X-Git-Url: http://git.cielonegro.org/gitweb.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fkuwo.py;h=f94804d06865c9e5b72c936818dbce5f4bf09229;hb=3c20208eff29fcd3e5d4b13f3d4ffa1be7c56309;hp=51137a982a9581fc6e2e39a1dafdc47351282b3a;hpb=c44c7895b8774fb819b0b664bfcf64a7ebeea4e8;p=youtube-dl.git diff --git a/youtube_dl/extractor/kuwo.py b/youtube_dl/extractor/kuwo.py index 51137a982..f94804d06 100644 --- a/youtube_dl/extractor/kuwo.py +++ b/youtube_dl/extractor/kuwo.py @@ -23,7 +23,7 @@ class KuwoBaseIE(InfoExtractor): {'format': 'aac', 'ext': 'aac', 'abr': 48, 'preference': 10} ] - def _get_formats(self, song_id): + def _get_formats(self, song_id, tolerate_ip_deny=False): formats = [] for file_format in self._FORMATS: song_url = self._download_webpage( @@ -31,6 +31,10 @@ class KuwoBaseIE(InfoExtractor): (file_format['ext'], file_format.get('br', ''), song_id), song_id, note='Download %s url info' % file_format['format'], ) + + if song_url == 'IPDeny' and not tolerate_ip_deny: + raise ExtractorError('This song is blocked in this region', expected=True) + if song_url.startswith('http://') or song_url.startswith('https://'): formats.append({ 'url': song_url, @@ -39,7 +43,12 @@ class KuwoBaseIE(InfoExtractor): 'preference': file_format['preference'], 'abr': file_format.get('abr'), }) - self._sort_formats(formats) + + # XXX _sort_formats fails if there are not formats, while it's not the + # desired behavior if 'IPDeny' is ignored + # This check can be removed if https://github.com/rg3/youtube-dl/pull/8051 is merged + if not tolerate_ip_deny: + self._sort_formats(formats) return formats @@ -64,6 +73,7 @@ class KuwoIE(KuwoBaseIE): 'id': '6446136', 'ext': 'mp3', 'title': '心', + 'description': 'md5:b2ab6295d014005bfc607525bfc1e38a', 'creator': 'IU', 'upload_date': '20150518', }, @@ -77,9 +87,11 @@ class KuwoIE(KuwoBaseIE): webpage = self._download_webpage( url, song_id, note='Download song detail info', errnote='Unable to get song detail info') + if '对不起,该歌曲由于版权问题已被下线,将返回网站首页' in webpage: + raise ExtractorError('this song has been offline because of copyright issues', expected=True) song_name = self._html_search_regex( - r'(?s)class="[^"]*title[^"]*".*?]+title="([^"]+)"', webpage, 'song name') + r'(?s)class="(?:[^"\s]+\s+)*title(?:\s+[^"\s]+)*".*?]+title="([^"]+)"', webpage, 'song name') singer_name = self._html_search_regex( r']+class="s_img">\s*]+title="([^>]+)"', webpage, 'singer name', fatal=False) @@ -281,10 +293,16 @@ class KuwoMvIE(KuwoBaseIE): 'url': 'http://www.kuwo.cn/mv/6480076/', 'info_dict': { 'id': '6480076', - 'ext': 'mkv', - 'title': '我们家MV', + 'ext': 'mp4', + 'title': 'My HouseMV', 'creator': '2PM', }, + # In this video, music URLs (anti.s) are blocked outside China and + # USA, while the MV URL (mvurl) is available globally, so force the MV + # URL for consistent results in different countries + 'params': { + 'format': 'mv', + }, } _FORMATS = KuwoBaseIE._FORMATS + [ {'format': 'mkv', 'ext': 'mkv', 'preference': 250}, @@ -306,7 +324,17 @@ class KuwoMvIE(KuwoBaseIE): else: raise ExtractorError('Unable to find song or singer names') - formats = self._get_formats(song_id) + formats = self._get_formats(song_id, tolerate_ip_deny=True) + + mv_url = self._download_webpage( + 'http://www.kuwo.cn/yy/st/mvurl?rid=MUSIC_%s' % song_id, + song_id, note='Download %s MV URL' % song_id) + formats.append({ + 'url': mv_url, + 'format_id': 'mv', + }) + + self._sort_formats(formats) return { 'id': song_id,