X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/fe979149c83b5a935f7d28baf75848a9137316fd..3cd33af594d34a8809318ebd0be4d606d212a7a0:/youtube_dl/extractor/douyutv.py diff --git a/youtube_dl/extractor/douyutv.py b/youtube_dl/extractor/douyutv.py index ce69627..9115944 100644 --- a/youtube_dl/extractor/douyutv.py +++ b/youtube_dl/extractor/douyutv.py @@ -3,14 +3,22 @@ from __future__ import unicode_literals import hashlib import time +import uuid + from .common import InfoExtractor -from ..utils import (ExtractorError, unescapeHTML) -from ..compat import (compat_str, compat_basestring) +from ..compat import ( + compat_str, + compat_urllib_parse_urlencode, +) +from ..utils import ( + ExtractorError, + unescapeHTML, +) class DouyuTVIE(InfoExtractor): IE_DESC = '斗鱼' - _VALID_URL = r'https?://(?:www\.)?douyu(?:tv)?\.com/(?P[A-Za-z0-9]+)' + _VALID_URL = r'https?://(?:www\.)?douyu(?:tv)?\.com/(?:[^/]+/)*(?P[A-Za-z0-9]+)' _TESTS = [{ 'url': 'http://www.douyutv.com/iseven', 'info_dict': { @@ -18,10 +26,9 @@ class DouyuTVIE(InfoExtractor): '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}$', - 'description': 're:.*m7show@163\.com.*', - 'thumbnail': 're:^https?://.*\.jpg$', + 'description': r're:.*m7show@163\.com.*', + 'thumbnail': r're:^https?://.*\.jpg$', 'uploader': '7师傅', - 'uploader_id': '431925', 'is_live': True, }, 'params': { @@ -35,9 +42,8 @@ class DouyuTVIE(InfoExtractor): 'ext': 'flv', 'title': 're:^小漠从零单排记！——CSOL2躲猫猫 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$', 'description': 'md5:746a2f7a253966a06755a912f0acc0d2', - 'thumbnail': 're:^https?://.*\.jpg$', + 'thumbnail': r're:^https?://.*\.jpg$', 'uploader': 'douyu小漠', - 'uploader_id': '3769985', 'is_live': True, }, 'params': { @@ -51,10 +57,9 @@ class DouyuTVIE(InfoExtractor): '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}$', - 'description': 're:.*m7show@163\.com.*', - 'thumbnail': 're:^https?://.*\.jpg$', + 'description': r're:.*m7show@163\.com.*', + 'thumbnail': r're:^https?://.*\.jpg$', 'uploader': '7师傅', - 'uploader_id': '431925', 'is_live': True, }, 'params': { @@ -63,8 +68,16 @@ class DouyuTVIE(InfoExtractor): }, { 'url': 'http://www.douyu.com/xiaocang', 'only_matching': True, + }, { + # \"room_id\" + 'url': 'http://www.douyu.com/t/lpl', + '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) @@ -73,76 +86,58 @@ class DouyuTVIE(InfoExtractor): else: page = self._download_webpage(url, video_id) room_id = self._html_search_regex( - r'"room_id"\s*:\s*(\d+),', page, 'room id') - - config = None - # Douyu API sometimes returns error "Unable to load the requested class: eticket_redis_cache" - # Retry with different parameters - same parameters cause same errors - for i in range(5): - prefix = 'room/%s?aid=android&client_sys=android&time=%d' % ( - room_id, int(time.time())) - auth = hashlib.md5((prefix + '1231').encode('ascii')).hexdigest() - - config_page = self._download_webpage( - 'http://www.douyutv.com/api/v1/%s&auth=%s' % (prefix, auth), - video_id) - try: - config = self._parse_json(config_page, video_id, fatal=False) - except ExtractorError: - # Wait some time before retrying to get a different time() value - self._sleep(1, video_id, msg_template='%(video_id)s: Error occurs. ' - 'Waiting for %(timeout)s seconds before retrying') - continue - else: - break - if config is None: - raise ExtractorError('Unable to fetch API result') - - data = config['data'] - - error_code = config.get('error', 0) - if error_code is not 0: - error_desc = 'Server reported error %i' % error_code - if isinstance(data, (compat_str, compat_basestring)): - error_desc += ': ' + data - raise ExtractorError(error_desc, expected=True) + r'"room_id\\?"\s*:\s*(\d+),', page, 'room id') + + room = self._download_json( + 'http://m.douyu.com/html5/live?roomId=%s' % room_id, video_id, + note='Downloading room info')['data'] - show_status = data.get('show_status') # 1 = live, 2 = offline - if show_status == '2': - raise ExtractorError( - 'Live stream is offline', expected=True) + 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() - base_url = data['rtmp_url'] - live_path = data['rtmp_live'] + flv_data = compat_urllib_parse_urlencode({ + 'cdn': 'ws', + 'rate': '0', + 'tt': tt, + 'did': did, + 'sign': sign, + }) - title = self._live_title(unescapeHTML(data['room_name'])) - description = data.get('show_details') - thumbnail = data.get('room_src') + 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) - uploader = data.get('nickname') - uploader_id = data.get('owner_uid') + base_url = video_info['data']['rtmp_url'] + live_path = video_info['data']['rtmp_live'] - multi_formats = data.get('rtmp_multi_bitrate') - if not isinstance(multi_formats, dict): - multi_formats = {} - multi_formats['live'] = live_path + video_url = '%s/%s' % (base_url, live_path) - formats = [{ - 'url': '%s/%s' % (base_url, format_path), - 'format_id': format_id, - 'preference': 1 if format_id == 'live' else 0, - } for format_id, format_path in multi_formats.items()] - self._sort_formats(formats) + title = self._live_title(unescapeHTML(room['room_name'])) + description = room.get('notice') + thumbnail = room.get('room_src') + uploader = room.get('nickname') return { 'id': room_id, 'display_id': video_id, + 'url': video_url, 'title': title, 'description': description, 'thumbnail': thumbnail, 'uploader': uploader, - 'uploader_id': uploader_id, - 'formats': formats, 'is_live': True, }