1
0
mirror of https://codeberg.org/polarisfm/youtube-dl synced 2024-12-05 14:47:55 +01:00
youtube-dl/youtube_dl/extractor/vlive.py

368 lines
13 KiB
Python
Raw Normal View History

2015-08-20 06:56:11 +02:00
# coding: utf-8
from __future__ import unicode_literals
2015-08-20 06:56:11 +02:00
2016-04-26 16:30:24 +02:00
import re
2017-01-24 06:52:17 +01:00
import time
import itertools
2016-04-29 13:29:00 +02:00
2015-08-20 06:56:11 +02:00
from .common import InfoExtractor
from .naver import NaverBaseIE
from ..compat import compat_str
2015-08-20 06:56:11 +02:00
from ..utils import (
2016-04-29 13:29:00 +02:00
ExtractorError,
merge_dicts,
remove_start,
2017-01-28 13:08:01 +01:00
try_get,
urlencode_postdata,
2015-08-20 06:56:11 +02:00
)
class VLiveIE(NaverBaseIE):
2015-08-20 06:56:11 +02:00
IE_NAME = 'vlive'
_VALID_URL = r'https?://(?:(?:www|m)\.)?vlive\.tv/video/(?P<id>[0-9]+)'
_NETRC_MACHINE = 'vlive'
2016-11-15 16:11:47 +01:00
_TESTS = [{
'url': 'http://www.vlive.tv/video/1326',
2015-08-20 06:56:11 +02:00
'md5': 'cc7314812855ce56de70a06a27314983',
'info_dict': {
'id': '1326',
'ext': 'mp4',
'title': "[V LIVE] Girl's Day's Broadcast",
'creator': "Girl's Day",
'view_count': int,
'uploader_id': 'muploader_a',
2015-08-20 06:56:11 +02:00
},
2016-11-15 16:11:47 +01:00
}, {
'url': 'http://www.vlive.tv/video/16937',
'info_dict': {
'id': '16937',
'ext': 'mp4',
'title': '[V LIVE] 첸백시 걍방',
'creator': 'EXO',
'view_count': int,
'subtitles': 'mincount:12',
'uploader_id': 'muploader_j',
2016-11-15 16:11:47 +01:00
},
'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',
2016-11-15 16:11:47 +01:00
}]
2015-08-20 06:56:11 +02:00
2017-07-09 23:24:04 +02:00
@classmethod
def suitable(cls, url):
return False if VLivePlaylistIE.suitable(url) else super(VLiveIE, cls).suitable(url)
def _real_initialize(self):
self._login()
def _login(self):
email, password = self._get_login_info()
if None in (email, password):
return
def is_logged_in():
login_info = self._download_json(
'https://www.vlive.tv/auth/loginInfo', None,
note='Downloading login info',
headers={'Referer': 'https://www.vlive.tv/home'})
return try_get(
login_info, lambda x: x['message']['login'], bool) or False
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': email, 'pwd': password}),
headers={
'Referer': LOGIN_URL,
'Content-Type': 'application/x-www-form-urlencoded'
})
if not is_logged_in():
raise ExtractorError('Unable to log in', expected=True)
2015-08-20 06:56:11 +02:00
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(
2018-07-01 16:19:17 +02:00
'https://www.vlive.tv/video/%s' % video_id, video_id)
2015-08-20 06:56:11 +02:00
VIDEO_PARAMS_RE = r'\bvlive\.video\.init\(([^)]+)'
VIDEO_PARAMS_FIELD = 'video params'
2016-12-05 22:16:12 +01:00
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_')
2016-04-26 16:30:24 +02:00
2017-04-12 21:38:43 +02:00
if status in ('LIVE_ON_AIR', 'BIG_EVENT_ON_AIR'):
2016-12-05 22:16:12 +01:00
return self._live(video_id, webpage)
2017-04-12 21:38:43 +02:00
elif status in ('VOD_ON_AIR', 'BIG_EVENT_INTRO'):
return self._replay(video_id, webpage, long_video_id, key)
2016-04-26 16:30:24 +02:00
if status == 'LIVE_END':
raise ExtractorError('Uploading for replay. Please wait...',
expected=True)
elif status == 'COMING_SOON':
raise ExtractorError('Coming soon!', expected=True)
2016-04-26 16:30:24 +02:00
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)
2016-04-26 16:30:24 +02:00
else:
raise ExtractorError('Unknown status %s' % status)
def _get_common_fields(self, webpage):
2015-08-20 06:56:11 +02:00
title = self._og_search_title(webpage)
2016-04-26 16:30:24 +02:00
creator = self._html_search_regex(
r'<div[^>]+class="info_area"[^>]*>\s*(?:<em[^>]*>.*?</em\s*>\s*)?<a\s+[^>]*>([^<]+)',
2016-04-26 16:30:24 +02:00
webpage, 'creator', fatal=False)
thumbnail = self._og_search_thumbnail(webpage)
return {
'title': title,
'creator': creator,
'thumbnail': thumbnail,
}
2015-09-05 09:25:19 +02:00
2016-12-05 22:16:12 +01:00
def _live(self, video_id, webpage):
init_page = self._download_init_page(video_id)
2016-12-05 22:16:12 +01:00
live_params = self._search_regex(
r'"liveStreamInfo"\s*:\s*(".*"),',
init_page, 'live stream info')
2016-12-05 22:16:12 +01:00
live_params = self._parse_json(live_params, video_id)
live_params = self._parse_json(live_params, video_id)
2016-04-26 16:30:24 +02:00
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
2016-04-26 16:30:24 +02:00
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))
2017-01-24 06:52:17 +01:00
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'
})
2017-01-24 06:52:17 +01:00
class VLiveChannelIE(InfoExtractor):
IE_NAME = 'vlive:channel'
2017-01-28 13:08:01 +01:00
_VALID_URL = r'https?://channels\.vlive\.tv/(?P<id>[0-9A-Z]+)'
2017-01-24 06:52:17 +01:00
_TEST = {
2017-01-28 13:08:01 +01:00
'url': 'http://channels.vlive.tv/FCD4B',
2017-01-24 06:52:17 +01:00
'info_dict': {
'id': 'FCD4B',
'title': 'MAMAMOO',
},
'playlist_mincount': 110
}
_APP_ID = '8c6cc7b45d2568fb668be6e05b6e5a3b'
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)
2017-01-28 13:08:01 +01:00
app_id = None
2017-01-24 06:52:17 +01:00
app_js_url = self._search_regex(
2017-01-28 13:08:01 +01:00
r'<script[^>]+src=(["\'])(?P<url>http.+?/app\.js.*?)\1',
webpage, 'app js', default=None, group='url')
2017-01-24 06:52:17 +01:00
if app_js_url:
2017-01-28 13:08:01 +01:00
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
2017-01-24 06:52:17 +01:00
channel_info = self._download_json(
'http://api.vfan.vlive.tv/vproxy/channelplus/decodeChannelCode',
2017-01-28 13:08:01 +01:00
channel_code, note='Downloading decode channel code',
query={
'app_id': app_id,
'channelCode': channel_code,
'_': int(time.time())
})
2017-01-24 06:52:17 +01:00
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',
2017-01-28 13:08:01 +01:00
channel_code, note='Downloading channel list page #%d' % page_num,
2017-01-24 06:52:17 +01:00
query={
'app_id': app_id,
'channelSeq': channel_seq,
# 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.
2019-03-09 13:14:41 +01:00
# 1. https://github.com/ytdl-org/youtube-dl/issues/13830
# 2. http://channels.vlive.tv/EDBF.
'maxNumOfRows': 100,
2017-01-24 06:52:17 +01:00
'_': int(time.time()),
'pageNo': page_num
}
)
2017-01-28 13:08:01 +01:00
if not channel_name:
channel_name = try_get(
video_list,
lambda x: x['result']['channelInfo']['channelName'],
compat_str)
videos = try_get(
video_list, lambda x: x['result']['videoList'], list)
if not videos:
2017-01-24 06:52:17 +01:00
break
2017-01-28 13:08:01 +01:00
for video in videos:
video_id = video.get('videoSeq')
if not video_id:
continue
video_id = compat_str(video_id)
2017-01-24 06:52:17 +01:00
entries.append(
self.url_result(
2017-01-28 13:08:01 +01:00
'http://www.vlive.tv/video/%s' % video_id,
ie=VLiveIE.ie_key(), video_id=video_id))
2017-01-24 06:52:17 +01:00
return self.playlist_result(
entries, channel_code, channel_name)
2017-07-09 23:24:04 +02:00
class VLivePlaylistIE(InfoExtractor):
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'
_TESTS = [{
# regular working playlist
'url': 'https://www.vlive.tv/video/117956/playlist/117963',
'info_dict': {
'id': '117963',
'title': '아이돌룸(IDOL ROOM) 41회 - (여자)아이들'
},
'playlist_mincount': 10
}, {
# playlist with no playlistVideoSeqs
2017-07-09 23:24:04 +02:00
'url': 'http://www.vlive.tv/video/22867/playlist/22912',
'info_dict': {
'id': '22867',
'ext': 'mp4',
'title': '[V LIVE] Valentine Day Message from MINA',
'creator': 'TWICE',
'view_count': int
2017-07-09 23:24:04 +02:00
},
'params': {
'skip_download': True,
}
}]
def _build_video_result(self, video_id, message):
self.to_screen(message)
return self.url_result(
self._VIDEO_URL_TEMPLATE % video_id,
ie=VLiveIE.ie_key(), video_id=video_id)
2017-07-09 23:24:04 +02:00
def _real_extract(self, url):
2017-07-09 23:32:24 +02:00
mobj = re.match(self._VALID_URL, url)
video_id, playlist_id = mobj.group('video_id', 'id')
2017-07-09 23:24:04 +02:00
if self._downloader.params.get('noplaylist'):
return self._build_video_result(
video_id,
'Downloading just video %s because of --no-playlist'
% video_id)
2017-07-09 23:24:04 +02:00
self.to_screen(
2017-07-09 23:32:24 +02:00
'Downloading playlist %s - add --no-playlist to just download video'
% playlist_id)
2017-07-09 23:24:04 +02:00
webpage = self._download_webpage(
2017-07-09 23:32:24 +02:00
'http://www.vlive.tv/video/%s/playlist/%s'
% (video_id, playlist_id), playlist_id)
2017-07-09 23:24:04 +02:00
raw_item_ids = self._search_regex(
r'playlistVideoSeqs\s*=\s*(\[[^]]+\])', webpage,
'playlist video seqs', default=None, fatal=False)
if not raw_item_ids:
return self._build_video_result(
video_id,
'Downloading just video %s because no playlist was found'
% video_id)
item_ids = self._parse_json(raw_item_ids, playlist_id)
2017-07-09 23:24:04 +02:00
2017-07-09 23:32:24 +02:00
entries = [
self.url_result(
self._VIDEO_URL_TEMPLATE % item_id, ie=VLiveIE.ie_key(),
2017-07-09 23:32:24 +02:00
video_id=compat_str(item_id))
for item_id in item_ids]
2017-07-09 23:24:04 +02:00
2017-07-09 23:32:24 +02:00
playlist_name = self._html_search_regex(
r'<div[^>]+class="[^"]*multicam_playlist[^>]*>\s*<h3[^>]+>([^<]+)',
webpage, 'playlist title', fatal=False)
2017-07-09 23:24:04 +02:00
2017-07-09 23:32:24 +02:00
return self.playlist_result(entries, playlist_id, playlist_name)