X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/3477c644417600d9ec8f8d2a44f82da0a4b15eb5..47d80ec0b18245caeb97018d4c1af18d0b5b972b:/youtube_dl/extractor/vlive.py?ds=sidebyside diff --git a/youtube_dl/extractor/vlive.py b/youtube_dl/extractor/vlive.py index 86c1cb5..acf9fda 100644 --- a/youtube_dl/extractor/vlive.py +++ b/youtube_dl/extractor/vlive.py @@ -1,86 +1,142 @@ # coding: utf-8 from __future__ import unicode_literals -import hmac -from hashlib import sha1 -from base64 import b64encode -from time import time +import re from .common import InfoExtractor from ..utils import ( + dict_get, ExtractorError, - determine_ext + float_or_none, + int_or_none, + remove_start, ) -from ..compat import compat_urllib_parse +from ..compat import compat_urllib_parse_urlencode class VLiveIE(InfoExtractor): IE_NAME = 'vlive' - # www.vlive.tv/video/ links redirect to m.vlive.tv/video/ for mobile devices - _VALID_URL = r'https?://(?:(www|m)\.)?vlive\.tv/video/(?P[0-9]+)' - _TEST = { - 'url': 'http://m.vlive.tv/video/1326', + _VALID_URL = r'https?://(?:(?:www|m)\.)?vlive\.tv/video/(?P[0-9]+)' + _TESTS = [{ + 'url': 'http://www.vlive.tv/video/1326', 'md5': 'cc7314812855ce56de70a06a27314983', 'info_dict': { 'id': '1326', 'ext': 'mp4', - 'title': '[V] Girl\'s Day\'s Broadcast', - 'creator': 'Girl\'s Day', + 'title': "[V LIVE] Girl's Day's Broadcast", + 'creator': "Girl's Day", + 'view_count': int, }, - } - _SECRET = 'rFkwZet6pqk1vQt6SxxUkAHX7YL3lmqzUMrU4IDusTo4jEBdtOhNfT4BYYAdArwH' + }, { + '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', + }, + 'params': { + 'skip_download': True, + }, + }] def _real_extract(self, url): video_id = self._match_id(url) webpage = self._download_webpage( - 'http://m.vlive.tv/video/%s' % video_id, - video_id, note='Download video page') + '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) + 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) - thumbnail = self._og_search_thumbnail(webpage) creator = self._html_search_regex( - r']+class="name">([^<>]+)', webpage, 'creator') + r']+class="info_area"[^>]*>\s*]*>([^<]+)', + webpage, 'creator', fatal=False) + thumbnail = self._og_search_thumbnail(webpage) + return { + 'title': title, + 'creator': creator, + 'thumbnail': thumbnail, + } - url = 'http://global.apis.naver.com/globalV/globalV/vod/%s/playinfo?' % video_id - msgpad = '%.0f' % (time() * 1000) - md = b64encode( - hmac.new(self._SECRET.encode('ascii'), - (url[:255] + msgpad).encode('ascii'), sha1).digest() - ) - url += '&' + compat_urllib_parse.urlencode({'msgpad': msgpad, 'md': md}) - playinfo = self._download_json(url, video_id, 'Downloading video json') + 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) - if playinfo.get('message', '') != 'success': - raise ExtractorError(playinfo.get('message', 'JSON request unsuccessful')) + return dict(self._get_common_fields(webpage), + id=video_id, + formats=formats, + is_live=True) - if not playinfo.get('result'): - raise ExtractorError('No videos found.') + 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 = [] - for vid in playinfo['result'].get('videos', {}).get('list', []): - formats.append({ - 'url': vid['source'], - 'ext': 'mp4', - 'abr': vid.get('bitrate', {}).get('audio'), - 'vbr': vid.get('bitrate', {}).get('video'), - 'format_id': vid['encodingOption']['name'], - 'height': vid.get('height'), - 'width': vid.get('width'), - }) + 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['result'].get('captions', {}).get('list', []): - subtitles[caption['language']] = [ - {'ext': determine_ext(caption['source'], default_ext='vtt'), - 'url': caption['source']}] + for caption in playinfo.get('captions', {}).get('list', []): + lang = dict_get(caption, ('locale', 'language', 'country', 'label')) + if lang and caption.get('source'): + subtitles[lang] = [{ + 'ext': 'vtt', + 'url': caption['source']}] - return { - 'id': video_id, - 'title': title, - 'creator': creator, - 'thumbnail': thumbnail, - 'formats': formats, - 'subtitles': subtitles, - } + return dict(self._get_common_fields(webpage), + id=video_id, + formats=formats, + view_count=view_count, + subtitles=subtitles)