import re
import time
import itertools
+import json
from .common import InfoExtractor
from .naver import NaverBaseIE
-from ..compat import compat_str
+from ..compat import (
+ compat_HTTPError,
+ compat_str,
+)
from ..utils import (
ExtractorError,
+ int_or_none,
merge_dicts,
- remove_start,
try_get,
urlencode_postdata,
)
-class VLiveIE(NaverBaseIE):
+class VLiveBaseIE(NaverBaseIE):
+ _APP_ID = '8c6cc7b45d2568fb668be6e05b6e5a3b'
+
+
+class VLiveIE(VLiveBaseIE):
IE_NAME = 'vlive'
- _VALID_URL = r'https?://(?:(?:www|m)\.)?vlive\.tv/video/(?P<id>[0-9]+)'
+ _VALID_URL = r'https?://(?:(?:www|m)\.)?vlive\.tv/(?:video|embed)/(?P<id>[0-9]+)'
_NETRC_MACHINE = 'vlive'
_TESTS = [{
'url': 'http://www.vlive.tv/video/1326',
'info_dict': {
'id': '1326',
'ext': 'mp4',
- 'title': "[V LIVE] Girl's Day's Broadcast",
+ 'title': "Girl's Day's Broadcast",
'creator': "Girl's Day",
'view_count': int,
'uploader_id': 'muploader_a',
'info_dict': {
'id': '16937',
'ext': 'mp4',
- 'title': '[V LIVE] 첸백시 걍방',
+ 'title': '첸백시 걍방',
'creator': 'EXO',
'view_count': int,
'subtitles': 'mincount:12',
'subtitles': 'mincount:10',
},
'skip': 'This video is only available for CH+ subscribers',
+ }, {
+ 'url': 'https://www.vlive.tv/embed/1326',
+ 'only_matching': True,
}]
- @classmethod
- def suitable(cls, url):
- return False if VLivePlaylistIE.suitable(url) else super(VLiveIE, cls).suitable(url)
-
def _real_initialize(self):
self._login()
if not is_logged_in():
raise ExtractorError('Unable to log in', expected=True)
+ def _call_api(self, path_template, video_id, fields=None):
+ query = {'appId': self._APP_ID}
+ if fields:
+ query['fields'] = fields
+ return self._download_json(
+ 'https://www.vlive.tv/globalv-web/vam-web/' + path_template % video_id, video_id,
+ 'Downloading %s JSON metadata' % path_template.split('/')[-1].split('-')[0],
+ headers={'Referer': 'https://www.vlive.tv/'}, query=query)
+
def _real_extract(self, url):
video_id = self._match_id(url)
- webpage = self._download_webpage(
- 'https://www.vlive.tv/video/%s' % video_id, video_id)
-
- VIDEO_PARAMS_RE = r'\bvlive\.video\.init\(([^)]+)'
- VIDEO_PARAMS_FIELD = 'video params'
-
- params = self._parse_json(self._search_regex(
- VIDEO_PARAMS_RE, webpage, VIDEO_PARAMS_FIELD, default=''), video_id,
- transform_source=lambda s: '[' + s + ']', fatal=False)
-
- if not params or len(params) < 7:
- params = self._search_regex(
- VIDEO_PARAMS_RE, webpage, VIDEO_PARAMS_FIELD)
- params = [p.strip(r'"') for p in re.split(r'\s*,\s*', params)]
-
- status, long_video_id, key = params[2], params[5], params[6]
- status = remove_start(status, 'PRODUCT_')
-
- if status in ('LIVE_ON_AIR', 'BIG_EVENT_ON_AIR'):
- return self._live(video_id, webpage)
- elif status in ('VOD_ON_AIR', 'BIG_EVENT_INTRO'):
- return self._replay(video_id, webpage, long_video_id, key)
-
- if status == 'LIVE_END':
- raise ExtractorError('Uploading for replay. Please wait...',
- expected=True)
- elif status == 'COMING_SOON':
- raise ExtractorError('Coming soon!', expected=True)
- elif status == 'CANCELED':
- raise ExtractorError('We are sorry, '
- 'but the live broadcast has been canceled.',
- expected=True)
- elif status == 'ONLY_APP':
- raise ExtractorError('Unsupported video type', expected=True)
- else:
- raise ExtractorError('Unknown status %s' % status)
-
- def _get_common_fields(self, webpage):
- title = self._og_search_title(webpage)
- creator = self._html_search_regex(
- r'<div[^>]+class="info_area"[^>]*>\s*(?:<em[^>]*>.*?</em\s*>\s*)?<a\s+[^>]*>([^<]+)',
- webpage, 'creator', fatal=False)
- thumbnail = self._og_search_thumbnail(webpage)
- return {
- 'title': title,
- 'creator': creator,
- 'thumbnail': thumbnail,
- }
-
- def _live(self, video_id, webpage):
- init_page = self._download_init_page(video_id)
-
- live_params = self._search_regex(
- r'"liveStreamInfo"\s*:\s*(".*"),',
- init_page, 'live stream info')
- live_params = self._parse_json(live_params, video_id)
- live_params = self._parse_json(live_params, video_id)
-
- formats = []
- for vid in live_params.get('resolutions', []):
- formats.extend(self._extract_m3u8_formats(
- vid['cdnUrl'], video_id, 'mp4',
- m3u8_id=vid.get('name'),
- fatal=False, live=True))
- self._sort_formats(formats)
-
- info = self._get_common_fields(webpage)
- info.update({
- 'title': self._live_title(info['title']),
- 'id': video_id,
- 'formats': formats,
- 'is_live': True,
- })
- return info
-
- def _replay(self, video_id, webpage, long_video_id, key):
- if '' in (long_video_id, key):
- init_page = self._download_init_page(video_id)
- video_info = self._parse_json(self._search_regex(
- (r'(?s)oVideoStatus\s*=\s*({.+?})\s*</script',
- r'(?s)oVideoStatus\s*=\s*({.+})'), init_page, 'video info'),
- video_id)
- if video_info.get('status') == 'NEED_CHANNEL_PLUS':
- self.raise_login_required(
- 'This video is only available for CH+ subscribers')
- long_video_id, key = video_info['vid'], video_info['inkey']
-
- return merge_dicts(
- self._get_common_fields(webpage),
- self._extract_video_info(video_id, long_video_id, key))
-
- def _download_init_page(self, video_id):
- return self._download_webpage(
- 'https://www.vlive.tv/video/init/view',
- video_id, note='Downloading live webpage',
- data=urlencode_postdata({'videoSeq': video_id}),
- headers={
- 'Referer': 'https://www.vlive.tv/video/%s' % video_id,
- 'Content-Type': 'application/x-www-form-urlencoded'
- })
-
-
-class VLiveChannelIE(InfoExtractor):
+ try:
+ 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}')
+ 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())['message'])
+ raise
+
+ video = post['officialVideo']
+
+ 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')),
+ }
+
+ 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']
+ return merge_dicts(
+ get_common_fields(),
+ self._extract_video_info(video_id, vod_id, inkey))
+ 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': self._live_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 VLiveChannelIE(VLiveBaseIE):
IE_NAME = 'vlive:channel'
- _VALID_URL = r'https?://channels\.vlive\.tv/(?P<id>[0-9A-Z]+)'
- _TEST = {
+ _VALID_URL = r'https?://(?:channels\.vlive\.tv|(?:(?:www|m)\.)?vlive\.tv/channel)/(?P<id>[0-9A-Z]+)'
+ _TESTS = [{
'url': 'http://channels.vlive.tv/FCD4B',
'info_dict': {
'id': 'FCD4B',
'title': 'MAMAMOO',
},
'playlist_mincount': 110
- }
- _APP_ID = '8c6cc7b45d2568fb668be6e05b6e5a3b'
+ }, {
+ 'url': 'https://www.vlive.tv/channel/FCD4B',
+ 'only_matching': True,
+ }]
+
+ def _call_api(self, path, channel_key_suffix, channel_value, note, query):
+ q = {
+ 'app_id': self._APP_ID,
+ 'channel' + channel_key_suffix: channel_value,
+ }
+ q.update(query)
+ return self._download_json(
+ 'http://api.vfan.vlive.tv/vproxy/channelplus/' + path,
+ channel_value, note='Downloading ' + note, query=q)['result']
def _real_extract(self, url):
channel_code = self._match_id(url)
- webpage = self._download_webpage(
- 'http://channels.vlive.tv/%s/video' % channel_code, channel_code)
-
- app_id = None
-
- app_js_url = self._search_regex(
- r'<script[^>]+src=(["\'])(?P<url>http.+?/app\.js.*?)\1',
- webpage, 'app js', default=None, group='url')
-
- if app_js_url:
- app_js = self._download_webpage(
- app_js_url, channel_code, 'Downloading app JS', fatal=False)
- if app_js:
- app_id = self._search_regex(
- r'Global\.VFAN_APP_ID\s*=\s*[\'"]([^\'"]+)[\'"]',
- app_js, 'app id', default=None)
-
- app_id = app_id or self._APP_ID
-
- channel_info = self._download_json(
- 'http://api.vfan.vlive.tv/vproxy/channelplus/decodeChannelCode',
- channel_code, note='Downloading decode channel code',
- query={
- 'app_id': app_id,
- 'channelCode': channel_code,
- '_': int(time.time())
- })
+ channel_seq = self._call_api(
+ 'decodeChannelCode', 'Code', channel_code,
+ 'decode channel code', {})['channelSeq']
- channel_seq = channel_info['result']['channelSeq']
channel_name = None
entries = []
for page_num in itertools.count(1):
- video_list = self._download_json(
- 'http://api.vfan.vlive.tv/vproxy/channelplus/getChannelVideoList',
- channel_code, note='Downloading channel list page #%d' % page_num,
- query={
- 'app_id': app_id,
- 'channelSeq': channel_seq,
+ video_list = self._call_api(
+ 'getChannelVideoList', 'Seq', channel_seq,
+ 'channel list page #%d' % page_num, {
# Large values of maxNumOfRows (~300 or above) may cause
# empty responses (see [1]), e.g. this happens for [2] that
# has more than 300 videos.
# 1. https://github.com/ytdl-org/youtube-dl/issues/13830
# 2. http://channels.vlive.tv/EDBF.
'maxNumOfRows': 100,
- '_': int(time.time()),
'pageNo': page_num
}
)
if not channel_name:
channel_name = try_get(
video_list,
- lambda x: x['result']['channelInfo']['channelName'],
+ lambda x: x['channelInfo']['channelName'],
compat_str)
videos = try_get(
- video_list, lambda x: x['result']['videoList'], list)
+ video_list, lambda x: x['videoList'], list)
if not videos:
break
entries, channel_code, channel_name)
-class VLivePlaylistIE(InfoExtractor):
+# old extractor. Rewrite?
+
+class VLivePlaylistIE(VLiveBaseIE):
IE_NAME = 'vlive:playlist'
_VALID_URL = r'https?://(?:(?:www|m)\.)?vlive\.tv/video/(?P<video_id>[0-9]+)/playlist/(?P<id>[0-9]+)'
_VIDEO_URL_TEMPLATE = 'http://www.vlive.tv/video/%s'