_ Git - youtube-dl/blob - youtube_dl/extractor/qqmusic.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import random
   5 import time
   6 import re
   7
   8 from .common import InfoExtractor
   9 from ..utils import (
  10     strip_jsonp,
  11     unescapeHTML,
  12     clean_html,
  13 )
  14 from ..compat import compat_urllib_request
  15
  16
  17 class QQMusicIE(InfoExtractor):
  18     IE_NAME = 'qqmusic'
  19     IE_DESC = 'QQ音乐'
  20     _VALID_URL = r'http://y.qq.com/#type=song&mid=(?P<id>[0-9A-Za-z]+)'
  21     _TESTS = [{
  22         'url': 'http://y.qq.com/#type=song&mid=004295Et37taLD',
  23         'md5': '9ce1c1c8445f561506d2e3cfb0255705',
  24         'info_dict': {
  25             'id': '004295Et37taLD',
  26             'ext': 'mp3',
  27             'title': '可惜没如果',
  28             'upload_date': '20141227',
  29             'creator': '林俊杰',
  30             'description': 'md5:d327722d0361576fde558f1ac68a7065',
  31             'thumbnail': 're:^https?://.*\.jpg$',
  32         }
  33     }, {
  34         'note': 'There is no mp3-320 version of this song.',
  35         'url': 'http://y.qq.com/#type=song&mid=004MsGEo3DdNxV',
  36         'md5': 'fa3926f0c585cda0af8fa4f796482e3e',
  37         'info_dict': {
  38             'id': '004MsGEo3DdNxV',
  39             'ext': 'mp3',
  40             'title': '如果',
  41             'upload_date': '20050626',
  42             'creator': '李季美',
  43             'description': 'md5:46857d5ed62bc4ba84607a805dccf437',
  44             'thumbnail': 're:^https?://.*\.jpg$',
  45         }
  46     }]
  47
  48     _FORMATS = {
  49         'mp3-320': {'prefix': 'M800', 'ext': 'mp3', 'preference': 40, 'abr': 320},
  50         'mp3-128': {'prefix': 'M500', 'ext': 'mp3', 'preference': 30, 'abr': 128},
  51         'm4a': {'prefix': 'C200', 'ext': 'm4a', 'preference': 10}
  52     }
  53
  54     # Reference: m_r_GetRUin() in top_player.js
  55     # http://imgcache.gtimg.cn/music/portal_v3/y/top_player.js
  56     @staticmethod
  57     def m_r_get_ruin():
  58         curMs = int(time.time() * 1000) % 1000
  59         return int(round(random.random() * 2147483647) * curMs % 1E10)
  60
  61     def _filter_lrc(self, data):
  62         lyrics_expr = r'(\[[0-9]{2}:[0-9]{2}\.[0-9]{2,}\][^\n]*|\[[^\]]*\])'
  63         texts = re.findall(lyrics_expr, data)
  64         return ''.join(i + "\n" for i in texts)
  65
  66     def _real_extract(self, url):
  67         mid = self._match_id(url)
  68
  69         detail_info_page = self._download_webpage(
  70             'http://s.plcloud.music.qq.com/fcgi-bin/fcg_yqq_song_detail_info.fcg?songmid=%s&play=0' % mid,
  71             mid, note='Download song detail info',
  72             errnote='Unable to get song detail info', encoding='gbk')
  73
  74         song_name = self._html_search_regex(
  75             r"songname:\s*'([^']+)'", detail_info_page, 'song name')
  76
  77         publish_time = self._html_search_regex(
  78             r'发行时间：(\d{4}-\d{2}-\d{2})', detail_info_page,
  79             'publish time', default=None)
  80         if publish_time:
  81             publish_time = publish_time.replace('-', '')
  82
  83         singer = self._html_search_regex(
  84             r"singer:\s*'([^']+)", detail_info_page, 'singer', default=None)
  85
  86         lrc_content = self._html_search_regex(
  87             r'<div class="content" id="lrc_content"[^<>]*>([^<>]+)</div>',
  88             detail_info_page, 'LRC lyrics', default=None)
  89         if lrc_content:
  90             lrc_content = lrc_content.replace('\\n', '\n')
  91
  92         thumbnail_url = None
  93         albummid = self._search_regex(
  94             [r'albummid:\'([0-9a-zA-Z]+)\'', r'"albummid":"([0-9a-zA-Z]+)"'],
  95             detail_info_page, 'album mid', default=None)
  96         if albummid:
  97             thumbnail_url = "http://i.gtimg.cn/music/photo/mid_album_500/%s/%s/%s.jpg" \
  98                             % (albummid[-2:-1], albummid[-1], albummid)
  99
 100         guid = self.m_r_get_ruin()
 101
 102         vkey = self._download_json(
 103             'http://base.music.qq.com/fcgi-bin/fcg_musicexpress.fcg?json=3&guid=%s' % guid,
 104             mid, note='Retrieve vkey', errnote='Unable to get vkey',
 105             transform_source=strip_jsonp)['key']
 106
 107         formats = []
 108         for format_id, details in self._FORMATS.items():
 109             formats.append({
 110                 'url': 'http://cc.stream.qqmusic.qq.com/%s%s.%s?vkey=%s&guid=%s&fromtag=0'
 111                        % (details['prefix'], mid, details['ext'], vkey, guid),
 112                 'format': format_id,
 113                 'format_id': format_id,
 114                 'preference': details['preference'],
 115                 'abr': details.get('abr'),
 116             })
 117         self._check_formats(formats, mid)
 118         self._sort_formats(formats)
 119
 120         actual_lrc_lyrics = self._filter_lrc(lrc_content)
 121
 122         info_dict = {
 123             'id': mid,
 124             'formats': formats,
 125             'title': song_name,
 126             'upload_date': publish_time,
 127             'creator': singer,
 128             'description': lrc_content,
 129             'thumbnail': thumbnail_url
 130         }
 131         if actual_lrc_lyrics:
 132             info_dict['subtitles'] = {
 133                 'origin': [{
 134                     'ext': 'lrc',
 135                     'data': actual_lrc_lyrics,
 136                 }]
 137             }
 138         return info_dict
 139
 140
 141 class QQPlaylistBaseIE(InfoExtractor):
 142     @staticmethod
 143     def qq_static_url(category, mid):
 144         return 'http://y.qq.com/y/static/%s/%s/%s/%s.html' % (category, mid[-2], mid[-1], mid)
 145
 146     @classmethod
 147     def get_entries_from_page(cls, page):
 148         entries = []
 149
 150         for item in re.findall(r'class="data"[^<>]*>([^<>]+)</', page):
 151             song_mid = unescapeHTML(item).split('|')[-5]
 152             entries.append(cls.url_result(
 153                 'http://y.qq.com/#type=song&mid=' + song_mid, 'QQMusic',
 154                 song_mid))
 155
 156         return entries
 157
 158
 159 class QQMusicSingerIE(QQPlaylistBaseIE):
 160     IE_NAME = 'qqmusic:singer'
 161     IE_DESC = 'QQ音乐 - 歌手'
 162     _VALID_URL = r'http://y.qq.com/#type=singer&mid=(?P<id>[0-9A-Za-z]+)'
 163     _TEST = {
 164         'url': 'http://y.qq.com/#type=singer&mid=001BLpXF2DyJe2',
 165         'info_dict': {
 166             'id': '001BLpXF2DyJe2',
 167             'title': '林俊杰',
 168             'description': 'md5:2a222d89ba4455a3af19940c0481bb78',
 169         },
 170         'playlist_count': 12,
 171     }
 172
 173     def _real_extract(self, url):
 174         mid = self._match_id(url)
 175
 176         singer_page = self._download_webpage(
 177             self.qq_static_url('singer', mid), mid, 'Download singer page')
 178
 179         entries = self.get_entries_from_page(singer_page)
 180
 181         singer_name = self._html_search_regex(
 182             r"singername\s*:\s*'([^']+)'", singer_page, 'singer name',
 183             default=None)
 184
 185         singer_id = self._html_search_regex(
 186             r"singerid\s*:\s*'([0-9]+)'", singer_page, 'singer id',
 187             default=None)
 188
 189         singer_desc = None
 190
 191         if singer_id:
 192             req = compat_urllib_request.Request(
 193                 'http://s.plcloud.music.qq.com/fcgi-bin/fcg_get_singer_desc.fcg?utf8=1&outCharset=utf-8&format=xml&singerid=%s' % singer_id)
 194             req.add_header(
 195                 'Referer', 'http://s.plcloud.music.qq.com/xhr_proxy_utf8.html')
 196             singer_desc_page = self._download_xml(
 197                 req, mid, 'Donwload singer description XML')
 198
 199             singer_desc = singer_desc_page.find('./data/info/desc').text
 200
 201         return self.playlist_result(entries, mid, singer_name, singer_desc)
 202
 203
 204 class QQMusicAlbumIE(QQPlaylistBaseIE):
 205     IE_NAME = 'qqmusic:album'
 206     IE_DESC = 'QQ音乐 - 专辑'
 207     _VALID_URL = r'http://y.qq.com/#type=album&mid=(?P<id>[0-9A-Za-z]+)'
 208
 209     _TESTS = [{
 210         'url': 'http://y.qq.com/#type=album&mid=000gXCTb2AhRR1',
 211         'info_dict': {
 212             'id': '000gXCTb2AhRR1',
 213             'title': '我们都是这样长大的',
 214             'description': 'md5:179c5dce203a5931970d306aa9607ea6',
 215         },
 216         'playlist_count': 4,
 217     }, {
 218         'url': 'http://y.qq.com/#type=album&mid=002Y5a3b3AlCu3',
 219         'info_dict': {
 220             'id': '002Y5a3b3AlCu3',
 221             'title': '그리고...',
 222             'description': 'md5:a48823755615508a95080e81b51ba729',
 223         },
 224         'playlist_count': 8,
 225     }]
 226
 227     def _real_extract(self, url):
 228         mid = self._match_id(url)
 229
 230         album = self._download_json(
 231             'http://i.y.qq.com/v8/fcg-bin/fcg_v8_album_info_cp.fcg?albummid=%s&format=json' % mid,
 232             mid, 'Download album page')['data']
 233
 234         entries = [
 235             self.url_result(
 236                 'http://y.qq.com/#type=song&mid=' + song['songmid'], 'QQMusic', song['songmid']
 237             ) for song in album['list']
 238         ]
 239         album_name = album.get('name')
 240         album_detail = album.get('desc')
 241         if album_detail is not None:
 242             album_detail = album_detail.strip()
 243
 244         return self.playlist_result(entries, mid, album_name, album_detail)
 245
 246
 247 class QQMusicToplistIE(QQPlaylistBaseIE):
 248     IE_NAME = 'qqmusic:toplist'
 249     IE_DESC = 'QQ音乐 - 排行榜'
 250     _VALID_URL = r'http://y\.qq\.com/#type=toplist&p=(?P<id>(top|global)_[0-9]+)'
 251
 252     _TESTS = [{
 253         'url': 'http://y.qq.com/#type=toplist&p=global_123',
 254         'info_dict': {
 255             'id': 'global_123',
 256             'title': '美国iTunes榜',
 257         },
 258         'playlist_count': 10,
 259     }, {
 260         'url': 'http://y.qq.com/#type=toplist&p=top_3',
 261         'info_dict': {
 262             'id': 'top_3',
 263             'title': 'QQ音乐巅峰榜·欧美',
 264             'description': 'QQ音乐巅峰榜·欧美根据用户收听行为自动生成，集结当下最流行的欧美新歌！:更新时间：每周四22点|统'
 265                            '计周期：一周（上周四至本周三）|统计对象：三个月内发行的欧美歌曲|统计数量：100首|统计算法：根据'
 266                            '歌曲在一周内的有效播放次数，由高到低取前100名（同一歌手最多允许5首歌曲同时上榜）|有效播放次数：'
 267                            '登录用户完整播放一首歌曲，记为一次有效播放；同一用户收听同一首歌曲，每天记录为1次有效播放'
 268         },
 269         'playlist_count': 100,
 270     }, {
 271         'url': 'http://y.qq.com/#type=toplist&p=global_106',
 272         'info_dict': {
 273             'id': 'global_106',
 274             'title': '韩国Mnet榜',
 275         },
 276         'playlist_count': 50,
 277     }]
 278
 279     def _real_extract(self, url):
 280         list_id = self._match_id(url)
 281
 282         list_type, num_id = list_id.split("_")
 283
 284         toplist_json = self._download_json(
 285             'http://i.y.qq.com/v8/fcg-bin/fcg_v8_toplist_cp.fcg?type=%s&topid=%s&format=json'
 286             % (list_type, num_id),
 287             list_id, 'Download toplist page')
 288
 289         entries = [
 290             self.url_result(
 291                 'http://y.qq.com/#type=song&mid=' + song['data']['songmid'], 'QQMusic', song['data']['songmid']
 292             ) for song in toplist_json['songlist']
 293         ]
 294
 295         topinfo = toplist_json.get('topinfo', {})
 296         list_name = topinfo.get('ListName')
 297         list_description = topinfo.get('info')
 298         return self.playlist_result(entries, list_id, list_name, list_description)
 299
 300
 301 class QQMusicPlaylistIE(QQPlaylistBaseIE):
 302     IE_NAME = 'qqmusic:playlist'
 303     IE_DESC = 'QQ音乐 - 歌单'
 304     _VALID_URL = r'http://y\.qq\.com/#type=taoge&id=(?P<id>[0-9]+)'
 305
 306     _TEST = {
 307         'url': 'http://y.qq.com/#type=taoge&id=3462654915',
 308         'info_dict': {
 309             'id': '3462654915',
 310             'title': '韩国5月新歌精选下旬',
 311             'description': 'md5:d2c9d758a96b9888cf4fe82f603121d4',
 312         },
 313         'playlist_count': 40,
 314     }
 315
 316     def _real_extract(self, url):
 317         list_id = self._match_id(url)
 318
 319         list_json = self._download_json(
 320             'http://i.y.qq.com/qzone-music/fcg-bin/fcg_ucc_getcdinfo_byids_cp.fcg?type=1&json=1&utf8=1&onlysong=0&disstid=%s'
 321             % list_id, list_id, 'Download list page',
 322             transform_source=strip_jsonp)['cdlist'][0]
 323
 324         entries = [
 325             self.url_result(
 326                 'http://y.qq.com/#type=song&mid=' + song['songmid'], 'QQMusic', song['songmid']
 327             ) for song in list_json['songlist']
 328         ]
 329
 330         list_name = list_json.get('dissname')
 331         list_description = clean_html(unescapeHTML(list_json.get('desc')))
 332         return self.playlist_result(entries, list_id, list_name, list_description)