X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fdouyutv.py;h=9757f442259d5678cf3bd6ebeb1dbec28eb03184;hb=HEAD;hp=33efc993eeb89b76af7779ca67de35ad92620150;hpb=545ef4f531e457f8d7ec0cfe79969b4934184f79;p=youtube-dl diff --git a/youtube_dl/extractor/douyutv.py b/youtube_dl/extractor/douyutv.py index 33efc993e..9757f4422 100644 --- a/youtube_dl/extractor/douyutv.py +++ b/youtube_dl/extractor/douyutv.py @@ -1,26 +1,31 @@ # coding: utf-8 from __future__ import unicode_literals -import hashlib import time -import uuid +import hashlib +import re + from .common import InfoExtractor -from ..utils import (ExtractorError, unescapeHTML) -from ..compat import (compat_str, compat_basestring, compat_urllib_parse_urlencode) +from ..utils import ( + ExtractorError, + unescapeHTML, + unified_strdate, + urljoin, +) 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': { '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}$', - 'description': 're:.*m7show@163\.com.*', - 'thumbnail': 're:^https?://.*\.jpg$', + 'title': 're:^清晨醒脑！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', 'is_live': True, }, @@ -35,7 +40,7 @@ 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$', 'uploader': 'douyu小漠', 'is_live': True, }, @@ -49,9 +54,9 @@ class DouyuTVIE(InfoExtractor): 'id': '17732', 'display_id': '17732', 'ext': 'flv', - 'title': 're:^清晨醒脑！T-ara根本停不下来！ [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$', - 'description': 're:.*m7show@163\.com.*', - 'thumbnail': 're:^https?://.*\.jpg$', + 'title': 're:^清晨醒脑！根本停不下来！ [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$', + 'description': r're:.*m7show@163\.com.*', + 'thumbnail': r're:^https?://.*\.jpg$', 'uploader': '7师傅', 'is_live': True, }, @@ -61,6 +66,10 @@ class DouyuTVIE(InfoExtractor): }, { 'url': 'http://www.douyu.com/xiaocang', 'only_matching': True, + }, { + # \"room_id\" + 'url': 'http://www.douyu.com/t/lpl', + 'only_matching': True, }] def _real_extract(self, url): @@ -71,67 +80,33 @@ 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') + r'"room_id\\?"\s*:\s*(\d+),', page, 'room id') - room_url = 'http://m.douyu.com/html5/live?roomId=%s' % room_id - room_content = self._download_webpage(room_url, video_id) - room_json = self._parse_json(room_content, video_id, fatal=False) + # 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'] - room = room_json['data'] - - show_status = room.get('show_status') # 1 = live, 2 = offline - if show_status == '2': - raise ExtractorError( - 'Live stream is offline', expected=True) - - flv_json = 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): - tt = int(time.time() / 60) - did = uuid.uuid4().hex.upper() - - # 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 - # If API changes in the future, just use this way to update - sign_content = '{room_id}{did}A12Svb&%1UUmf@hC{tt}'.format(room_id = room_id, did = did, tt = tt) - sign = hashlib.md5((sign_content).encode('utf-8')).hexdigest() - - payload = {'cdn': 'ws', 'rate': '0', 'tt': tt, 'did': did, 'sign': sign} - flv_data = compat_urllib_parse_urlencode(payload) - - flv_request_url = 'http://www.douyu.com/lapi/live/getPlay/%s' % room_id - flv_content = self._download_webpage(flv_request_url, video_id, data=flv_data, - headers={'Content-Type': 'application/x-www-form-urlencoded'}) - try: - flv_json = self._parse_json(flv_content, 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 flv_json is None: - raise ExtractorError('Unable to fetch API result') - - flv = flv_json['data'] - - error_code = flv_json.get('error', 0) - if error_code is not 0: - error_desc = 'Server reported error %i' % error_code - if isinstance(flv, (compat_str, compat_basestring)): - error_desc += ': ' + flv - raise ExtractorError(error_desc, expected=True) - - base_url = flv['rtmp_url'] - live_path = flv['rtmp_live'] - - video_url = '%s/%s' % (base_url, live_path) + if room.get('show_status') == '2': + raise ExtractorError('Live stream is offline', expected=True) + + # 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') @@ -145,3 +120,82 @@ class DouyuTVIE(InfoExtractor): 'uploader': uploader, 'is_live': True, } + + +class DouyuShowIE(InfoExtractor): + _VALID_URL = r'https?://v(?:mobile)?\.douyu\.com/show/(?P[0-9a-zA-Z]+)' + + _TESTS = [{ + 'url': 'https://v.douyu.com/show/rjNBdvnVXNzvE2yw', + 'md5': '0c2cfd068ee2afe657801269b2d86214', + 'info_dict': { + 'id': 'rjNBdvnVXNzvE2yw', + 'ext': 'mp4', + 'title': '陈一发儿：砒霜 我有个室友系列！04-01 22点场', + 'duration': 7150.08, + 'thumbnail': r're:^https?://.*\.jpg$', + 'uploader': '陈一发儿', + 'uploader_id': 'XrZwYelr5wbK', + 'uploader_url': 'https://v.douyu.com/author/XrZwYelr5wbK', + 'upload_date': '20170402', + }, + }, { + 'url': 'https://vmobile.douyu.com/show/rjNBdvnVXNzvE2yw', + 'only_matching': True, + }] + + def _real_extract(self, url): + url = url.replace('vmobile.', 'v.') + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + + room_info = self._parse_json(self._search_regex( + r'var\s+\$ROOM\s*=\s*({.+});', webpage, 'room info'), video_id) + + video_info = None + + for trial in range(5): + # Sometimes Douyu rejects our request. Let's try it more times + try: + video_info = self._download_json( + 'https://vmobile.douyu.com/video/getInfo', video_id, + query={'vid': video_id}, + headers={ + 'Referer': url, + 'x-requested-with': 'XMLHttpRequest', + }) + break + except ExtractorError: + self._sleep(1, video_id) + + if not video_info: + raise ExtractorError('Can\'t fetch video info') + + formats = self._extract_m3u8_formats( + video_info['data']['video_url'], video_id, + entry_protocol='m3u8_native', ext='mp4') + + upload_date = unified_strdate(self._html_search_regex( + r'上传时间：([^<]+)', webpage, + 'upload date', fatal=False)) + + uploader = uploader_id = uploader_url = None + mobj = re.search( + r'(?m)]+href="/author/([0-9a-zA-Z]+)".+?]+title="([^"]+)"', + webpage) + if mobj: + uploader_id, uploader = mobj.groups() + uploader_url = urljoin(url, '/author/' + uploader_id) + + return { + 'id': video_id, + 'title': room_info['name'], + 'formats': formats, + 'duration': room_info.get('duration'), + 'thumbnail': room_info.get('pic'), + 'upload_date': upload_date, + 'uploader': uploader, + 'uploader_id': uploader_id, + 'uploader_url': uploader_url, + }