[qqmusic] Use _check_formats instead
[youtube-dl] / 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 )
13 from ..compat import compat_urllib_request
14
15
16 class QQMusicIE(InfoExtractor):
17     IE_NAME = 'qqmusic'
18     _VALID_URL = r'http://y.qq.com/#type=song&mid=(?P<id>[0-9A-Za-z]+)'
19     _TESTS = [{
20         'url': 'http://y.qq.com/#type=song&mid=004295Et37taLD',
21         'md5': '9ce1c1c8445f561506d2e3cfb0255705',
22         'info_dict': {
23             'id': '004295Et37taLD',
24             'ext': 'mp3',
25             'title': '可惜没如果',
26             'upload_date': '20141227',
27             'creator': '林俊杰',
28             'description': 'md5:d327722d0361576fde558f1ac68a7065',
29             'thumbnail': 'http://i.gtimg.cn/music/photo/mid_album_500/7/p/001IV22P1RDX7p.jpg',
30         }
31     }, {
32         'note': 'There is no mp3-320 version of this song.',
33         'url': 'http://y.qq.com/#type=song&mid=004MsGEo3DdNxV',
34         'md5': 'fa3926f0c585cda0af8fa4f796482e3e',
35         'info_dict': {
36             'id': '004MsGEo3DdNxV',
37             'ext': 'mp3',
38             'title': '如果',
39             'upload_date': '20050626',
40             'creator': '李季美',
41             'description': 'md5:46857d5ed62bc4ba84607a805dccf437',
42             'thumbnail': 'http://i.gtimg.cn/music/photo/mid_album_500/r/Q/0042owYj46IxrQ.jpg',
43         }
44     }]
45
46     _FORMATS = {
47         'mp3-320': {'prefix': 'M800', 'ext': 'mp3', 'preference': 40, 'abr': 320},
48         'mp3-128': {'prefix': 'M500', 'ext': 'mp3', 'preference': 30, 'abr': 128},
49         'm4a': {'prefix': 'C200', 'ext': 'm4a', 'preference': 10}
50     }
51
52     # Reference: m_r_GetRUin() in top_player.js
53     # http://imgcache.gtimg.cn/music/portal_v3/y/top_player.js
54     @staticmethod
55     def m_r_get_ruin():
56         curMs = int(time.time() * 1000) % 1000
57         return int(round(random.random() * 2147483647) * curMs % 1E10)
58
59     def _real_extract(self, url):
60         mid = self._match_id(url)
61
62         detail_info_page = self._download_webpage(
63             'http://s.plcloud.music.qq.com/fcgi-bin/fcg_yqq_song_detail_info.fcg?songmid=%s&play=0' % mid,
64             mid, note='Download song detail info',
65             errnote='Unable to get song detail info', encoding='gbk')
66
67         song_name = self._html_search_regex(
68             r"songname:\s*'([^']+)'", detail_info_page, 'song name')
69
70         publish_time = self._html_search_regex(
71             r'发行时间:(\d{4}-\d{2}-\d{2})', detail_info_page,
72             'publish time', default=None)
73         if publish_time:
74             publish_time = publish_time.replace('-', '')
75
76         singer = self._html_search_regex(
77             r"singer:\s*'([^']+)", detail_info_page, 'singer', default=None)
78
79         lrc_content = self._html_search_regex(
80             r'<div class="content" id="lrc_content"[^<>]*>([^<>]+)</div>',
81             detail_info_page, 'LRC lyrics', default=None)
82         if lrc_content:
83             lrc_content = lrc_content.replace('\\n', '\n')
84
85         thumbnail_url = None
86         albummid = self._search_regex(
87             [r'albummid:\'([0-9a-zA-Z]+)\'', r'"albummid":"([0-9a-zA-Z]+)"'],
88             detail_info_page, 'album mid', default=None)
89         if albummid:
90             thumbnail_url = "http://i.gtimg.cn/music/photo/mid_album_500/%s/%s/%s.jpg" \
91                             % (albummid[-2:-1], albummid[-1], albummid)
92
93         guid = self.m_r_get_ruin()
94
95         vkey = self._download_json(
96             'http://base.music.qq.com/fcgi-bin/fcg_musicexpress.fcg?json=3&guid=%s' % guid,
97             mid, note='Retrieve vkey', errnote='Unable to get vkey',
98             transform_source=strip_jsonp)['key']
99
100         formats = []
101         for format_id, details in self._FORMATS.items():
102             formats.append({
103                 'url': 'http://cc.stream.qqmusic.qq.com/%s%s.%s?vkey=%s&guid=%s&fromtag=0'
104                        % (details['prefix'], mid, details['ext'], vkey, guid),
105                 'format': format_id,
106                 'format_id': format_id,
107                 'preference': details['preference'],
108                 'abr': details.get('abr'),
109             })
110         self._check_formats(formats, mid)
111         self._sort_formats(formats)
112
113         return {
114             'id': mid,
115             'formats': formats,
116             'title': song_name,
117             'upload_date': publish_time,
118             'creator': singer,
119             'description': lrc_content,
120             'thumbnail': thumbnail_url,
121         }
122
123
124 class QQPlaylistBaseIE(InfoExtractor):
125     @staticmethod
126     def qq_static_url(category, mid):
127         return 'http://y.qq.com/y/static/%s/%s/%s/%s.html' % (category, mid[-2], mid[-1], mid)
128
129     @classmethod
130     def get_entries_from_page(cls, page):
131         entries = []
132
133         for item in re.findall(r'class="data"[^<>]*>([^<>]+)</', page):
134             song_mid = unescapeHTML(item).split('|')[-5]
135             entries.append(cls.url_result(
136                 'http://y.qq.com/#type=song&mid=' + song_mid, 'QQMusic',
137                 song_mid))
138
139         return entries
140
141
142 class QQMusicSingerIE(QQPlaylistBaseIE):
143     IE_NAME = 'qqmusic:singer'
144     _VALID_URL = r'http://y.qq.com/#type=singer&mid=(?P<id>[0-9A-Za-z]+)'
145     _TEST = {
146         'url': 'http://y.qq.com/#type=singer&mid=001BLpXF2DyJe2',
147         'info_dict': {
148             'id': '001BLpXF2DyJe2',
149             'title': '林俊杰',
150             'description': 'md5:2a222d89ba4455a3af19940c0481bb78',
151         },
152         'playlist_count': 12,
153     }
154
155     def _real_extract(self, url):
156         mid = self._match_id(url)
157
158         singer_page = self._download_webpage(
159             self.qq_static_url('singer', mid), mid, 'Download singer page')
160
161         entries = self.get_entries_from_page(singer_page)
162
163         singer_name = self._html_search_regex(
164             r"singername\s*:\s*'([^']+)'", singer_page, 'singer name',
165             default=None)
166
167         singer_id = self._html_search_regex(
168             r"singerid\s*:\s*'([0-9]+)'", singer_page, 'singer id',
169             default=None)
170
171         singer_desc = None
172
173         if singer_id:
174             req = compat_urllib_request.Request(
175                 'http://s.plcloud.music.qq.com/fcgi-bin/fcg_get_singer_desc.fcg?utf8=1&outCharset=utf-8&format=xml&singerid=%s' % singer_id)
176             req.add_header(
177                 'Referer', 'http://s.plcloud.music.qq.com/xhr_proxy_utf8.html')
178             singer_desc_page = self._download_xml(
179                 req, mid, 'Donwload singer description XML')
180
181             singer_desc = singer_desc_page.find('./data/info/desc').text
182
183         return self.playlist_result(entries, mid, singer_name, singer_desc)
184
185
186 class QQMusicAlbumIE(QQPlaylistBaseIE):
187     IE_NAME = 'qqmusic:album'
188     _VALID_URL = r'http://y.qq.com/#type=album&mid=(?P<id>[0-9A-Za-z]+)'
189
190     _TEST = {
191         'url': 'http://y.qq.com/#type=album&mid=000gXCTb2AhRR1&play=0',
192         'info_dict': {
193             'id': '000gXCTb2AhRR1',
194             'title': '我们都是这样长大的',
195             'description': 'md5:d216c55a2d4b3537fe4415b8767d74d6',
196         },
197         'playlist_count': 4,
198     }
199
200     def _real_extract(self, url):
201         mid = self._match_id(url)
202
203         album_page = self._download_webpage(
204             self.qq_static_url('album', mid), mid, 'Download album page')
205
206         entries = self.get_entries_from_page(album_page)
207
208         album_name = self._html_search_regex(
209             r"albumname\s*:\s*'([^']+)',", album_page, 'album name',
210             default=None)
211
212         album_detail = self._html_search_regex(
213             r'<div class="album_detail close_detail">\s*<p>((?:[^<>]+(?:<br />)?)+)</p>',
214             album_page, 'album details', default=None)
215
216         return self.playlist_result(entries, mid, album_name, album_detail)
217
218
219 class QQMusicToplistIE(QQPlaylistBaseIE):
220     IE_NAME = 'qqmusic:toplist'
221     _VALID_URL = r'http://y\.qq\.com/#type=toplist&p=(?P<id>(top|global)_[0-9]+)'
222
223     _TESTS = [{
224         'url': 'http://y.qq.com/#type=toplist&p=global_123',
225         'info_dict': {
226             'id': 'global_123',
227             'title': '美国iTunes榜',
228         },
229         'playlist_count': 10,
230     }, {
231         'url': 'http://y.qq.com/#type=toplist&p=top_3',
232         'info_dict': {
233             'id': 'top_3',
234             'title': 'QQ音乐巅峰榜·欧美',
235             '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'
236                            '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'
237                            '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:'
238                            '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'
239         },
240         'playlist_count': 100,
241     }, {
242         'url': 'http://y.qq.com/#type=toplist&p=global_106',
243         'info_dict': {
244             'id': 'global_106',
245             'title': '韩国Mnet榜',
246         },
247         'playlist_count': 50,
248     }]
249
250     def _real_extract(self, url):
251         list_id = self._match_id(url)
252
253         list_type, num_id = list_id.split("_")
254
255         toplist_json = self._download_json(
256             'http://i.y.qq.com/v8/fcg-bin/fcg_v8_toplist_cp.fcg?type=%s&topid=%s&format=json'
257             % (list_type, num_id),
258             list_id, 'Download toplist page')
259
260         entries = [
261             self.url_result(
262                 'http://y.qq.com/#type=song&mid=' + song['data']['songmid'], 'QQMusic', song['data']['songmid']
263             ) for song in toplist_json['songlist']
264         ]
265
266         topinfo = toplist_json.get('topinfo', {})
267         list_name = topinfo.get('ListName')
268         list_description = topinfo.get('info')
269         return self.playlist_result(entries, list_id, list_name, list_description)