]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/qqmusic.py
1654a641f00db6833056571394b4dbe0b2856150
[youtubedl] / 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 _real_extract(self, url):
62 mid = self._match_id(url)
63
64 detail_info_page = self._download_webpage(
65 'http://s.plcloud.music.qq.com/fcgi-bin/fcg_yqq_song_detail_info.fcg?songmid=%s&play=0' % mid,
66 mid, note='Download song detail info',
67 errnote='Unable to get song detail info', encoding='gbk')
68
69 song_name = self._html_search_regex(
70 r"songname:\s*'([^']+)'", detail_info_page, 'song name')
71
72 publish_time = self._html_search_regex(
73 r'发行时间:(\d{4}-\d{2}-\d{2})', detail_info_page,
74 'publish time', default=None)
75 if publish_time:
76 publish_time = publish_time.replace('-', '')
77
78 singer = self._html_search_regex(
79 r"singer:\s*'([^']+)", detail_info_page, 'singer', default=None)
80
81 lrc_content = self._html_search_regex(
82 r'<div class="content" id="lrc_content"[^<>]*>([^<>]+)</div>',
83 detail_info_page, 'LRC lyrics', default=None)
84 if lrc_content:
85 lrc_content = lrc_content.replace('\\n', '\n')
86
87 thumbnail_url = None
88 albummid = self._search_regex(
89 [r'albummid:\'([0-9a-zA-Z]+)\'', r'"albummid":"([0-9a-zA-Z]+)"'],
90 detail_info_page, 'album mid', default=None)
91 if albummid:
92 thumbnail_url = "http://i.gtimg.cn/music/photo/mid_album_500/%s/%s/%s.jpg" \
93 % (albummid[-2:-1], albummid[-1], albummid)
94
95 guid = self.m_r_get_ruin()
96
97 vkey = self._download_json(
98 'http://base.music.qq.com/fcgi-bin/fcg_musicexpress.fcg?json=3&guid=%s' % guid,
99 mid, note='Retrieve vkey', errnote='Unable to get vkey',
100 transform_source=strip_jsonp)['key']
101
102 formats = []
103 for format_id, details in self._FORMATS.items():
104 formats.append({
105 'url': 'http://cc.stream.qqmusic.qq.com/%s%s.%s?vkey=%s&guid=%s&fromtag=0'
106 % (details['prefix'], mid, details['ext'], vkey, guid),
107 'format': format_id,
108 'format_id': format_id,
109 'preference': details['preference'],
110 'abr': details.get('abr'),
111 })
112 self._check_formats(formats, mid)
113 self._sort_formats(formats)
114
115 return {
116 'id': mid,
117 'formats': formats,
118 'title': song_name,
119 'upload_date': publish_time,
120 'creator': singer,
121 'description': lrc_content,
122 'thumbnail': thumbnail_url,
123 }
124
125
126 class QQPlaylistBaseIE(InfoExtractor):
127 @staticmethod
128 def qq_static_url(category, mid):
129 return 'http://y.qq.com/y/static/%s/%s/%s/%s.html' % (category, mid[-2], mid[-1], mid)
130
131 @classmethod
132 def get_entries_from_page(cls, page):
133 entries = []
134
135 for item in re.findall(r'class="data"[^<>]*>([^<>]+)</', page):
136 song_mid = unescapeHTML(item).split('|')[-5]
137 entries.append(cls.url_result(
138 'http://y.qq.com/#type=song&mid=' + song_mid, 'QQMusic',
139 song_mid))
140
141 return entries
142
143
144 class QQMusicSingerIE(QQPlaylistBaseIE):
145 IE_NAME = 'qqmusic:singer'
146 IE_DESC = 'QQ音乐 - 歌手'
147 _VALID_URL = r'http://y.qq.com/#type=singer&mid=(?P<id>[0-9A-Za-z]+)'
148 _TEST = {
149 'url': 'http://y.qq.com/#type=singer&mid=001BLpXF2DyJe2',
150 'info_dict': {
151 'id': '001BLpXF2DyJe2',
152 'title': '林俊杰',
153 'description': 'md5:2a222d89ba4455a3af19940c0481bb78',
154 },
155 'playlist_count': 12,
156 }
157
158 def _real_extract(self, url):
159 mid = self._match_id(url)
160
161 singer_page = self._download_webpage(
162 self.qq_static_url('singer', mid), mid, 'Download singer page')
163
164 entries = self.get_entries_from_page(singer_page)
165
166 singer_name = self._html_search_regex(
167 r"singername\s*:\s*'([^']+)'", singer_page, 'singer name',
168 default=None)
169
170 singer_id = self._html_search_regex(
171 r"singerid\s*:\s*'([0-9]+)'", singer_page, 'singer id',
172 default=None)
173
174 singer_desc = None
175
176 if singer_id:
177 req = compat_urllib_request.Request(
178 'http://s.plcloud.music.qq.com/fcgi-bin/fcg_get_singer_desc.fcg?utf8=1&outCharset=utf-8&format=xml&singerid=%s' % singer_id)
179 req.add_header(
180 'Referer', 'http://s.plcloud.music.qq.com/xhr_proxy_utf8.html')
181 singer_desc_page = self._download_xml(
182 req, mid, 'Donwload singer description XML')
183
184 singer_desc = singer_desc_page.find('./data/info/desc').text
185
186 return self.playlist_result(entries, mid, singer_name, singer_desc)
187
188
189 class QQMusicAlbumIE(QQPlaylistBaseIE):
190 IE_NAME = 'qqmusic:album'
191 IE_DESC = 'QQ音乐 - 专辑'
192 _VALID_URL = r'http://y.qq.com/#type=album&mid=(?P<id>[0-9A-Za-z]+)'
193
194 _TESTS = [{
195 'url': 'http://y.qq.com/#type=album&mid=000gXCTb2AhRR1',
196 'info_dict': {
197 'id': '000gXCTb2AhRR1',
198 'title': '我们都是这样长大的',
199 'description': 'md5:179c5dce203a5931970d306aa9607ea6',
200 },
201 'playlist_count': 4,
202 }, {
203 'url': 'http://y.qq.com/#type=album&mid=002Y5a3b3AlCu3',
204 'info_dict': {
205 'id': '002Y5a3b3AlCu3',
206 'title': '그리고...',
207 'description': 'md5:a48823755615508a95080e81b51ba729',
208 },
209 'playlist_count': 8,
210 }]
211
212 def _real_extract(self, url):
213 mid = self._match_id(url)
214
215 album = self._download_json(
216 'http://i.y.qq.com/v8/fcg-bin/fcg_v8_album_info_cp.fcg?albummid=%s&format=json' % mid,
217 mid, 'Download album page')['data']
218
219 entries = [
220 self.url_result(
221 'http://y.qq.com/#type=song&mid=' + song['songmid'], 'QQMusic', song['songmid']
222 ) for song in album['list']
223 ]
224 album_name = album.get('name')
225 album_detail = album.get('desc')
226 if album_detail is not None:
227 album_detail = album_detail.strip()
228
229 return self.playlist_result(entries, mid, album_name, album_detail)
230
231
232 class QQMusicToplistIE(QQPlaylistBaseIE):
233 IE_NAME = 'qqmusic:toplist'
234 IE_DESC = 'QQ音乐 - 排行榜'
235 _VALID_URL = r'http://y\.qq\.com/#type=toplist&p=(?P<id>(top|global)_[0-9]+)'
236
237 _TESTS = [{
238 'url': 'http://y.qq.com/#type=toplist&p=global_123',
239 'info_dict': {
240 'id': 'global_123',
241 'title': '美国iTunes榜',
242 },
243 'playlist_count': 10,
244 }, {
245 'url': 'http://y.qq.com/#type=toplist&p=top_3',
246 'info_dict': {
247 'id': 'top_3',
248 'title': 'QQ音乐巅峰榜·欧美',
249 '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'
250 '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'
251 '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:'
252 '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'
253 },
254 'playlist_count': 100,
255 }, {
256 'url': 'http://y.qq.com/#type=toplist&p=global_106',
257 'info_dict': {
258 'id': 'global_106',
259 'title': '韩国Mnet榜',
260 },
261 'playlist_count': 50,
262 }]
263
264 def _real_extract(self, url):
265 list_id = self._match_id(url)
266
267 list_type, num_id = list_id.split("_")
268
269 toplist_json = self._download_json(
270 'http://i.y.qq.com/v8/fcg-bin/fcg_v8_toplist_cp.fcg?type=%s&topid=%s&format=json'
271 % (list_type, num_id),
272 list_id, 'Download toplist page')
273
274 entries = [
275 self.url_result(
276 'http://y.qq.com/#type=song&mid=' + song['data']['songmid'], 'QQMusic', song['data']['songmid']
277 ) for song in toplist_json['songlist']
278 ]
279
280 topinfo = toplist_json.get('topinfo', {})
281 list_name = topinfo.get('ListName')
282 list_description = topinfo.get('info')
283 return self.playlist_result(entries, list_id, list_name, list_description)
284
285
286 class QQMusicPlaylistIE(QQPlaylistBaseIE):
287 IE_NAME = 'qqmusic:playlist'
288 IE_DESC = 'QQ音乐 - 歌单'
289 _VALID_URL = r'http://y\.qq\.com/#type=taoge&id=(?P<id>[0-9]+)'
290
291 _TEST = {
292 'url': 'http://y.qq.com/#type=taoge&id=3462654915',
293 'info_dict': {
294 'id': '3462654915',
295 'title': '韩国5月新歌精选下旬',
296 'description': 'md5:d2c9d758a96b9888cf4fe82f603121d4',
297 },
298 'playlist_count': 40,
299 }
300
301 def _real_extract(self, url):
302 list_id = self._match_id(url)
303
304 list_json = self._download_json(
305 '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'
306 % list_id, list_id, 'Download list page',
307 transform_source=strip_jsonp)['cdlist'][0]
308
309 entries = [
310 self.url_result(
311 'http://y.qq.com/#type=song&mid=' + song['songmid'], 'QQMusic', song['songmid']
312 ) for song in list_json['songlist']
313 ]
314
315 list_name = list_json.get('dissname')
316 list_description = clean_html(unescapeHTML(list_json.get('desc')))
317 return self.playlist_result(entries, list_id, list_name, list_description)