExposure%22%3Afalse%7D%5D',
'adu': '%2F',
'videoId': vod_id,
'cc': user_country,
})
formats = []
streams = traverse_obj(vod_data, ('streams', ...))
if len(streams) > 1:
self.report_warning('Multiple streams found. Only the first stream will be downloaded.')
stream = streams[0]
max_stream = max(
stream.get('videos') or [],
key=lambda v: traverse_obj(v, ('bitrate', 'video'), default=0), default=None)
if max_stream is not None:
params = {arg.get('name'): arg.get('value') for arg in stream.get('keys', []) if arg.get('type') == 'param'}
formats = self._extract_formats_from_m3u8(max_stream.get('source'), params, video_id) or []
# For parts of the project MP4 files were archived
max_video = max(
traverse_obj(vod_data, ('videos', 'list', ...)),
key=lambda v: traverse_obj(v, ('bitrate', 'video'), default=0), default=None)
if max_video is not None:
video_url = self._WAYBACK_BASE_URL + max_video.get('source')
urlh = self._request_webpage(HEADRequest(video_url), video_id, errnote=False,
fatal=False, note='Check video availablity')
if urlh:
formats.append({'url': video_url})
return {
'id': video_id,
'formats': formats,
**traverse_obj(player_info, ('postDetail', 'post', {
'title': ('officialVideo', 'title', {str}),
'creator': ('author', 'nickname', {str}),
'channel': ('channel', 'channelName', {str}),
'channel_id': ('channel', 'channelCode', {str}),
'duration': ('officialVideo', 'playTime', {int_or_none}),
'view_count': ('officialVideo', 'playCount', {int_or_none}),
'like_count': ('officialVideo', 'likeCount', {int_or_none}),
'comment_count': ('officialVideo', 'commentCount', {int_or_none}),
'timestamp': ('officialVideo', 'createdAt', {lambda x: int_or_none(x, scale=1000)}),
'release_timestamp': ('officialVideo', 'willStartAt', {lambda x: int_or_none(x, scale=1000)}),
})),
**traverse_obj(vod_data, ('meta', {
'uploader_id': ('user', 'id', {str}),
'uploader': ('user', 'name', {str}),
'uploader_url': ('user', 'url', {url_or_none}),
'thumbnail': ('cover', 'source', {url_or_none}),
}), expected_type=lambda x: x or None),
**NaverBaseIE.process_subtitles(vod_data, lambda x: [self._WAYBACK_BASE_URL + x]),
}

View File

@ -21,6 +21,23 @@ from ..utils import (
class NaverBaseIE(InfoExtractor):
_CAPTION_EXT_RE = r'\.(?:ttml|vtt)'
@staticmethod # NB: Used in VLiveWebArchiveIE
def process_subtitles(vod_data, process_url):
ret = {'subtitles': {}, 'automatic_captions': {}}
for caption in traverse_obj(vod_data, ('captions', 'list', ...)):
caption_url = caption.get('source')
if not caption_url:
continue
type_ = 'automatic_captions' if caption.get('type') == 'auto' else 'subtitles'
lang = caption.get('locale') or join_nonempty('language', 'country', from_dict=caption) or 'und'
if caption.get('type') == 'fan':
lang += '_fan%d' % next(i for i in itertools.count(1) if f'{lang}_fan{i}' not in ret[type_])
ret[type_].setdefault(lang, []).extend({
'url': sub_url,
'name': join_nonempty('label', 'fanName', from_dict=caption, delim=' - '),
} for sub_url in process_url(caption_url))
return ret
def _extract_video_info(self, video_id, vid, key):
video_data = self._download_json(
'http://play.rmcnmv.naver.com/vod/play/v2.0/' + vid,
@ -79,34 +96,18 @@ class NaverBaseIE(InfoExtractor):
]
return [caption_url]
automatic_captions = {}
subtitles = {}
for caption in get_list('caption'):
caption_url = caption.get('source')
if not caption_url:
continue
sub_dict = automatic_captions if caption.get('type') == 'auto' else subtitles
lang = caption.get('locale') or join_nonempty('language', 'country', from_dict=caption) or 'und'
if caption.get('type') == 'fan':
lang += '_fan%d' % next(i for i in itertools.count(1) if f'{lang}_fan{i}' not in sub_dict)
sub_dict.setdefault(lang, []).extend({
'url': sub_url,
'name': join_nonempty('label', 'fanName', from_dict=caption, delim=' - '),
} for sub_url in get_subs(caption_url))
user = meta.get('user', {})
return {
'id': video_id,
'title': title,
'formats': formats,
'subtitles': subtitles,
'automatic_captions': automatic_captions,
'thumbnail': try_get(meta, lambda x: x['cover']['source']),
'view_count': int_or_none(meta.get('count')),
'uploader_id': user.get('id'),
'uploader': user.get('name'),
'uploader_url': user.get('url'),
**self.process_subtitles(video_data, get_subs),
}

View File

@ -1,372 +0,0 @@
import itertools
import json
from .naver import NaverBaseIE
from ..compat import (
compat_HTTPError,
compat_str,
)
from ..utils import (
ExtractorError,
int_or_none,
LazyList,
merge_dicts,
str_or_none,
strip_or_none,
traverse_obj,
try_get,
urlencode_postdata,
url_or_none,
)
class VLiveBaseIE(NaverBaseIE):
_NETRC_MACHINE = 'vlive'
_logged_in = False
def _perform_login(self, username, password):
if self._logged_in:
return
LOGIN_URL = 'https://www.vlive.tv/auth/email/login'
self._request_webpage(
LOGIN_URL, None, note='Downloading login cookies')
self._download_webpage(
LOGIN_URL, None, note='Logging in',
data=urlencode_postdata({'email': username, 'pwd': password}),
headers={
'Referer': LOGIN_URL,
'Content-Type': 'application/x-www-form-urlencoded'
})
login_info = self._download_json(
'https://www.vlive.tv/auth/loginInfo', None,
note='Checking login status',
headers={'Referer': 'https://www.vlive.tv/home'})
if not try_get(login_info, lambda x: x['message']['login'], bool):
raise ExtractorError('Unable to log in', expected=True)
VLiveBaseIE._logged_in = True
def _call_api(self, path_template, video_id, fields=None, query_add={}, note=None):
if note is None:
note = 'Downloading %s JSON metadata' % path_template.split('/')[-1].split('-')[0]
query = {'appId': '8c6cc7b45d2568fb668be6e05b6e5a3b', 'gcc': 'KR', 'platformType': 'PC'}
if fields:
query['fields'] = fields
if query_add:
query.update(query_add)
try:
return self._download_json(
'https://www.vlive.tv/globalv-web/vam-web/' + path_template % video_id, video_id,
note, headers={'Referer': 'https://www.vlive.tv/'}, query=query)
except ExtractorError as e:
if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
self.raise_login_required(json.loads(e.cause.read().decode('utf-8'))['message'])
raise
class VLiveIE(VLiveBaseIE):
IE_NAME = 'vlive'
_VALID_URL = r'https?://(?:(?:www|m)\.)?vlive\.tv/(?:video|embed)/(?P<id>[0-9]+)'
_TESTS = [{
'url': 'http://www.vlive.tv/video/1326',
'md5': 'cc7314812855ce56de70a06a27314983',
'info_dict': {
'id': '1326',
'ext': 'mp4',
'title': "Girl's Day's Broadcast",
'creator': "Girl's Day",
'view_count': int,
'uploader_id': 'muploader_a',
'upload_date': '20150817',
'thumbnail': r're:^https?://.*\.(?:jpg|png)$',
'timestamp': 1439816449,
'like_count': int,
'channel': 'Girl\'s Day',
'channel_id': 'FDF27',
'comment_count': int,
'release_timestamp': 1439818140,
'release_date': '20150817',
'duration': 1014,
},
'params': {
'skip_download': True,
},
}, {
'url': 'http://www.vlive.tv/video/16937',
'info_dict': {
'id': '16937',
'ext': 'mp4',
'title': '첸백시 걍방',
'creator': 'EXO',
'view_count': int,
'subtitles': 'mincount:12',
'uploader_id': 'muploader_j',
'upload_date': '20161112',
'thumbnail': r're:^https?://.*\.(?:jpg|png)$',
'timestamp': 1478923074,
'like_count': int,
'channel': 'EXO',
'channel_id': 'F94BD',
'comment_count': int,
'release_timestamp': 1478924280,
'release_date': '20161112',
'duration': 906,
},
'params': {
'skip_download': True,
},
}, {
'url': 'https://www.vlive.tv/video/129100',
'md5': 'ca2569453b79d66e5b919e5d308bff6b',
'info_dict': {
'id': '129100',
'ext': 'mp4',
'title': '[V LIVE] [BTS+] Run BTS! 2019 - EP.71 :: Behind the scene',
'creator': 'BTS+',
'view_count': int,
'subtitles': 'mincount:10',
},
'skip': 'This video is only available for CH+ subscribers',
}, {
'url': 'https://www.vlive.tv/embed/1326',
'only_matching': True,
}, {
# works only with gcc=KR
'url': 'https://www.vlive.tv/video/225019',
'only_matching': True,
}, {
'url': 'https://www.vlive.tv/video/223906',
'info_dict': {
'id': '58',
'title': 'RUN BTS!'
},
'playlist_mincount': 120
}]
def _real_extract(self, url):
video_id = self._match_id(url)
post = self._call_api(
'post/v1.0/officialVideoPost-%s', video_id,
'author{nickname},channel{channelCode,channelName},officialVideo{commentCount,exposeStatus,likeCount,playCount,playTime,status,title,type,vodId},playlist{playlistSeq,totalCount,name}')
playlist_id = str_or_none(try_get(post, lambda x: x['playlist']['playlistSeq']))
if not self._yes_playlist(playlist_id, video_id):
video = post['officialVideo']
return self._get_vlive_info(post, video, video_id)
playlist_name = str_or_none(try_get(post, lambda x: x['playlist']['name']))
playlist_count = str_or_none(try_get(post, lambda x: x['playlist']['totalCount']))
playlist = self._call_api(
'playlist/v1.0/playlist-%s/posts', playlist_id, 'data', {'limit': playlist_count})
entries = []
for video_data in playlist['data']:
video = video_data.get('officialVideo')
video_id = str_or_none(video.get('videoSeq'))
entries.append(self._get_vlive_info(video_data, video, video_id))
return self.playlist_result(entries, playlist_id, playlist_name)
def _get_vlive_info(self, post, video, video_id):
def get_common_fields():
channel = post.get('channel') or {}
return {
'title': video.get('title'),
'creator': post.get('author', {}).get('nickname'),
'channel': channel.get('channelName'),
'channel_id': channel.get('channelCode'),
'duration': int_or_none(video.get('playTime')),
'view_count': int_or_none(video.get('playCount')),
'like_count': int_or_none(video.get('likeCount')),
'comment_count': int_or_none(video.get('commentCount')),
'timestamp': int_or_none(video.get('createdAt'), scale=1000),
'release_timestamp': int_or_none(traverse_obj(video, 'onAirStartAt', 'willStartAt'), scale=1000),
'thumbnail': video.get('thumb'),
}
video_type = video.get('type')
if video_type == 'VOD':
inkey = self._call_api('video/v1.0/vod/%s/inkey', video_id)['inkey']
vod_id = video['vodId']
info_dict = merge_dicts(
get_common_fields(),
self._extract_video_info(video_id, vod_id, inkey))
thumbnail = video.get('thumb')
if thumbnail:
if not info_dict.get('thumbnails') and info_dict.get('thumbnail'):
info_dict['thumbnails'] = [{'url': info_dict.pop('thumbnail')}]
info_dict.setdefault('thumbnails', []).append({'url': thumbnail, 'preference': 1})
return info_dict
elif video_type == 'LIVE':
status = video.get('status')
if status == 'ON_AIR':
stream_url = self._call_api(
'old/v3/live/%s/playInfo',
video_id)['result']['adaptiveStreamUrl']
formats = self._extract_m3u8_formats(stream_url, video_id, 'mp4')
info = get_common_fields()
info.update({
'title': video['title'],
'id': video_id,
'formats': formats,
'is_live': True,
})
return info
elif status == 'ENDED':
raise ExtractorError(
'Uploading for replay. Please wait...', expected=True)
elif status == 'RESERVED':
raise ExtractorError('Coming soon!', expected=True)
elif video.get('exposeStatus') == 'CANCEL':
raise ExtractorError(
'We are sorry, but the live broadcast has been canceled.',
expected=True)
else:
raise ExtractorError('Unknown status ' + status)
class VLivePostIE(VLiveBaseIE):
IE_NAME = 'vlive:post'
_VALID_URL = r'https?://(?:(?:www|m)\.)?vlive\.tv/post/(?P<id>\d-\d+)'
_TESTS = [{
# uploadType = SOS
'url': 'https://www.vlive.tv/post/1-20088044',
'info_dict': {
'id': '1-20088044',
'title': 'Hola estrellitas la tierra les dice hola (si era así no?) Ha...',
'description': 'md5:fab8a1e50e6e51608907f46c7fa4b407',
},
'playlist_count': 3,
}, {
# uploadType = V
'url': 'https://www.vlive.tv/post/1-20087926',
'info_dict': {
'id': '1-20087926',
'title': 'James Corden: And so, the baby becamos the Papa💜😭💪😭',
},
'playlist_count': 1,
}]
_FVIDEO_TMPL = 'fvideo/v1.0/fvideo-%%s/%s'
def _real_extract(self, url):
post_id = self._match_id(url)
post = self._call_api(
'post/v1.0/post-%s', post_id,
'attachments{video},officialVideo{videoSeq},plainBody,title')
video_seq = str_or_none(try_get(
post, lambda x: x['officialVideo']['videoSeq']))
if video_seq:
return self.url_result(
'http://www.vlive.tv/video/' + video_seq,
VLiveIE.ie_key(), video_seq)
title = post['title']
entries = []
for idx, video in enumerate(post['attachments']['video'].values()):
video_id = video.get('videoId')
if not video_id:
continue
upload_type = video.get('uploadType')
upload_info = video.get('uploadInfo') or {}
entry = None
if upload_type == 'SOS':
download = self._call_api(
self._FVIDEO_TMPL % 'sosPlayInfo', video_id)['videoUrl']['download']
formats = []
for f_id, f_url in download.items():
formats.append({
'format_id': f_id,
'url': f_url,
'height': int_or_none(f_id[:-1]),
})
entry = {
'formats': formats,
'id': video_id,
'thumbnail': upload_info.get('imageUrl'),
}
elif upload_type == 'V':
vod_id = upload_info.get('videoId')
if not vod_id:
continue
inkey = self._call_api(self._FVIDEO_TMPL % 'inKey', video_id)['inKey']
entry = self._extract_video_info(video_id, vod_id, inkey)
if entry:
entry['title'] = '%s_part%s' % (title, idx)
entries.append(entry)
return self.playlist_result(
entries, post_id, title, strip_or_none(post.get('plainBody')))
class VLiveChannelIE(VLiveBaseIE):
IE_NAME = 'vlive:channel'
_VALID_URL = r'https?://(?:channels\.vlive\.tv|(?:(?:www|m)\.)?vlive\.tv/channel)/(?P<channel_id>[0-9A-Z]+)(?:/board/(?P<posts_id>\d+))?'
_TESTS = [{
'url': 'http://channels.vlive.tv/FCD4B',
'info_dict': {
'id': 'FCD4B',
'title': 'MAMAMOO',
},
'playlist_mincount': 110
}, {
'url': 'https://www.vlive.tv/channel/FCD4B',
'only_matching': True,
}, {
'url': 'https://www.vlive.tv/channel/FCD4B/board/3546',
'info_dict': {
'id': 'FCD4B-3546',
'title': 'MAMAMOO - Star Board',
},
'playlist_mincount': 880
}]
def _entries(self, posts_id, board_name):
if board_name:
posts_path = 'post/v1.0/board-%s/posts'
query_add = {'limit': 100, 'sortType': 'LATEST'}
else:
posts_path = 'post/v1.0/channel-%s/starPosts'
query_add = {'limit': 100}
for page_num in itertools.count(1):
video_list = self._call_api(
posts_path, posts_id, 'channel{channelName},contentType,postId,title,url', query_add,
note=f'Downloading playlist page {page_num}')
for video in try_get(video_list, lambda x: x['data'], list) or []:
video_id = str(video.get('postId'))
video_title = str_or_none(video.get('title'))
video_url = url_or_none(video.get('url'))
if not all((video_id, video_title, video_url)) or video.get('contentType') != 'VIDEO':
continue
channel_name = try_get(video, lambda x: x['channel']['channelName'], compat_str)
yield self.url_result(video_url, VLivePostIE.ie_key(), video_id, video_title, channel=channel_name)
after = try_get(video_list, lambda x: x['paging']['nextParams']['after'], compat_str)
if not after:
break
query_add['after'] = after
def _real_extract(self, url):
channel_id, posts_id = self._match_valid_url(url).groups()
board_name = None
if posts_id:
board = self._call_api(
'board/v1.0/board-%s', posts_id, 'title,boardType')
board_name = board.get('title') or 'Unknown'
if board.get('boardType') not in ('STAR', 'VLIVE_PLUS'):
raise ExtractorError(f'Board {board_name!r} is not supported', expected=True)
entries = LazyList(self._entries(posts_id or channel_id, board_name))
channel_name = entries[0]['channel']
return self.playlist_result(
entries,
f'{channel_id}-{posts_id}' if posts_id else channel_id,
f'{channel_name} - {board_name}' if channel_name and board_name else channel_name)