Mirroring
/
yt-dlc
mirror of https://github.com/blackjack4494/yt-dlc
1
0
Fork 0
Code Issues Releases Wiki Activity

[vlive] merge youtube-dl

Browse Source
pull/248/head
Kyu Yeun Kim 5 years ago
parent d052b9a112
commit 452ec863a3
2 changed files with 108 additions and 253 deletions
Show Stats Download Patch File Download Diff File

1
youtube_dlc/extractor/extractors.py
Unescape Escape View File

@ -1388,7 +1388,6 @@ from .vk import (
from .vlive import ( from .vlive import (
VLiveIE, VLiveIE,
VLiveChannelIE, VLiveChannelIE,
VLivePlaylistIE
) )
from .vodlocker import VodlockerIE from .vodlocker import VodlockerIE
from .vodpl import VODPlIE from .vodpl import VODPlIE

340
youtube_dlc/extractor/vlive.py
Unescape Escape View File

@ -1,55 +1,48 @@
# coding: utf-8 # coding: utf-8
from __future__ import unicode_literals from __future__ import unicode_literals
import re
import time
import itertools import itertools
import json
from .common import InfoExtractor
from .naver import NaverBaseIE from .naver import NaverBaseIE
from ..compat import compat_str from ..compat import (
compat_HTTPError,
compat_str,
)
from ..utils import ( from ..utils import (
ExtractorError, ExtractorError,
int_or_none,
merge_dicts, merge_dicts,
try_get, try_get,
urlencode_postdata, urlencode_postdata,
) )
class VLiveIE(NaverBaseIE): class VLiveBaseIE(NaverBaseIE):
_APP_ID = '8c6cc7b45d2568fb668be6e05b6e5a3b'
class VLiveIE(VLiveBaseIE):
IE_NAME = 'vlive' IE_NAME = 'vlive'
_VALID_URL = r'https?://(?:(?:www|m)\.)?vlive\.tv/(?:video|post)/(?P<id>(?:\d-)?[0-9]+)' _VALID_URL = r'https?://(?:(?:www|m)\.)?vlive\.tv/(?:video|embed)/(?P<id>[0-9]+)'
_NETRC_MACHINE = 'vlive' _NETRC_MACHINE = 'vlive'
_TESTS = [{ _TESTS = [{
'url': 'https://www.vlive.tv/video/1326', 'url': 'http://www.vlive.tv/video/1326',
'md5': 'cc7314812855ce56de70a06a27314983', 'md5': 'cc7314812855ce56de70a06a27314983',
'info_dict': { 'info_dict': {
'id': '1326', 'id': '1326',
'ext': 'mp4', 'ext': 'mp4',
'title': "[V LIVE] Girl's Day's Broadcast", 'title': "Girl's Day's Broadcast",
'creator': "Girl's Day", 'creator': "Girl's Day",
'view_count': int, 'view_count': int,
'uploader_id': 'muploader_a', 'uploader_id': 'muploader_a',
}, },
}, }, {
{ 'url': 'http://www.vlive.tv/video/16937',
'url': 'https://vlive.tv/post/1-18244258',
'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',
},
},
{
'url': 'https://www.vlive.tv/video/16937',
'info_dict': { 'info_dict': {
'id': '16937', 'id': '16937',
'ext': 'mp4', 'ext': 'mp4',
'title': '[V LIVE] 첸백시 걍방', 'title': '첸백시 걍방',
'creator': 'EXO', 'creator': 'EXO',
'view_count': int, 'view_count': int,
'subtitles': 'mincount:12', 'subtitles': 'mincount:12',
@ -70,12 +63,11 @@ class VLiveIE(NaverBaseIE):
'subtitles': 'mincount:10', 'subtitles': 'mincount:10',
}, },
'skip': 'This video is only available for CH+ subscribers', '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): def _real_initialize(self):
self._login() self._login()
@ -107,118 +99,82 @@ class VLiveIE(NaverBaseIE):
if not is_logged_in(): if not is_logged_in():
raise ExtractorError('Unable to log in', expected=True) raise ExtractorError('Unable to log in', expected=True)
def _real_extract(self, url): def _call_api(self, path_template, video_id, fields=None):
# url may match on a post or a video url with a post_id potentially matching a video_id query = {'appId': self._APP_ID}
working_id = self._match_id(url) if fields:
webpage = self._download_webpage(url, working_id) query['fields'] = fields
return self._download_json(
PARAMS_RE = r'window\.__PRELOADED_STATE__\s*=\s*({.*});?\s*</script>' 'https://www.vlive.tv/globalv-web/vam-web/' + path_template % video_id, video_id,
PARAMS_FIELD = 'params' 'Downloading %s JSON metadata' % path_template.split('/')[-1].split('-')[0],
headers={'Referer': 'https://www.vlive.tv/'}, query=query)
params = self._search_regex(
PARAMS_RE, webpage, PARAMS_FIELD, default='', flags=re.DOTALL)
params = self._parse_json(params, working_id, fatal=False)
video_params = try_get(params, lambda x: x["postDetail"]["post"]["officialVideo"], dict)
if video_params is None: def _real_extract(self, url):
error = try_get(params, lambda x: x["postDetail"]["error"], dict) video_id = self._match_id(url)
error_data = try_get(error, lambda x: x["data"], dict)
error_video = try_get(error_data, lambda x: x["officialVideo"], dict)
error_msg = try_get(error, lambda x: x["message"], compat_str)
product_type = try_get(error_data,
[lambda x: x["officialVideo"]["productType"],
lambda x: x["board"]["boardType"]],
compat_str)
if error_video is not None:
if product_type in ('VLIVE_PLUS', 'VLIVE+'):
self.raise_login_required('This video is only available with V LIVE+.')
elif error_msg is not None:
raise ExtractorError('V LIVE reported the following error: %s' % error_msg)
else:
raise ExtractorError('Failed to extract video parameters.')
elif 'post' in url:
raise ExtractorError('Url does not appear to be a video post.', expected=True)
else:
raise ExtractorError('Failed to extract video parameters.')
video_id = working_id if 'video' in url else str(video_params["videoSeq"]) 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_type = video_params["type"] video = post['officialVideo']
if video_type in ('VOD'):
encoding_status = video_params["encodingStatus"]
if encoding_status == 'COMPLETE':
return self._replay(video_id, webpage, params, video_params)
else:
raise ExtractorError('VOD encoding not yet complete. Please try again later.',
expected=True)
elif video_type in ('LIVE'):
video_status = video_params["status"]
if video_status in ('RESERVED'):
raise ExtractorError('Coming soon!', expected=True)
elif video_status in ('ENDED', 'END'):
raise ExtractorError('Uploading for replay. Please wait...', expected=True)
else:
return self._live(video_id, webpage, params)
else:
raise ExtractorError('Unknown video type %s' % video_type)
def _get_common_fields(self, webpage, params): def get_common_fields():
title = self._og_search_title(webpage) channel = post.get('channel') or {}
description = self._html_search_meta(
['og:description', 'description', 'twitter:description'],
webpage, 'description', default=None)
creator = (try_get(params, lambda x: x["channel"]["channel"]["channelName"], compat_str)
or self._search_regex(r'on (.*) channel', description or '', 'creator', fatal=False))
thumbnail = self._og_search_thumbnail(webpage)
return { return {
'title': title, 'title': video.get('title'),
'creator': creator, 'creator': post.get('author', {}).get('nickname'),
'thumbnail': thumbnail, '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')),
} }
def _live(self, video_id, webpage, params): video_type = video.get('type')
LIVE_INFO_ENDPOINT = 'https://www.vlive.tv/globalv-web/vam-web/old/v3/live/%s/playInfo' % video_id if video_type == 'VOD':
play_info = self._download_json(LIVE_INFO_ENDPOINT, video_id, inkey = self._call_api('video/v1.0/vod/%s/inkey', video_id)['inkey']
headers={"referer": "https://www.vlive.tv"}) vod_id = video['vodId']
return merge_dicts(
streams = try_get(play_info, lambda x: x["result"]["streamList"], list) or [] get_common_fields(),
self._extract_video_info(video_id, vod_id, inkey))
formats = [] elif video_type == 'LIVE':
for stream in streams: status = video.get('status')
formats.extend(self._extract_m3u8_formats( if status == 'ON_AIR':
stream['serviceUrl'], video_id, 'mp4', stream_url = self._call_api(
fatal=False, live=True)) 'old/v3/live/%s/playInfo',
self._sort_formats(formats) video_id)['result']['adaptiveStreamUrl']
formats = self._extract_m3u8_formats(stream_url, video_id, 'mp4')
info = self._get_common_fields(webpage, params) info = get_common_fields()
info.update({ info.update({
'title': self._live_title(info['title']), 'title': self._live_title(video['title']),
'id': video_id, 'id': video_id,
'formats': formats, 'formats': formats,
'is_live': True, 'is_live': True,
}) })
return info return info
elif status == 'ENDED':
def _replay(self, video_id, webpage, params, video_params): raise ExtractorError(
long_video_id = video_params["vodId"] 'Uploading for replay. Please wait...', expected=True)
elif status == 'RESERVED':
VOD_KEY_ENDPOINT = 'https://www.vlive.tv/globalv-web/vam-web/video/v1.0/vod/%s/inkey' % video_id raise ExtractorError('Coming soon!', expected=True)
key_json = self._download_json(VOD_KEY_ENDPOINT, video_id, elif video.get('exposeStatus') == 'CANCEL':
headers={"referer": "https://www.vlive.tv"}) raise ExtractorError(
key = key_json["inkey"] 'We are sorry, but the live broadcast has been canceled.',
expected=True)
return merge_dicts( else:
self._get_common_fields(webpage, params), raise ExtractorError('Unknown status ' + status)
self._extract_video_info(video_id, long_video_id, key))
class VLiveChannelIE(InfoExtractor): class VLiveChannelIE(VLiveBaseIE):
IE_NAME = 'vlive:channel' IE_NAME = 'vlive:channel'
_VALID_URL = r'https?://(?:(?:www|m)\.)?(?:channels\.vlive\.tv/|vlive\.tv/channels?/)(?P<id>[0-9A-Z]+)' _VALID_URL = r'https?://(?:channels\.vlive\.tv|(?:(?:www|m)\.)?vlive\.tv/channel)/(?P<id>[0-9A-Z]+)'
_TESTS = [{ _TESTS = [{
'url': 'https://channels.vlive.tv/FCD4B', 'url': 'http://channels.vlive.tv/FCD4B',
'info_dict': { 'info_dict': {
'id': 'FCD4B', 'id': 'FCD4B',
'title': 'MAMAMOO', 'title': 'MAMAMOO',
@ -226,63 +182,39 @@ class VLiveChannelIE(InfoExtractor):
'playlist_mincount': 110 'playlist_mincount': 110
}, { }, {
'url': 'https://www.vlive.tv/channel/FCD4B', 'url': 'https://www.vlive.tv/channel/FCD4B',
'info_dict': { 'only_matching': True,
'id': 'FCD4B',
'title': 'MAMAMOO',
},
'playlist_mincount': 110
}] }]
_APP_ID = '8c6cc7b45d2568fb668be6e05b6e5a3b'
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): def _real_extract(self, url):
channel_code = self._match_id(url) channel_code = self._match_id(url)
webpage = self._download_webpage( channel_seq = self._call_api(
'http://channels.vlive.tv/%s/video' % channel_code, channel_code) 'decodeChannelCode', 'Code', channel_code,
'decode channel code', {})['channelSeq']
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 = channel_info['result']['channelSeq']
channel_name = None channel_name = None
entries = [] entries = []
for page_num in itertools.count(1): for page_num in itertools.count(1):
video_list = self._download_json( video_list = self._call_api(
'http://api.vfan.vlive.tv/vproxy/channelplus/getChannelVideoList', 'getChannelVideoList', 'Seq', channel_seq,
channel_code, note='Downloading channel list page #%d' % page_num, 'channel list page #%d' % page_num, {
query={
'app_id': app_id,
'channelSeq': channel_seq,
# Large values of maxNumOfRows (~300 or above) may cause # Large values of maxNumOfRows (~300 or above) may cause
# empty responses (see [1]), e.g. this happens for [2] that # empty responses (see [1]), e.g. this happens for [2] that
# has more than 300 videos. # has more than 300 videos.
# 1. https://github.com/ytdl-org/youtube-dl/issues/13830 # 1. https://github.com/ytdl-org/youtube-dl/issues/13830
# 2. http://channels.vlive.tv/EDBF. # 2. http://channels.vlive.tv/EDBF.
'maxNumOfRows': 100, 'maxNumOfRows': 100,
'_': int(time.time()),
'pageNo': page_num 'pageNo': page_num
} }
) )
@ -290,11 +222,11 @@ class VLiveChannelIE(InfoExtractor):
if not channel_name: if not channel_name:
channel_name = try_get( channel_name = try_get(
video_list, video_list,
lambda x: x['result']['channelInfo']['channelName'], lambda x: x['channelInfo']['channelName'],
compat_str) compat_str)
videos = try_get( videos = try_get(
video_list, lambda x: x['result']['videoList'], list) video_list, lambda x: x['videoList'], list)
if not videos: if not videos:
break break
@ -310,79 +242,3 @@ class VLiveChannelIE(InfoExtractor):
return self.playlist_result( return self.playlist_result(
entries, channel_code, channel_name) entries, channel_code, channel_name)
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
'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
},
'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)
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
video_id, playlist_id = mobj.group('video_id', 'id')
if self._downloader.params.get('noplaylist'):
return self._build_video_result(
video_id,
'Downloading just video %s because of --no-playlist'
% video_id)
self.to_screen(
'Downloading playlist %s - add --no-playlist to just download video'
% playlist_id)
webpage = self._download_webpage(
'http://www.vlive.tv/video/%s/playlist/%s'
% (video_id, playlist_id), playlist_id)
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)
entries = [
self.url_result(
self._VIDEO_URL_TEMPLATE % item_id, ie=VLiveIE.ie_key(),
video_id=compat_str(item_id))
for item_id in item_ids]
playlist_name = self._html_search_regex(
r'<div[^>]+class="[^"]*multicam_playlist[^>]*>\s*<h3[^>]+>([^<]+)',
webpage, 'playlist title', fatal=False)
return self.playlist_result(entries, playlist_id, playlist_name)

Write Preview
Loading…
Cancel
Save