2 from __future__ import unicode_literals
8 from .common import InfoExtractor
10 compat_etree_fromstring,
13 compat_xml_parse_error,
23 class BiliBiliIE(InfoExtractor):
24 _VALID_URL = r'https?://www\.bilibili\.(?:tv|com)/video/av(?P<id>\d+)'
27 'url': 'http://www.bilibili.tv/video/av1074402/',
28 'md5': '5f7d29e1a2872f3df0cf76b1f87d3788',
33 'description': 'md5:ce18c2a2d2193f0df2917d270f2e5923',
35 'timestamp': 1398012660,
36 'upload_date': '20140420',
37 'thumbnail': 're:^https?://.+\.jpg',
39 'uploader_id': '156160',
42 'url': 'http://www.bilibili.com/video/av1041170/',
45 'title': '【BD1080P】刀语【诸神&异域】',
46 'description': '这是个神奇的故事~每个人不留弹幕不给走哦~切利哦!~',
51 # BiliBili blocks keys from time to time. The current key is extracted from
53 # TODO: find the sign algorithm used in the flash player
54 _APP_KEY = '86385cdc024c0f6c'
56 def _real_extract(self, url):
57 mobj = re.match(self._VALID_URL, url)
58 video_id = mobj.group('id')
60 webpage = self._download_webpage(url, video_id)
62 params = compat_parse_qs(self._search_regex(
63 [r'EmbedPlayer\([^)]+,\s*"([^"]+)"\)',
64 r'<iframe[^>]+src="https://secure\.bilibili\.com/secure,([^"]+)"'],
65 webpage, 'player parameters'))
66 cid = params['cid'][0]
68 info_xml_str = self._download_webpage(
69 'http://interface.bilibili.com/v_cdn_play',
70 cid, query={'appkey': self._APP_KEY, 'cid': cid},
71 note='Downloading video info page')
77 info_xml = compat_etree_fromstring(info_xml_str.encode('utf-8'))
78 except compat_xml_parse_error:
79 info_json = self._parse_json(info_xml_str, video_id, fatal=False)
80 err_msg = (info_json or {}).get('error_text')
82 err_msg = xpath_text(info_xml, './message')
84 if info_xml is not None:
85 durls = info_xml.findall('./durl')
88 raise ExtractorError('%s said: %s' % (self.IE_NAME, err_msg), expected=True)
90 raise ExtractorError('No videos found!')
95 size = xpath_text(durl, ['./filesize', './size'])
97 'url': durl.find('./url').text,
98 'filesize': int_or_none(size),
100 for backup_url in durl.findall('./backup_url/url'):
102 'url': backup_url.text,
103 # backup URLs have lower priorities
104 'preference': -2 if 'hd.mp4' in backup_url.text else -3,
107 self._sort_formats(formats)
110 'id': '%s_part%s' % (cid, xpath_text(durl, './order')),
111 'duration': int_or_none(xpath_text(durl, './length'), 1000),
115 title = self._html_search_regex('<h1[^>]+title="([^"]+)">', webpage, 'title')
116 description = self._html_search_meta('description', webpage)
117 datetime_str = self._html_search_regex(
118 r'<time[^>]+datetime="([^"]+)"', webpage, 'upload time', fatal=False)
120 timestamp = calendar.timegm(datetime.datetime.strptime(datetime_str, '%Y-%m-%dT%H:%M').timetuple())
122 # TODO 'view_count' requires deobfuscating Javascript
124 'id': compat_str(cid),
126 'description': description,
127 'timestamp': timestamp,
128 'thumbnail': self._html_search_meta('thumbnailUrl', webpage),
129 'duration': float_or_none(xpath_text(info_xml, './timelength'), scale=1000),
132 uploader_mobj = re.search(
133 r'<a[^>]+href="https?://space\.bilibili\.com/(?P<id>\d+)"[^>]+title="(?P<name>[^"]+)"',
137 'uploader': uploader_mobj.group('name'),
138 'uploader_id': uploader_mobj.group('id'),
141 for entry in entries:
144 if len(entries) == 1:
148 '_type': 'multi_video',
151 'description': description,