X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/b8d8e13c1f9e4d3cdd7d41c5c9d711a36dd5f9c3..9865587cf7b579793ba274067c3e44065d5ff77d:/youtube_dl/extractor/bilibili.py diff --git a/youtube_dl/extractor/bilibili.py b/youtube_dl/extractor/bilibili.py index 80dd838..3746671 100644 --- a/youtube_dl/extractor/bilibili.py +++ b/youtube_dl/extractor/bilibili.py @@ -27,14 +27,14 @@ class BiliBiliIE(InfoExtractor): _TESTS = [{ 'url': 'http://www.bilibili.tv/video/av1074402/', - 'md5': '9fa226fe2b8a9a4d5a69b4c6a183417e', + 'md5': '5f7d29e1a2872f3df0cf76b1f87d3788', 'info_dict': { 'id': '1074402', - 'ext': 'mp4', + 'ext': 'flv', 'title': '【金坷垃】金泡沫', 'description': 'md5:ce18c2a2d2193f0df2917d270f2e5923', - 'duration': 308.315, - 'timestamp': 1398012660, + 'duration': 308.067, + 'timestamp': 1398012678, 'upload_date': '20140420', 'thumbnail': r're:^https?://.+\.jpg', 'uploader': '菊子桑', @@ -54,10 +54,47 @@ class BiliBiliIE(InfoExtractor): 'description': '如果你是神明，并且能够让妄想成为现实。那你会进行怎么样的妄想？是淫靡的世界？独裁社会？毁灭性的制裁？还是……2015年，涩谷。从6年前发生的大灾害“涩谷地震”之后复兴了的这个街区里新设立的私立高中...', }, 'skip': 'Geo-restricted to China', + }, { + # Title with double quotes + 'url': 'http://www.bilibili.com/video/av8903802/', + 'info_dict': { + 'id': '8903802', + 'title': '阿滴英文｜英文歌分享#6 "Closer', + 'description': '滴妹今天唱Closer給你聽! 有史以来，被推最多次也是最久的歌曲，其实歌词跟我原本想像差蛮多的，不过还是好听！ 微博@阿滴英文', + }, + 'playlist': [{ + 'info_dict': { + 'id': '8903802_part1', + 'ext': 'flv', + 'title': '阿滴英文｜英文歌分享#6 "Closer', + 'description': 'md5:3b1b9e25b78da4ef87e9b548b88ee76a', + 'uploader': '阿滴英文', + 'uploader_id': '65880958', + 'timestamp': 1488382634, + 'upload_date': '20170301', + }, + 'params': { + 'skip_download': True, # Test metadata only + }, + }, { + 'info_dict': { + 'id': '8903802_part2', + 'ext': 'flv', + 'title': '阿滴英文｜英文歌分享#6 "Closer', + 'description': 'md5:3b1b9e25b78da4ef87e9b548b88ee76a', + 'uploader': '阿滴英文', + 'uploader_id': '65880958', + 'timestamp': 1488382634, + 'upload_date': '20170301', + }, + 'params': { + 'skip_download': True, # Test metadata only + }, + }] }] - _APP_KEY = '84956560bc028eb7' - _BILIBILI_KEY = '94aba54af9065f71de72f5508f1cd42e' + _APP_KEY = 'iVGUTjsxvpLeuDCf' + _BILIBILI_KEY = 'aHRmhWMLkdeMuILqORnYZocwMBpMEOdt' def _report_error(self, result): if 'message' in result: @@ -76,8 +113,12 @@ class BiliBiliIE(InfoExtractor): webpage = self._download_webpage(url, video_id) if 'anime/' not in url: - cid = compat_parse_qs(self._search_regex( + cid = self._search_regex( + r'\bcid(?:["\']:|=)(\d+)', webpage, 'cid', + default=None + ) or compat_parse_qs(self._search_regex( [r'EmbedPlayer\([^)]+,\s*"([^"]+)"\)', + r'EmbedPlayer\([^)]+,\s*\\"([^"]+)\\"\)', r']+src="https://secure\.bilibili\.com/secure,([^"]+)"'], webpage, 'player parameters'))['cid'][0] else: @@ -86,6 +127,7 @@ class BiliBiliIE(InfoExtractor): video_id, anime_id, compat_urlparse.urljoin(url, '//bangumi.bilibili.com/anime/%s' % anime_id))) headers = { 'Content-Type': 'application/x-www-form-urlencoded; charset=UTF-8', + 'Referer': url } headers.update(self.geo_verification_headers()) @@ -97,43 +139,66 @@ class BiliBiliIE(InfoExtractor): self._report_error(js) cid = js['result']['cid'] - payload = 'appkey=%s&cid=%s&otype=json&quality=2&type=mp4' % (self._APP_KEY, cid) - sign = hashlib.md5((payload + self._BILIBILI_KEY).encode('utf-8')).hexdigest() - - video_info = self._download_json( - 'http://interface.bilibili.com/playurl?%s&sign=%s' % (payload, sign), - video_id, note='Downloading video info page', - headers=self.geo_verification_headers()) - - if 'durl' not in video_info: - self._report_error(video_info) + headers = { + 'Referer': url + } + headers.update(self.geo_verification_headers()) entries = [] - for idx, durl in enumerate(video_info['durl']): - formats = [{ - 'url': durl['url'], - 'filesize': int_or_none(durl['size']), - }] - for backup_url in durl.get('backup_url', []): - formats.append({ - 'url': backup_url, - # backup URLs have lower priorities - 'preference': -2 if 'hd.mp4' in backup_url else -3, + RENDITIONS = ('qn=80&quality=80&type=', 'quality=2&type=mp4') + for num, rendition in enumerate(RENDITIONS, start=1): + payload = 'appkey=%s&cid=%s&otype=json&%s' % (self._APP_KEY, cid, rendition) + sign = hashlib.md5((payload + self._BILIBILI_KEY).encode('utf-8')).hexdigest() + + video_info = self._download_json( + 'http://interface.bilibili.com/v2/playurl?%s&sign=%s' % (payload, sign), + video_id, note='Downloading video info page', + headers=headers, fatal=num == len(RENDITIONS)) + + if not video_info: + continue + + if 'durl' not in video_info: + if num < len(RENDITIONS): + continue + self._report_error(video_info) + + for idx, durl in enumerate(video_info['durl']): + formats = [{ + 'url': durl['url'], + 'filesize': int_or_none(durl['size']), + }] + for backup_url in durl.get('backup_url', []): + formats.append({ + 'url': backup_url, + # backup URLs have lower priorities + 'preference': -2 if 'hd.mp4' in backup_url else -3, + }) + + for a_format in formats: + a_format.setdefault('http_headers', {}).update({ + 'Referer': url, + }) + + self._sort_formats(formats) + + entries.append({ + 'id': '%s_part%s' % (video_id, idx), + 'duration': float_or_none(durl.get('length'), 1000), + 'formats': formats, }) + break - self._sort_formats(formats) - - entries.append({ - 'id': '%s_part%s' % (video_id, idx), - 'duration': float_or_none(durl.get('length'), 1000), - 'formats': formats, - }) - - title = self._html_search_regex(']+title="([^"]+)">', webpage, 'title') + title = self._html_search_regex( + (']+\btitle=(["\'])(?P(?:(?!\1).)+)\1', + '(?s)<h1[^>]*>(?P<title>.+?)</h1>'), webpage, 'title', + group='title') description = self._html_search_meta('description', webpage) timestamp = unified_timestamp(self._html_search_regex( - r'<time[^>]+datetime="([^"]+)"', webpage, 'upload time', default=None)) + r'<time[^>]+datetime="([^"]+)"', webpage, 'upload time', + default=None) or self._html_search_meta( + 'uploadDate', webpage, 'timestamp', default=None)) thumbnail = self._html_search_meta(['og:image', 'thumbnailUrl'], webpage) # TODO 'view_count' requires deobfuscating Javascript @@ -147,13 +212,16 @@ class BiliBiliIE(InfoExtractor): } uploader_mobj = re.search( - r'<a[^>]+href="(?:https?:)?//space\.bilibili\.com/(?P<id>\d+)"[^>]+title="(?P<name>[^"]+)"', + r'<a[^>]+href="(?:https?:)?//space\.bilibili\.com/(?P<id>\d+)"[^>]*>(?P<name>[^<]+)', webpage) if uploader_mobj: info.update({ 'uploader': uploader_mobj.group('name'), 'uploader_id': uploader_mobj.group('id'), }) + if not info.get('uploader'): + info['uploader'] = self._html_search_meta( + 'author', webpage, 'uploader', default=None) for entry in entries: entry.update(info)