X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/b8d8e13c1f9e4d3cdd7d41c5c9d711a36dd5f9c3..b0b67c374528aca7bc76a9476d554a1d3ab96a9e:/youtube_dl/extractor/vlive.py?ds=sidebyside diff --git a/youtube_dl/extractor/vlive.py b/youtube_dl/extractor/vlive.py index b971890..64d0224 100644 --- a/youtube_dl/extractor/vlive.py +++ b/youtube_dl/extractor/vlive.py @@ -49,6 +49,10 @@ class VLiveIE(InfoExtractor): }, }] + @classmethod + def suitable(cls, url): + return False if VLivePlaylistIE.suitable(url) else super(VLiveIE, cls).suitable(url) + def _real_extract(self, url): video_id = self._match_id(url) @@ -70,9 +74,9 @@ class VLiveIE(InfoExtractor): status, long_video_id, key = params[2], params[5], params[6] status = remove_start(status, 'PRODUCT_') - if status == 'LIVE_ON_AIR' or status == 'BIG_EVENT_ON_AIR': + if status in ('LIVE_ON_AIR', 'BIG_EVENT_ON_AIR'): return self._live(video_id, webpage) - elif status == 'VOD_ON_AIR' or status == 'BIG_EVENT_INTRO': + elif status in ('VOD_ON_AIR', 'BIG_EVENT_INTRO'): if long_video_id and key: return self._replay(video_id, webpage, long_video_id, key) else: @@ -232,7 +236,12 @@ class VLiveChannelIE(InfoExtractor): query={ 'app_id': app_id, 'channelSeq': channel_seq, - 'maxNumOfRows': 1000, + # 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/rg3/youtube-dl/issues/13830 + # 2. http://channels.vlive.tv/EDBF. + 'maxNumOfRows': 100, '_': int(time.time()), 'pageNo': page_num } @@ -261,3 +270,54 @@ class VLiveChannelIE(InfoExtractor): 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]+)' + _TEST = { + 'url': 'http://www.vlive.tv/video/22867/playlist/22912', + 'info_dict': { + 'id': '22912', + 'title': 'Valentine Day Message from TWICE' + }, + 'playlist_mincount': 9 + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id, playlist_id = mobj.group('video_id', 'id') + + VIDEO_URL_TEMPLATE = 'http://www.vlive.tv/video/%s' + if self._downloader.params.get('noplaylist'): + self.to_screen( + 'Downloading just video %s because of --no-playlist' % video_id) + return self.url_result( + VIDEO_URL_TEMPLATE % video_id, + ie=VLiveIE.ie_key(), video_id=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) + + item_ids = self._parse_json( + self._search_regex( + r'playlistVideoSeqs\s*=\s*(\[[^]]+\])', webpage, + 'playlist video seqs'), + playlist_id) + + entries = [ + self.url_result( + 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)