diff options
Diffstat (limited to 'youtube_dl/extractor/vlive.py')
| -rw-r--r-- | youtube_dl/extractor/vlive.py | 405 |
1 files changed, 302 insertions, 103 deletions
diff --git a/youtube_dl/extractor/vlive.py b/youtube_dl/extractor/vlive.py index 8d671cca7..42da34d44 100644 --- a/youtube_dl/extractor/vlive.py +++ b/youtube_dl/extractor/vlive.py @@ -1,129 +1,328 @@ # coding: utf-8 from __future__ import unicode_literals -import re +import itertools +import json -from .common import InfoExtractor +from .naver import NaverBaseIE +from ..compat import ( + compat_HTTPError, + compat_str, +) from ..utils import ( - dict_get, ExtractorError, - float_or_none, int_or_none, - remove_start, + merge_dicts, + str_or_none, + strip_or_none, + try_get, + urlencode_postdata, ) -from ..compat import compat_urllib_parse_urlencode -class VLiveIE(InfoExtractor): +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]+)' - _TEST = { + _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', 'md5': 'cc7314812855ce56de70a06a27314983', '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', + }, + }, { + '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', + }, + '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, + }] + + 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 _call_api(self, path_template, video_id, fields=None): + query = {'appId': self._APP_ID, 'gcc': 'KR', 'platformType': 'PC'} + if fields: + query['fields'] = fields + try: + 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) + 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 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) - - video_params = self._search_regex( - r'\bvlive\.video\.init\(([^)]+)\)', - webpage, 'video params') - status, _, _, live_params, long_video_id, key = re.split( - r'"\s*,\s*"', video_params)[2:8] - status = remove_start(status, 'PRODUCT_') - - if status == 'LIVE_ON_AIR' or status == 'BIG_EVENT_ON_AIR': - live_params = self._parse_json('"%s"' % live_params, video_id) - live_params = self._parse_json(live_params, video_id) - return self._live(video_id, webpage, live_params) - elif status == 'VOD_ON_AIR' or status == 'BIG_EVENT_INTRO': - if long_video_id and key: - return self._replay(video_id, webpage, long_video_id, key) + 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}') + + 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') + self._sort_formats(formats) + 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: - status = 'COMING_SOON' - - 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) - 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*<a\s+[^>]*>([^<]+)', - webpage, 'creator', fatal=False) - thumbnail = self._og_search_thumbnail(webpage) - return { - 'title': title, - 'creator': creator, - 'thumbnail': thumbnail, + raise ExtractorError('Unknown status ' + status) + + +class VLivePostIE(VLiveIE): + 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' + _SOS_TMPL = _FVIDEO_TMPL % 'sosPlayInfo' + _INKEY_TMPL = _FVIDEO_TMPL % 'inKey' + + 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._SOS_TMPL, 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]), + }) + self._sort_formats(formats) + 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._INKEY_TMPL, 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<id>[0-9A-Z]+)' + _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, + }] + + 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) + + channel_seq = self._call_api( + 'decodeChannelCode', 'Code', channel_code, + 'decode channel code', {})['channelSeq'] + + channel_name = None + entries = [] + + for page_num in itertools.count(1): + 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, + 'pageNo': page_num + } + ) + + if not channel_name: + channel_name = try_get( + video_list, + lambda x: x['channelInfo']['channelName'], + compat_str) + + videos = try_get( + video_list, lambda x: x['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)) - def _live(self, video_id, webpage, live_params): - 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) - - return dict(self._get_common_fields(webpage), - id=video_id, - formats=formats, - is_live=True) - - def _replay(self, video_id, webpage, long_video_id, key): - 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) - - 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']}] - - return dict(self._get_common_fields(webpage), - id=video_id, - formats=formats, - view_count=view_count, - subtitles=subtitles) + return self.playlist_result( + entries, channel_code, channel_name) |