Raphaël G. Git Repositories - youtubedl/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     sanitized_Request,
  11     strip_jsonp,
  12     unescapeHTML,
  13     clean_html,
  14 )
  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             'release_date': '20141227',
  29             'creator': '林俊杰',
  30             'description': 'md5:d327722d0361576fde558f1ac68a7065',
  31             'thumbnail': 're:^https?://.*\.jpg$',
  32         }
  33     }, {
  34         'note': 'There is no mp3-320 version of this song.',
  35         'url': 'http://y.qq.com/#type=song&mid=004MsGEo3DdNxV',
  36         'md5': 'fa3926f0c585cda0af8fa4f796482e3e',
  37         'info_dict': {
  38             'id': '004MsGEo3DdNxV',
  39             'ext': 'mp3',
  40             'title': '如果',
  41             'release_date': '20050626',
  42             'creator': '李季美',
  43             'description': 'md5:46857d5ed62bc4ba84607a805dccf437',
  44             'thumbnail': 're:^https?://.*\.jpg$',
  45         }
  46     }, {
  47         'note': 'lyrics not in .lrc format',
  48         'url': 'http://y.qq.com/#type=song&mid=001JyApY11tIp6',
  49         'info_dict': {
  50             'id': '001JyApY11tIp6',
  51             'ext': 'mp3',
  52             'title': 'Shadows Over Transylvania',
  53             'release_date': '19970225',
  54             'creator': 'Dark Funeral',
  55             'description': 'md5:ed14d5bd7ecec19609108052c25b2c11',
  56             'thumbnail': 're:^https?://.*\.jpg$',
  57         },
  58         'params': {
  59             'skip_download': True,
  60         },
  61     }]
  62
  63     _FORMATS = {
  64         'mp3-320': {'prefix': 'M800', 'ext': 'mp3', 'preference': 40, 'abr': 320},
  65         'mp3-128': {'prefix': 'M500', 'ext': 'mp3', 'preference': 30, 'abr': 128},
  66         'm4a': {'prefix': 'C200', 'ext': 'm4a', 'preference': 10}
  67     }
  68
  69     # Reference: m_r_GetRUin() in top_player.js
  70     # http://imgcache.gtimg.cn/music/portal_v3/y/top_player.js
  71     @staticmethod
  72     def m_r_get_ruin():
  73         curMs = int(time.time() * 1000) % 1000
  74         return int(round(random.random() * 2147483647) * curMs % 1E10)
  75
  76     def _real_extract(self, url):
  77         mid = self._match_id(url)
  78
  79         detail_info_page = self._download_webpage(
  80             'http://s.plcloud.music.qq.com/fcgi-bin/fcg_yqq_song_detail_info.fcg?songmid=%s&play=0' % mid,
  81             mid, note='Download song detail info',
  82             errnote='Unable to get song detail info', encoding='gbk')
  83
  84         song_name = self._html_search_regex(
  85             r"songname:\s*'([^']+)'", detail_info_page, 'song name')
  86
  87         publish_time = self._html_search_regex(
  88             r'发行时间：(\d{4}-\d{2}-\d{2})', detail_info_page,
  89             'publish time', default=None)
  90         if publish_time:
  91             publish_time = publish_time.replace('-', '')
  92
  93         singer = self._html_search_regex(
  94             r"singer:\s*'([^']+)", detail_info_page, 'singer', default=None)
  95
  96         lrc_content = self._html_search_regex(
  97             r'<div class="content" id="lrc_content"[^<>]*>([^<>]+)</div>',
  98             detail_info_page, 'LRC lyrics', default=None)
  99         if lrc_content:
 100             lrc_content = lrc_content.replace('\\n', '\n')
 101
 102         thumbnail_url = None
 103         albummid = self._search_regex(
 104             [r'albummid:\'([0-9a-zA-Z]+)\'', r'"albummid":"([0-9a-zA-Z]+)"'],
 105             detail_info_page, 'album mid', default=None)
 106         if albummid:
 107             thumbnail_url = "http://i.gtimg.cn/music/photo/mid_album_500/%s/%s/%s.jpg" \
 108                             % (albummid[-2:-1], albummid[-1], albummid)
 109
 110         guid = self.m_r_get_ruin()
 111
 112         vkey = self._download_json(
 113             'http://base.music.qq.com/fcgi-bin/fcg_musicexpress.fcg?json=3&guid=%s' % guid,
 114             mid, note='Retrieve vkey', errnote='Unable to get vkey',
 115             transform_source=strip_jsonp)['key']
 116
 117         formats = []
 118         for format_id, details in self._FORMATS.items():
 119             formats.append({
 120                 'url': 'http://cc.stream.qqmusic.qq.com/%s%s.%s?vkey=%s&guid=%s&fromtag=0'
 121                        % (details['prefix'], mid, details['ext'], vkey, guid),
 122                 'format': format_id,
 123                 'format_id': format_id,
 124                 'preference': details['preference'],
 125                 'abr': details.get('abr'),
 126             })
 127         self._check_formats(formats, mid)
 128         self._sort_formats(formats)
 129
 130         actual_lrc_lyrics = ''.join(
 131             line + '\n' for line in re.findall(
 132                 r'(?m)^(\[[0-9]{2}:[0-9]{2}(?:\.[0-9]{2,})?\][^\n]*|\[[^\]]*\])', lrc_content))
 133
 134         info_dict = {
 135             'id': mid,
 136             'formats': formats,
 137             'title': song_name,
 138             'release_date': publish_time,
 139             'creator': singer,
 140             'description': lrc_content,
 141             'thumbnail': thumbnail_url
 142         }
 143         if actual_lrc_lyrics:
 144             info_dict['subtitles'] = {
 145                 'origin': [{
 146                     'ext': 'lrc',
 147                     'data': actual_lrc_lyrics,
 148                 }]
 149             }
 150         return info_dict
 151
 152
 153 class QQPlaylistBaseIE(InfoExtractor):
 154     @staticmethod
 155     def qq_static_url(category, mid):
 156         return 'http://y.qq.com/y/static/%s/%s/%s/%s.html' % (category, mid[-2], mid[-1], mid)
 157
 158     @classmethod
 159     def get_entries_from_page(cls, page):
 160         entries = []
 161
 162         for item in re.findall(r'class="data"[^<>]*>([^<>]+)</', page):
 163             song_mid = unescapeHTML(item).split('|')[-5]
 164             entries.append(cls.url_result(
 165                 'http://y.qq.com/#type=song&mid=' + song_mid, 'QQMusic',
 166                 song_mid))
 167
 168         return entries
 169
 170
 171 class QQMusicSingerIE(QQPlaylistBaseIE):
 172     IE_NAME = 'qqmusic:singer'
 173     IE_DESC = 'QQ音乐 - 歌手'
 174     _VALID_URL = r'http://y.qq.com/#type=singer&mid=(?P<id>[0-9A-Za-z]+)'
 175     _TEST = {
 176         'url': 'http://y.qq.com/#type=singer&mid=001BLpXF2DyJe2',
 177         'info_dict': {
 178             'id': '001BLpXF2DyJe2',
 179             'title': '林俊杰',
 180             'description': 'md5:2a222d89ba4455a3af19940c0481bb78',
 181         },
 182         'playlist_count': 12,
 183     }
 184
 185     def _real_extract(self, url):
 186         mid = self._match_id(url)
 187
 188         singer_page = self._download_webpage(
 189             self.qq_static_url('singer', mid), mid, 'Download singer page')
 190
 191         entries = self.get_entries_from_page(singer_page)
 192
 193         singer_name = self._html_search_regex(
 194             r"singername\s*:\s*'([^']+)'", singer_page, 'singer name',
 195             default=None)
 196
 197         singer_id = self._html_search_regex(
 198             r"singerid\s*:\s*'([0-9]+)'", singer_page, 'singer id',
 199             default=None)
 200
 201         singer_desc = None
 202
 203         if singer_id:
 204             req = sanitized_Request(
 205                 'http://s.plcloud.music.qq.com/fcgi-bin/fcg_get_singer_desc.fcg?utf8=1&outCharset=utf-8&format=xml&singerid=%s' % singer_id)
 206             req.add_header(
 207                 'Referer', 'http://s.plcloud.music.qq.com/xhr_proxy_utf8.html')
 208             singer_desc_page = self._download_xml(
 209                 req, mid, 'Donwload singer description XML')
 210
 211             singer_desc = singer_desc_page.find('./data/info/desc').text
 212
 213         return self.playlist_result(entries, mid, singer_name, singer_desc)
 214
 215
 216 class QQMusicAlbumIE(QQPlaylistBaseIE):
 217     IE_NAME = 'qqmusic:album'
 218     IE_DESC = 'QQ音乐 - 专辑'
 219     _VALID_URL = r'http://y.qq.com/#type=album&mid=(?P<id>[0-9A-Za-z]+)'
 220
 221     _TESTS = [{
 222         'url': 'http://y.qq.com/#type=album&mid=000gXCTb2AhRR1',
 223         'info_dict': {
 224             'id': '000gXCTb2AhRR1',
 225             'title': '我们都是这样长大的',
 226             'description': 'md5:179c5dce203a5931970d306aa9607ea6',
 227         },
 228         'playlist_count': 4,
 229     }, {
 230         'url': 'http://y.qq.com/#type=album&mid=002Y5a3b3AlCu3',
 231         'info_dict': {
 232             'id': '002Y5a3b3AlCu3',
 233             'title': '그리고...',
 234             'description': 'md5:a48823755615508a95080e81b51ba729',
 235         },
 236         'playlist_count': 8,
 237     }]
 238
 239     def _real_extract(self, url):
 240         mid = self._match_id(url)
 241
 242         album = self._download_json(
 243             'http://i.y.qq.com/v8/fcg-bin/fcg_v8_album_info_cp.fcg?albummid=%s&format=json' % mid,
 244             mid, 'Download album page')['data']
 245
 246         entries = [
 247             self.url_result(
 248                 'http://y.qq.com/#type=song&mid=' + song['songmid'], 'QQMusic', song['songmid']
 249             ) for song in album['list']
 250         ]
 251         album_name = album.get('name')
 252         album_detail = album.get('desc')
 253         if album_detail is not None:
 254             album_detail = album_detail.strip()
 255
 256         return self.playlist_result(entries, mid, album_name, album_detail)
 257
 258
 259 class QQMusicToplistIE(QQPlaylistBaseIE):
 260     IE_NAME = 'qqmusic:toplist'
 261     IE_DESC = 'QQ音乐 - 排行榜'
 262     _VALID_URL = r'http://y\.qq\.com/#type=toplist&p=(?P<id>(top|global)_[0-9]+)'
 263
 264     _TESTS = [{
 265         'url': 'http://y.qq.com/#type=toplist&p=global_123',
 266         'info_dict': {
 267             'id': 'global_123',
 268             'title': '美国iTunes榜',
 269         },
 270         'playlist_count': 10,
 271     }, {
 272         'url': 'http://y.qq.com/#type=toplist&p=top_3',
 273         'info_dict': {
 274             'id': 'top_3',
 275             'title': 'QQ音乐巅峰榜·欧美',
 276             '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'
 277                            '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'
 278                            '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：'
 279                            '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'
 280         },
 281         'playlist_count': 100,
 282     }, {
 283         'url': 'http://y.qq.com/#type=toplist&p=global_106',
 284         'info_dict': {
 285             'id': 'global_106',
 286             'title': '韩国Mnet榜',
 287         },
 288         'playlist_count': 50,
 289     }]
 290
 291     def _real_extract(self, url):
 292         list_id = self._match_id(url)
 293
 294         list_type, num_id = list_id.split("_")
 295
 296         toplist_json = self._download_json(
 297             'http://i.y.qq.com/v8/fcg-bin/fcg_v8_toplist_cp.fcg?type=%s&topid=%s&format=json'
 298             % (list_type, num_id),
 299             list_id, 'Download toplist page')
 300
 301         entries = [
 302             self.url_result(
 303                 'http://y.qq.com/#type=song&mid=' + song['data']['songmid'], 'QQMusic', song['data']['songmid']
 304             ) for song in toplist_json['songlist']
 305         ]
 306
 307         topinfo = toplist_json.get('topinfo', {})
 308         list_name = topinfo.get('ListName')
 309         list_description = topinfo.get('info')
 310         return self.playlist_result(entries, list_id, list_name, list_description)
 311
 312
 313 class QQMusicPlaylistIE(QQPlaylistBaseIE):
 314     IE_NAME = 'qqmusic:playlist'
 315     IE_DESC = 'QQ音乐 - 歌单'
 316     _VALID_URL = r'http://y\.qq\.com/#type=taoge&id=(?P<id>[0-9]+)'
 317
 318     _TEST = {
 319         'url': 'http://y.qq.com/#type=taoge&id=3462654915',
 320         'info_dict': {
 321             'id': '3462654915',
 322             'title': '韩国5月新歌精选下旬',
 323             'description': 'md5:d2c9d758a96b9888cf4fe82f603121d4',
 324         },
 325         'playlist_count': 40,
 326     }
 327
 328     def _real_extract(self, url):
 329         list_id = self._match_id(url)
 330
 331         list_json = self._download_json(
 332             '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'
 333             % list_id, list_id, 'Download list page',
 334             transform_source=strip_jsonp)['cdlist'][0]
 335
 336         entries = [
 337             self.url_result(
 338                 'http://y.qq.com/#type=song&mid=' + song['songmid'], 'QQMusic', song['songmid']
 339             ) for song in list_json['songlist']
 340         ]
 341
 342         list_name = list_json.get('dissname')
 343         list_description = clean_html(unescapeHTML(list_json.get('desc')))
 344         return self.playlist_result(entries, list_id, list_name, list_description)