youtube_dl/extractor/kuwo.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     get_element_by_id,
   9     clean_html,
  10     ExtractorError,
  11     InAdvancePagedList,
  12     remove_start,
  13 )
  14
  15
  16 class KuwoBaseIE(InfoExtractor):
  17     _FORMATS = [
  18         {'format': 'ape', 'ext': 'ape', 'preference': 100},
  19         {'format': 'mp3-320', 'ext': 'mp3', 'br': '320kmp3', 'abr': 320, 'preference': 80},
  20         {'format': 'mp3-192', 'ext': 'mp3', 'br': '192kmp3', 'abr': 192, 'preference': 70},
  21         {'format': 'mp3-128', 'ext': 'mp3', 'br': '128kmp3', 'abr': 128, 'preference': 60},
  22         {'format': 'wma', 'ext': 'wma', 'preference': 20},
  23         {'format': 'aac', 'ext': 'aac', 'abr': 48, 'preference': 10}
  24     ]
  25
  26     def _get_formats(self, song_id, tolerate_ip_deny=False):
  27         formats = []
  28         for file_format in self._FORMATS:
  29             song_url = self._download_webpage(
  30                 'http://antiserver.kuwo.cn/anti.s?format=%s&br=%s&rid=MUSIC_%s&type=convert_url&response=url' %
  31                 (file_format['ext'], file_format.get('br', ''), song_id),
  32                 song_id, note='Download %s url info' % file_format['format'],
  33             )
  34
  35             if song_url == 'IPDeny' and not tolerate_ip_deny:
  36                 raise ExtractorError('This song is blocked in this region', expected=True)
  37
  38             if song_url.startswith('http://') or song_url.startswith('https://'):
  39                 formats.append({
  40                     'url': song_url,
  41                     'format_id': file_format['format'],
  42                     'format': file_format['format'],
  43                     'preference': file_format['preference'],
  44                     'abr': file_format.get('abr'),
  45                 })
  46
  47         # XXX _sort_formats fails if there are not formats, while it's not the
  48         # desired behavior if 'IPDeny' is ignored
  49         # This check can be removed if https://github.com/rg3/youtube-dl/pull/8051 is merged
  50         if not tolerate_ip_deny:
  51             self._sort_formats(formats)
  52         return formats
  53
  54
  55 class KuwoIE(KuwoBaseIE):
  56     IE_NAME = 'kuwo:song'
  57     IE_DESC = '酷我音乐'
  58     _VALID_URL = r'https?://www\.kuwo\.cn/yinyue/(?P<id>\d+?)'
  59     _TESTS = [{
  60         'url': 'http://www.kuwo.cn/yinyue/635632/',
  61         'info_dict': {
  62             'id': '635632',
  63             'ext': 'ape',
  64             'title': '爱我别走',
  65             'creator': '张震岳',
  66             'upload_date': '20080122',
  67             'description': 'md5:ed13f58e3c3bf3f7fd9fbc4e5a7aa75c'
  68         },
  69         'skip': 'this song has been offline because of copyright issues',
  70     }, {
  71         'url': 'http://www.kuwo.cn/yinyue/6446136/',
  72         'info_dict': {
  73             'id': '6446136',
  74             'ext': 'mp3',
  75             'title': '心',
  76             'description': 'md5:b2ab6295d014005bfc607525bfc1e38a',
  77             'creator': 'IU',
  78             'upload_date': '20150518',
  79         },
  80         'params': {
  81             'format': 'mp3-320'
  82         },
  83     }, {
  84         'url': 'http://www.kuwo.cn/yinyue/3197154?catalog=yueku2016',
  85         'only_matching': True,
  86     }]
  87
  88     def _real_extract(self, url):
  89         song_id = self._match_id(url)
  90         webpage = self._download_webpage(
  91             url, song_id, note='Download song detail info',
  92             errnote='Unable to get song detail info')
  93         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:
  94             raise ExtractorError('this song has been offline because of copyright issues', expected=True)
  95
  96         song_name = self._html_search_regex(
  97             r'(?s)class="(?:[^"\s]+\s+)*title(?:\s+[^"\s]+)*".*?<h1[^>]+title="([^"]+)"', webpage, 'song name')
  98         singer_name = self._html_search_regex(
  99             r'<div[^>]+class="s_img">\s*<a[^>]+title="([^>]+)"',
 100             webpage, 'singer name', fatal=False)
 101         lrc_content = clean_html(get_element_by_id('lrcContent', webpage))
 102         if lrc_content == '暂无':     # indicates no lyrics
 103             lrc_content = None
 104
 105         formats = self._get_formats(song_id)
 106
 107         album_id = self._html_search_regex(
 108             r'<p[^>]+class="album"[^<]+<a[^>]+href="http://www\.kuwo\.cn/album/(\d+)/"',
 109             webpage, 'album id', fatal=False)
 110
 111         publish_time = None
 112         if album_id is not None:
 113             album_info_page = self._download_webpage(
 114                 'http://www.kuwo.cn/album/%s/' % album_id, song_id,
 115                 note='Download album detail info',
 116                 errnote='Unable to get album detail info')
 117
 118             publish_time = self._html_search_regex(
 119                 r'发行时间：(\d{4}-\d{2}-\d{2})', album_info_page,
 120                 'publish time', fatal=False)
 121             if publish_time:
 122                 publish_time = publish_time.replace('-', '')
 123
 124         return {
 125             'id': song_id,
 126             'title': song_name,
 127             'creator': singer_name,
 128             'upload_date': publish_time,
 129             'description': lrc_content,
 130             'formats': formats,
 131         }
 132
 133
 134 class KuwoAlbumIE(InfoExtractor):
 135     IE_NAME = 'kuwo:album'
 136     IE_DESC = '酷我音乐 - 专辑'
 137     _VALID_URL = r'https?://www\.kuwo\.cn/album/(?P<id>\d+?)/'
 138     _TEST = {
 139         'url': 'http://www.kuwo.cn/album/502294/',
 140         'info_dict': {
 141             'id': '502294',
 142             'title': 'M',
 143             'description': 'md5:6a7235a84cc6400ec3b38a7bdaf1d60c',
 144         },
 145         'playlist_count': 2,
 146     }
 147
 148     def _real_extract(self, url):
 149         album_id = self._match_id(url)
 150
 151         webpage = self._download_webpage(
 152             url, album_id, note='Download album info',
 153             errnote='Unable to get album info')
 154
 155         album_name = self._html_search_regex(
 156             r'<div[^>]+class="comm"[^<]+<h1[^>]+title="([^"]+)"', webpage,
 157             'album name')
 158         album_intro = remove_start(
 159             clean_html(get_element_by_id('intro', webpage)),
 160             '%s简介：' % album_name)
 161
 162         entries = [
 163             self.url_result(song_url, 'Kuwo') for song_url in re.findall(
 164                 r'<p[^>]+class="listen"><a[^>]+href="(http://www\.kuwo\.cn/yinyue/\d+/)"',
 165                 webpage)
 166         ]
 167         return self.playlist_result(entries, album_id, album_name, album_intro)
 168
 169
 170 class KuwoChartIE(InfoExtractor):
 171     IE_NAME = 'kuwo:chart'
 172     IE_DESC = '酷我音乐 - 排行榜'
 173     _VALID_URL = r'https?://yinyue\.kuwo\.cn/billboard_(?P<id>[^.]+).htm'
 174     _TEST = {
 175         'url': 'http://yinyue.kuwo.cn/billboard_香港中文龙虎榜.htm',
 176         'info_dict': {
 177             'id': '香港中文龙虎榜',
 178         },
 179         'playlist_mincount': 10,
 180     }
 181
 182     def _real_extract(self, url):
 183         chart_id = self._match_id(url)
 184         webpage = self._download_webpage(
 185             url, chart_id, note='Download chart info',
 186             errnote='Unable to get chart info')
 187
 188         entries = [
 189             self.url_result(song_url, 'Kuwo') for song_url in re.findall(
 190                 r'<a[^>]+href="(http://www\.kuwo\.cn/yinyue/\d+)', webpage)
 191         ]
 192         return self.playlist_result(entries, chart_id)
 193
 194
 195 class KuwoSingerIE(InfoExtractor):
 196     IE_NAME = 'kuwo:singer'
 197     IE_DESC = '酷我音乐 - 歌手'
 198     _VALID_URL = r'https?://www\.kuwo\.cn/mingxing/(?P<id>[^/]+)'
 199     _TESTS = [{
 200         'url': 'http://www.kuwo.cn/mingxing/bruno+mars/',
 201         'info_dict': {
 202             'id': 'bruno+mars',
 203             'title': 'Bruno Mars',
 204         },
 205         'playlist_mincount': 329,
 206     }, {
 207         'url': 'http://www.kuwo.cn/mingxing/Ali/music.htm',
 208         'info_dict': {
 209             'id': 'Ali',
 210             'title': 'Ali',
 211         },
 212         'playlist_mincount': 95,
 213         'skip': 'Regularly stalls travis build',  # See https://travis-ci.org/rg3/youtube-dl/jobs/78878540
 214     }]
 215
 216     PAGE_SIZE = 15
 217
 218     def _real_extract(self, url):
 219         singer_id = self._match_id(url)
 220         webpage = self._download_webpage(
 221             url, singer_id, note='Download singer info',
 222             errnote='Unable to get singer info')
 223
 224         singer_name = self._html_search_regex(
 225             r'<h1>([^<]+)</h1>', webpage, 'singer name')
 226
 227         artist_id = self._html_search_regex(
 228             r'data-artistid="(\d+)"', webpage, 'artist id')
 229
 230         page_count = int(self._html_search_regex(
 231             r'data-page="(\d+)"', webpage, 'page count'))
 232
 233         def page_func(page_num):
 234             webpage = self._download_webpage(
 235                 'http://www.kuwo.cn/artist/contentMusicsAjax',
 236                 singer_id, note='Download song list page #%d' % (page_num + 1),
 237                 errnote='Unable to get song list page #%d' % (page_num + 1),
 238                 query={'artistId': artist_id, 'pn': page_num, 'rn': self.PAGE_SIZE})
 239
 240             return [
 241                 self.url_result(song_url, 'Kuwo') for song_url in re.findall(
 242                     r'<div[^>]+class="name"><a[^>]+href="(http://www\.kuwo\.cn/yinyue/\d+)',
 243                     webpage)
 244             ]
 245
 246         entries = InAdvancePagedList(page_func, page_count, self.PAGE_SIZE)
 247
 248         return self.playlist_result(entries, singer_id, singer_name)
 249
 250
 251 class KuwoCategoryIE(InfoExtractor):
 252     IE_NAME = 'kuwo:category'
 253     IE_DESC = '酷我音乐 - 分类'
 254     _VALID_URL = r'https?://yinyue\.kuwo\.cn/yy/cinfo_(?P<id>\d+?).htm'
 255     _TEST = {
 256         'url': 'http://yinyue.kuwo.cn/yy/cinfo_86375.htm',
 257         'info_dict': {
 258             'id': '86375',
 259             'title': '八十年代精选',
 260             'description': '这些都是属于八十年代的回忆！',
 261         },
 262         'playlist_count': 30,
 263     }
 264
 265     def _real_extract(self, url):
 266         category_id = self._match_id(url)
 267         webpage = self._download_webpage(
 268             url, category_id, note='Download category info',
 269             errnote='Unable to get category info')
 270
 271         category_name = self._html_search_regex(
 272             r'<h1[^>]+title="([^<>]+?)">[^<>]+?</h1>', webpage, 'category name')
 273
 274         category_desc = remove_start(
 275             get_element_by_id('intro', webpage).strip(),
 276             '%s简介：' % category_name)
 277
 278         jsonm = self._parse_json(self._html_search_regex(
 279             r'var\s+jsonm\s*=\s*([^;]+);', webpage, 'category songs'), category_id)
 280
 281         entries = [
 282             self.url_result('http://www.kuwo.cn/yinyue/%s/' % song['musicrid'], 'Kuwo')
 283             for song in jsonm['musiclist']
 284         ]
 285         return self.playlist_result(entries, category_id, category_name, category_desc)
 286
 287
 288 class KuwoMvIE(KuwoBaseIE):
 289     IE_NAME = 'kuwo:mv'
 290     IE_DESC = '酷我音乐 - MV'
 291     _VALID_URL = r'https?://www\.kuwo\.cn/mv/(?P<id>\d+?)/'
 292     _TEST = {
 293         'url': 'http://www.kuwo.cn/mv/6480076/',
 294         'info_dict': {
 295             'id': '6480076',
 296             'ext': 'mp4',
 297             'title': 'My HouseMV',
 298             'creator': '2PM',
 299         },
 300         # In this video, music URLs (anti.s) are blocked outside China and
 301         # USA, while the MV URL (mvurl) is available globally, so force the MV
 302         # URL for consistent results in different countries
 303         'params': {
 304             'format': 'mv',
 305         },
 306     }
 307     _FORMATS = KuwoBaseIE._FORMATS + [
 308         {'format': 'mkv', 'ext': 'mkv', 'preference': 250},
 309         {'format': 'mp4', 'ext': 'mp4', 'preference': 200},
 310     ]
 311
 312     def _real_extract(self, url):
 313         song_id = self._match_id(url)
 314         webpage = self._download_webpage(
 315             url, song_id, note='Download mv detail info: %s' % song_id,
 316             errnote='Unable to get mv detail info: %s' % song_id)
 317
 318         mobj = re.search(
 319             r'<h1[^>]+title="(?P<song>[^"]+)">[^<]+<span[^>]+title="(?P<singer>[^"]+)"',
 320             webpage)
 321         if mobj:
 322             song_name = mobj.group('song')
 323             singer_name = mobj.group('singer')
 324         else:
 325             raise ExtractorError('Unable to find song or singer names')
 326
 327         formats = self._get_formats(song_id, tolerate_ip_deny=True)
 328
 329         mv_url = self._download_webpage(
 330             'http://www.kuwo.cn/yy/st/mvurl?rid=MUSIC_%s' % song_id,
 331             song_id, note='Download %s MV URL' % song_id)
 332         formats.append({
 333             'url': mv_url,
 334             'format_id': 'mv',
 335         })
 336
 337         self._sort_formats(formats)
 338
 339         return {
 340             'id': song_id,
 341             'title': song_name,
 342             'creator': singer_name,
 343             'formats': formats,
 344         }