X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/575f75429e2e479957d006a26940f4a5a51347ce..2b4515201a54c38e34434ea2817d606be134a7c9:/youtube_dl/extractor/douyutv.py?ds=inline diff --git a/youtube_dl/extractor/douyutv.py b/youtube_dl/extractor/douyutv.py index 9115944..82d8a04 100644 --- a/youtube_dl/extractor/douyutv.py +++ b/youtube_dl/extractor/douyutv.py @@ -1,15 +1,10 @@ # coding: utf-8 from __future__ import unicode_literals -import hashlib import time -import uuid +import hashlib from .common import InfoExtractor -from ..compat import ( - compat_str, - compat_urllib_parse_urlencode, -) from ..utils import ( ExtractorError, unescapeHTML, @@ -25,7 +20,7 @@ class DouyuTVIE(InfoExtractor): 'id': '17732', 'display_id': 'iseven', 'ext': 'flv', - 'title': 're:^清晨醒脑!T-ara根本停不下来! [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$', + 'title': 're:^清晨醒脑!T-ARA根本停不下来! [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$', 'description': r're:.*m7show@163\.com.*', 'thumbnail': r're:^https?://.*\.jpg$', 'uploader': '7师傅', @@ -56,7 +51,7 @@ class DouyuTVIE(InfoExtractor): 'id': '17732', 'display_id': '17732', 'ext': 'flv', - 'title': 're:^清晨醒脑!T-ara根本停不下来! [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$', + 'title': 're:^清晨醒脑!T-ARA根本停不下来! [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$', 'description': r're:.*m7show@163\.com.*', 'thumbnail': r're:^https?://.*\.jpg$', 'uploader': '7师傅', @@ -74,10 +69,6 @@ class DouyuTVIE(InfoExtractor): 'only_matching': True, }] - # Decompile core.swf in webpage by ffdec "Search SWFs in memory". core.swf - # is encrypted originally, but ffdec can dump memory to get the decrypted one. - _API_KEY = 'A12Svb&%1UUmf@hC' - def _real_extract(self, url): video_id = self._match_id(url) @@ -88,6 +79,7 @@ class DouyuTVIE(InfoExtractor): room_id = self._html_search_regex( r'"room_id\\?"\s*:\s*(\d+),', page, 'room id') + # Grab metadata from mobile API room = self._download_json( 'http://m.douyu.com/html5/live?roomId=%s' % room_id, video_id, note='Downloading room info')['data'] @@ -96,38 +88,22 @@ class DouyuTVIE(InfoExtractor): if room.get('show_status') == '2': raise ExtractorError('Live stream is offline', expected=True) - tt = compat_str(int(time.time() / 60)) - did = uuid.uuid4().hex.upper() - - sign_content = ''.join((room_id, did, self._API_KEY, tt)) - sign = hashlib.md5((sign_content).encode('utf-8')).hexdigest() - - flv_data = compat_urllib_parse_urlencode({ - 'cdn': 'ws', - 'rate': '0', - 'tt': tt, - 'did': did, - 'sign': sign, - }) - - video_info = self._download_json( - 'http://www.douyu.com/lapi/live/getPlay/%s' % room_id, video_id, - data=flv_data, note='Downloading video info', - headers={'Content-Type': 'application/x-www-form-urlencoded'}) - - error_code = video_info.get('error', 0) - if error_code is not 0: - raise ExtractorError( - '%s reported error %i' % (self.IE_NAME, error_code), - expected=True) - - base_url = video_info['data']['rtmp_url'] - live_path = video_info['data']['rtmp_live'] - - video_url = '%s/%s' % (base_url, live_path) + # Grab the URL from PC client API + # The m3u8 url from mobile API requires re-authentication every 5 minutes + tt = int(time.time()) + signContent = 'lapi/live/thirdPart/getPlay/%s?aid=pcclient&rate=0&time=%d9TUk5fjjUjg9qIMH3sdnh' % (room_id, tt) + sign = hashlib.md5(signContent.encode('ascii')).hexdigest() + video_url = self._download_json( + 'http://coapi.douyucdn.cn/lapi/live/thirdPart/getPlay/' + room_id, + video_id, note='Downloading video URL info', + query={'rate': 0}, headers={ + 'auth': sign, + 'time': str(tt), + 'aid': 'pcclient' + })['data']['live_url'] title = self._live_title(unescapeHTML(room['room_name'])) - description = room.get('notice') + description = room.get('show_details') thumbnail = room.get('room_src') uploader = room.get('nickname')