X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/9f2b33881274af98a9145c533a1d295fad71521a..4fbf6829491780534e93bd27e5a749e608c97b46:/youtube_dl/extractor/vlive.py?ds=inline diff --git a/youtube_dl/extractor/vlive.py b/youtube_dl/extractor/vlive.py index 9e2aa58..f79531e 100644 --- a/youtube_dl/extractor/vlive.py +++ b/youtube_dl/extractor/vlive.py @@ -1,88 +1,367 @@ # coding: utf-8 from __future__ import unicode_literals +import re +import time +import itertools + from .common import InfoExtractor +from .naver import NaverBaseIE +from ..compat import compat_str from ..utils import ( - dict_get, - float_or_none, - int_or_none, + ExtractorError, + merge_dicts, + remove_start, + try_get, + urlencode_postdata, ) -from ..compat import compat_urllib_parse -class VLiveIE(InfoExtractor): +class VLiveIE(NaverBaseIE): IE_NAME = 'vlive' _VALID_URL = r'https?://(?:(?:www|m)\.)?vlive\.tv/video/(?P[0-9]+)' - _TEST = { + _NETRC_MACHINE = 'vlive' + _TESTS = [{ 'url': 'http://www.vlive.tv/video/1326', 'md5': 'cc7314812855ce56de70a06a27314983', 'info_dict': { 'id': '1326', 'ext': 'mp4', - 'title': "[V] Girl's Day's Broadcast", + 'title': "[V LIVE] Girl's Day's Broadcast", 'creator': "Girl's Day", 'view_count': int, + 'uploader_id': 'muploader_a', }, - } + }, { + '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', + }, + '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', + }] + + @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) def _real_extract(self, url): video_id = self._match_id(url) webpage = self._download_webpage( - 'http://www.vlive.tv/video/%s' % video_id, video_id) + 'https://www.vlive.tv/video/%s' % video_id, video_id) - long_video_id = self._search_regex( - r'vlive\.tv\.video\.ajax\.request\.handler\.init\(\s*"[0-9]+"\s*,\s*"[^"]*"\s*,\s*"([^"]+)"', - webpage, 'long video id') + VIDEO_PARAMS_RE = r'\bvlive\.video\.init\(([^)]+)' + VIDEO_PARAMS_FIELD = 'video params' - key = self._search_regex( - r'vlive\.tv\.video\.ajax\.request\.handler\.init\(\s*"[0-9]+"\s*,\s*"[^"]*"\s*,\s*"[^"]+"\s*,\s*"([^"]+)"', - webpage, 'key') + params = self._parse_json(self._search_regex( + VIDEO_PARAMS_RE, webpage, VIDEO_PARAMS_FIELD, default=''), video_id, + transform_source=lambda s: '[' + s + ']', fatal=False) - title = self._og_search_title(webpage) + 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)] - playinfo = self._download_json( - 'http://global.apis.naver.com/rmcnmv/rmcnmv/vod_play_videoInfo.json?%s' - % compat_urllib_parse.urlencode({ - 'videoId': long_video_id, - 'key': key, - 'ptc': 'http', - 'doct': 'json', # document type (xml or json) - 'cpt': 'vtt', # captions type (vtt or ttml) - }), video_id) - - formats = [{ - 'url': vid['source'], - 'format_id': vid.get('encodingOption', {}).get('name'), - 'abr': float_or_none(vid.get('bitrate', {}).get('audio')), - 'vbr': float_or_none(vid.get('bitrate', {}).get('video')), - 'width': int_or_none(vid.get('encodingOption', {}).get('width')), - 'height': int_or_none(vid.get('encodingOption', {}).get('height')), - 'filesize': int_or_none(vid.get('size')), - } for vid in playinfo.get('videos', {}).get('list', []) if vid.get('source')] - self._sort_formats(formats) + status, long_video_id, key = params[2], params[5], params[6] + status = remove_start(status, 'PRODUCT_') - thumbnail = self._og_search_thumbnail(webpage) - creator = self._html_search_regex( - r']+class="info_area"[^>]*>\s*]+class="name"[^>]*>([^<]+)', - webpage, 'creator', fatal=False) + 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) - view_count = int_or_none(playinfo.get('meta', {}).get('count')) - - subtitles = {} - for caption in playinfo.get('captions', {}).get('list', []): - lang = dict_get(caption, ('language', 'locale', 'country', 'label')) - if lang and caption.get('source'): - subtitles[lang] = [{ - 'ext': 'vtt', - 'url': caption['source']}] + 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']+class="info_area"[^>]*>\s*(?:]*>.*?\s*)?]*>([^<]+)', + webpage, 'creator', fatal=False) + thumbnail = self._og_search_thumbnail(webpage) return { - 'id': video_id, 'title': title, 'creator': creator, 'thumbnail': thumbnail, - 'view_count': view_count, + } + + 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, - 'subtitles': subtitles, + '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*[0-9A-Z]+)' + _TEST = { + 'url': 'http://channels.vlive.tv/FCD4B', + '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) + + app_id = None + + app_js_url = self._search_regex( + r']+src=(["\'])(?Phttp.+?/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 + 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, + # 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'], + compat_str) + + videos = try_get( + video_list, lambda x: x['result']['videoList'], list) + if not videos: + break + + for video in videos: + video_id = video.get('videoSeq') + if not video_id: + continue + video_id = compat_str(video_id) + entries.append( + self.url_result( + 'http://www.vlive.tv/video/%s' % video_id, + ie=VLiveIE.ie_key(), video_id=video_id)) + + return self.playlist_result( + entries, channel_code, channel_name) + + +class VLivePlaylistIE(InfoExtractor): + IE_NAME = 'vlive:playlist' + _VALID_URL = r'https?://(?:(?:www|m)\.)?vlive\.tv/video/(?P[0-9]+)/playlist/(?P[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']+class="[^"]*multicam_playlist[^>]*>\s*]+>([^<]+)', + webpage, 'playlist title', fatal=False) + + return self.playlist_result(entries, playlist_id, playlist_name)