1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 from base64 import b64encode
5 from binascii import hexlify
6 from datetime import datetime
7 from hashlib import md5
8 from random import randint
9 import json
10 import re
11 import time
12
13 from .common import InfoExtractor
14 from ..aes import aes_ecb_encrypt, pkcs7_padding
15 from ..compat import (
16 compat_urllib_parse_urlencode,
17 compat_str,
18 compat_itertools_count,
19 )
20 from ..utils import (
21 ExtractorError,
22 bytes_to_intlist,
23 error_to_compat_str,
24 float_or_none,
25 int_or_none,
26 intlist_to_bytes,
27 sanitized_Request,
28 std_headers,
29 try_get,
30 )
31
32
33 class NetEaseMusicBaseIE(InfoExtractor):
34 _FORMATS = ['bMusic', 'mMusic', 'hMusic']
35 _NETEASE_SALT = '3go8&$8*3*3h0k(2)2'
36 _API_BASE = 'http://music.163.com/api/'
37
38 @classmethod
39 def _encrypt(cls, dfsid):
40 salt_bytes = bytearray(cls._NETEASE_SALT.encode('utf-8'))
41 string_bytes = bytearray(compat_str(dfsid).encode('ascii'))
42 salt_len = len(salt_bytes)
43 for i in range(len(string_bytes)):
44 string_bytes[i] = string_bytes[i] ^ salt_bytes[i % salt_len]
45 m = md5()
46 m.update(bytes(string_bytes))
47 result = b64encode(m.digest()).decode('ascii')
48 return result.replace('/', '_').replace('+', '-')
49
50 @classmethod
51 def make_player_api_request_data_and_headers(cls, song_id, bitrate):
52 KEY = b'e82ckenh8dichen8'
53 URL = '/api/song/enhance/player/url'
54 now = int(time.time() * 1000)
55 rand = randint(0, 1000)
56 cookie = {
57 'osver': None,
58 'deviceId': None,
59 'appver': '8.0.0',
60 'versioncode': '140',
61 'mobilename': None,
62 'buildver': '1623435496',
63 'resolution': '1920x1080',
64 '__csrf': '',
65 'os': 'pc',
66 'channel': None,
67 'requestId': '{0}_{1:04}'.format(now, rand),
68 }
69 request_text = json.dumps(
70 {'ids': '[{0}]'.format(song_id), 'br': bitrate, 'header': cookie},
71 separators=(',', ':'))
72 message = 'nobody{0}use{1}md5forencrypt'.format(
73 URL, request_text).encode('latin1')
74 msg_digest = md5(message).hexdigest()
75
76 data = '{0}-36cd479b6b5-{1}-36cd479b6b5-{2}'.format(
77 URL, request_text, msg_digest)
78 data = pkcs7_padding(bytes_to_intlist(data))
79 encrypted = intlist_to_bytes(aes_ecb_encrypt(data, bytes_to_intlist(KEY)))
80 encrypted_params = hexlify(encrypted).decode('ascii').upper()
81
82 cookie = '; '.join(
83 ['{0}={1}'.format(k, v if v is not None else 'undefined')
84 for [k, v] in cookie.items()])
85
86 headers = {
87 'User-Agent': std_headers['User-Agent'],
88 'Content-Type': 'application/x-www-form-urlencoded',
89 'Referer': 'https://music.163.com',
90 'Cookie': cookie,
91 }
92 return ('params={0}'.format(encrypted_params), headers)
93
94 def _call_player_api(self, song_id, bitrate):
95 url = 'https://interface3.music.163.com/eapi/song/enhance/player/url'
96 data, headers = self.make_player_api_request_data_and_headers(song_id, bitrate)
97 try:
98 msg = 'empty result'
99 result = self._download_json(
100 url, song_id, data=data.encode('ascii'), headers=headers)
101 if result:
102 return result
103 except ExtractorError as e:
104 if type(e.cause) in (ValueError, TypeError):
105 # JSON load failure
106 raise
107 except Exception as e:
108 msg = error_to_compat_str(e)
109 self.report_warning('%s API call (%s) failed: %s' % (
110 song_id, bitrate, msg))
111 return {}
112
113 def extract_formats(self, info):
114 err = 0
115 formats = []
116 song_id = info['id']
117 for song_format in self._FORMATS:
118 details = info.get(song_format)
119 if not details:
120 continue
121
122 bitrate = int_or_none(details.get('bitrate')) or 999000
123 data = self._call_player_api(song_id, bitrate)
124 for song in try_get(data, lambda x: x['data'], list) or []:
125 song_url = try_get(song, lambda x: x['url'])
126 if not song_url:
127 continue
128 if self._is_valid_url(song_url, info['id'], 'song'):
129 formats.append({
130 'url': song_url,
131 'ext': details.get('extension'),
132 'abr': float_or_none(song.get('br'), scale=1000),
133 'format_id': song_format,
134 'filesize': int_or_none(song.get('size')),
135 'asr': int_or_none(details.get('sr')),
136 })
137 elif err == 0:
138 err = try_get(song, lambda x: x['code'], int)
139
140 if not formats:
141 msg = 'No media links found'
142 if err != 0 and (err < 200 or err >= 400):
143 raise ExtractorError(
144 '%s (site code %d)' % (msg, err, ), expected=True)
145 else:
146 self.raise_geo_restricted(
147 msg + ': probably this video is not available from your location due to geo restriction.',
148 countries=['CN'])
149
150 return formats
151
152 @classmethod
153 def convert_milliseconds(cls, ms):
154 return int(round(ms / 1000.0))
155
156 def query_api(self, endpoint, video_id, note):
157 req = sanitized_Request('%s%s' % (self._API_BASE, endpoint))
158 req.add_header('Referer', self._API_BASE)
159 return self._download_json(req, video_id, note)
160
161
162 class NetEaseMusicIE(NetEaseMusicBaseIE):
163 IE_NAME = 'netease:song'
164 IE_DESC = '网易云音乐'
165 _VALID_URL = r'https?://music\.163\.com/(#/)?song\?id=(?P<id>[0-9]+)'
166 _TESTS = [{
167 'url': 'http://music.163.com/#/song?id=32102397',
168 'md5': '3e909614ce09b1ccef4a3eb205441190',
169 'info_dict': {
170 'id': '32102397',
171 'ext': 'mp3',
172 'title': 'Bad Blood',
173 'creator': 'Taylor Swift / Kendrick Lamar',
174 'upload_date': '20150516',
175 'timestamp': 1431792000,
176 'description': 'md5:25fc5f27e47aad975aa6d36382c7833c',
177 },
178 }, {
179 'note': 'No lyrics.',
180 'url': 'http://music.163.com/song?id=17241424',
181 'info_dict': {
182 'id': '17241424',
183 'ext': 'mp3',
184 'title': 'Opus 28',
185 'creator': 'Dustin O\'Halloran',
186 'upload_date': '20080211',
187 'description': 'md5:f12945b0f6e0365e3b73c5032e1b0ff4',
188 'timestamp': 1202745600,
189 },
190 }, {
191 'note': 'Has translated name.',
192 'url': 'http://music.163.com/#/song?id=22735043',
193 'info_dict': {
194 'id': '22735043',
195 'ext': 'mp3',
196 'title': '소원을 말해봐 (Genie)',
197 'creator': '少女时代',
198 'description': 'md5:79d99cc560e4ca97e0c4d86800ee4184',
199 'upload_date': '20100127',
200 'timestamp': 1264608000,
201 'alt_title': '说出愿望吧(Genie)',
202 },
203 }]
204
205 def _process_lyrics(self, lyrics_info):
206 original = lyrics_info.get('lrc', {}).get('lyric')
207 translated = lyrics_info.get('tlyric', {}).get('lyric')
208
209 if not translated:
210 return original
211
212 lyrics_expr = r'(\[[0-9]{2}:[0-9]{2}\.[0-9]{2,}\])([^\n]+)'
213 original_ts_texts = re.findall(lyrics_expr, original)
214 translation_ts_dict = dict(
215 (time_stamp, text) for time_stamp, text in re.findall(lyrics_expr, translated)
216 )
217 lyrics = '\n'.join([
218 '%s%s / %s' % (time_stamp, text, translation_ts_dict.get(time_stamp, ''))
219 for time_stamp, text in original_ts_texts
220 ])
221 return lyrics
222
223 def _real_extract(self, url):
224 song_id = self._match_id(url)
225
226 params = {
227 'id': song_id,
228 'ids': '[%s]' % song_id
229 }
230 info = self.query_api(
231 'song/detail?' + compat_urllib_parse_urlencode(params),
232 song_id, 'Downloading song info')['songs'][0]
233
234 formats = self.extract_formats(info)
235 self._sort_formats(formats)
236
237 lyrics_info = self.query_api(
238 'song/lyric?id=%s&lv=-1&tv=-1' % song_id,
239 song_id, 'Downloading lyrics data')
240 lyrics = self._process_lyrics(lyrics_info)
241
242 alt_title = None
243 if info.get('transNames'):
244 alt_title = '/'.join(info.get('transNames'))
245
246 return {
247 'id': song_id,
248 'title': info['name'],
249 'alt_title': alt_title,
250 'creator': ' / '.join([artist['name'] for artist in info.get('artists', [])]),
251 'timestamp': self.convert_milliseconds(info.get('album', {}).get('publishTime')),
252 'thumbnail': info.get('album', {}).get('picUrl'),
253 'duration': self.convert_milliseconds(info.get('duration', 0)),
254 'description': lyrics,
255 'formats': formats,
256 }
257
258
259 class NetEaseMusicAlbumIE(NetEaseMusicBaseIE):
260 IE_NAME = 'netease:album'
261 IE_DESC = '网易云音乐 - 专辑'
262 _VALID_URL = r'https?://music\.163\.com/(#/)?album\?id=(?P<id>[0-9]+)'
263 _TEST = {
264 'url': 'http://music.163.com/#/album?id=220780',
265 'info_dict': {
266 'id': '220780',
267 'title': 'B\'day',
268 },
269 'playlist_count': 23,
270 'skip': 'Blocked outside Mainland China',
271 }
272
273 def _real_extract(self, url):
274 album_id = self._match_id(url)
275
276 info = self.query_api(
277 'album/%s?id=%s' % (album_id, album_id),
278 album_id, 'Downloading album data')['album']
279
280 name = info['name']
281 desc = info.get('description')
282 entries = [
283 self.url_result('http://music.163.com/#/song?id=%s' % song['id'],
284 'NetEaseMusic', song['id'])
285 for song in info['songs']
286 ]
287 return self.playlist_result(entries, album_id, name, desc)
288
289
290 class NetEaseMusicSingerIE(NetEaseMusicBaseIE):
291 IE_NAME = 'netease:singer'
292 IE_DESC = '网易云音乐 - 歌手'
293 _VALID_URL = r'https?://music\.163\.com/(#/)?artist\?id=(?P<id>[0-9]+)'
294 _TESTS = [{
295 'note': 'Singer has aliases.',
296 'url': 'http://music.163.com/#/artist?id=10559',
297 'info_dict': {
298 'id': '10559',
299 'title': '张惠妹 - aMEI;阿密特',
300 },
301 'playlist_count': 50,
302 'skip': 'Blocked outside Mainland China',
303 }, {
304 'note': 'Singer has translated name.',
305 'url': 'http://music.163.com/#/artist?id=124098',
306 'info_dict': {
307 'id': '124098',
308 'title': '李昇基 - 이승기',
309 },
310 'playlist_count': 50,
311 'skip': 'Blocked outside Mainland China',
312 }]
313
314 def _real_extract(self, url):
315 singer_id = self._match_id(url)
316
317 info = self.query_api(
318 'artist/%s?id=%s' % (singer_id, singer_id),
319 singer_id, 'Downloading singer data')
320
321 name = info['artist']['name']
322 if info['artist']['trans']:
323 name = '%s - %s' % (name, info['artist']['trans'])
324 if info['artist']['alias']:
325 name = '%s - %s' % (name, ';'.join(info['artist']['alias']))
326
327 entries = [
328 self.url_result('http://music.163.com/#/song?id=%s' % song['id'],
329 'NetEaseMusic', song['id'])
330 for song in info['hotSongs']
331 ]
332 return self.playlist_result(entries, singer_id, name)
333
334
335 class NetEaseMusicListIE(NetEaseMusicBaseIE):
336 IE_NAME = 'netease:playlist'
337 IE_DESC = '网易云音乐 - 歌单'
338 _VALID_URL = r'https?://music\.163\.com/(#/)?(playlist|discover/toplist)\?id=(?P<id>[0-9]+)'
339 _TESTS = [{
340 'url': 'http://music.163.com/#/playlist?id=79177352',
341 'info_dict': {
342 'id': '79177352',
343 'title': 'Billboard 2007 Top 100',
344 'description': 'md5:12fd0819cab2965b9583ace0f8b7b022'
345 },
346 'playlist_count': 99,
347 'skip': 'Blocked outside Mainland China',
348 }, {
349 'note': 'Toplist/Charts sample',
350 'url': 'http://music.163.com/#/discover/toplist?id=3733003',
351 'info_dict': {
352 'id': '3733003',
353 'title': 're:韩国Melon排行榜周榜 [0-9]{4}-[0-9]{2}-[0-9]{2}',
354 'description': 'md5:73ec782a612711cadc7872d9c1e134fc',
355 },
356 'playlist_count': 50,
357 'skip': 'Blocked outside Mainland China',
358 }]
359
360 def _real_extract(self, url):
361 list_id = self._match_id(url)
362
363 info = self.query_api(
364 'playlist/detail?id=%s&lv=-1&tv=-1' % list_id,
365 list_id, 'Downloading playlist data')['result']
366
367 name = info['name']
368 desc = info.get('description')
369
370 if info.get('specialType') == 10: # is a chart/toplist
371 datestamp = datetime.fromtimestamp(
372 self.convert_milliseconds(info['updateTime'])).strftime('%Y-%m-%d')
373 name = '%s %s' % (name, datestamp)
374
375 entries = [
376 self.url_result('http://music.163.com/#/song?id=%s' % song['id'],
377 'NetEaseMusic', song['id'])
378 for song in info['tracks']
379 ]
380 return self.playlist_result(entries, list_id, name, desc)
381
382
383 class NetEaseMusicMvIE(NetEaseMusicBaseIE):
384 IE_NAME = 'netease:mv'
385 IE_DESC = '网易云音乐 - MV'
386 _VALID_URL = r'https?://music\.163\.com/(#/)?mv\?id=(?P<id>[0-9]+)'
387 _TEST = {
388 'url': 'http://music.163.com/#/mv?id=415350',
389 'info_dict': {
390 'id': '415350',
391 'ext': 'mp4',
392 'title': '이럴거면 그러지말지',
393 'description': '白雅言自作曲唱甜蜜爱情',
394 'creator': '白雅言',
395 'upload_date': '20150520',
396 },
397 'skip': 'Blocked outside Mainland China',
398 }
399
400 def _real_extract(self, url):
401 mv_id = self._match_id(url)
402
403 info = self.query_api(
404 'mv/detail?id=%s&type=mp4' % mv_id,
405 mv_id, 'Downloading mv info')['data']
406
407 formats = [
408 {'url': mv_url, 'ext': 'mp4', 'format_id': '%sp' % brs, 'height': int(brs)}
409 for brs, mv_url in info['brs'].items()
410 ]
411 self._sort_formats(formats)
412
413 return {
414 'id': mv_id,
415 'title': info['name'],
416 'description': info.get('desc') or info.get('briefDesc'),
417 'creator': info['artistName'],
418 'upload_date': info['publishTime'].replace('-', ''),
419 'formats': formats,
420 'thumbnail': info.get('cover'),
421 'duration': self.convert_milliseconds(info.get('duration', 0)),
422 }
423
424
425 class NetEaseMusicProgramIE(NetEaseMusicBaseIE):
426 IE_NAME = 'netease:program'
427 IE_DESC = '网易云音乐 - 电台节目'
428 _VALID_URL = r'https?://music\.163\.com/(#/?)program\?id=(?P<id>[0-9]+)'
429 _TESTS = [{
430 'url': 'http://music.163.com/#/program?id=10109055',
431 'info_dict': {
432 'id': '10109055',
433 'ext': 'mp3',
434 'title': '不丹足球背后的故事',
435 'description': '喜马拉雅人的足球梦 ...',
436 'creator': '大话西藏',
437 'timestamp': 1434179342,
438 'upload_date': '20150613',
439 'duration': 900,
440 },
441 'skip': 'Blocked outside Mainland China',
442 }, {
443 'note': 'This program has accompanying songs.',
444 'url': 'http://music.163.com/#/program?id=10141022',
445 'info_dict': {
446 'id': '10141022',
447 'title': '25岁，你是自在如风的少年<27°C>',
448 'description': 'md5:8d594db46cc3e6509107ede70a4aaa3b',
449 },
450 'playlist_count': 4,
451 'skip': 'Blocked outside Mainland China',
452 }, {
453 'note': 'This program has accompanying songs.',
454 'url': 'http://music.163.com/#/program?id=10141022',
455 'info_dict': {
456 'id': '10141022',
457 'ext': 'mp3',
458 'title': '25岁，你是自在如风的少年<27°C>',
459 'description': 'md5:8d594db46cc3e6509107ede70a4aaa3b',
460 'timestamp': 1434450841,
461 'upload_date': '20150616',
462 },
463 'params': {
464 'noplaylist': True
465 },
466 'skip': 'Blocked outside Mainland China',
467 }]
468
469 def _real_extract(self, url):
470 program_id = self._match_id(url)
471
472 info = self.query_api(
473 'dj/program/detail?id=%s' % program_id,
474 program_id, 'Downloading program info')['program']
475
476 name = info['name']
477 description = info['description']
478
479 if not info['songs'] or self._downloader.params.get('noplaylist'):
480 if info['songs']:
481 self.to_screen(
482 'Downloading just the main audio %s because of --no-playlist'
483 % info['mainSong']['id'])
484
485 formats = self.extract_formats(info['mainSong'])
486 self._sort_formats(formats)
487
488 return {
489 'id': program_id,
490 'title': name,
491 'description': description,
492 'creator': info['dj']['brand'],
493 'timestamp': self.convert_milliseconds(info['createTime']),
494 'thumbnail': info['coverUrl'],
495 'duration': self.convert_milliseconds(info.get('duration', 0)),
496 'formats': formats,
497 }
498
499 self.to_screen(
500 'Downloading playlist %s - add --no-playlist to just download the main audio %s'
501 % (program_id, info['mainSong']['id']))
502
503 song_ids = [info['mainSong']['id']]
504 song_ids.extend([song['id'] for song in info['songs']])
505 entries = [
506 self.url_result('http://music.163.com/#/song?id=%s' % song_id,
507 'NetEaseMusic', song_id)
508 for song_id in song_ids
509 ]
510 return self.playlist_result(entries, program_id, name, description)
511
512
513 class NetEaseMusicDjRadioIE(NetEaseMusicBaseIE):
514 IE_NAME = 'netease:djradio'
515 IE_DESC = '网易云音乐 - 电台'
516 _VALID_URL = r'https?://music\.163\.com/(#/)?djradio\?id=(?P<id>[0-9]+)'
517 _TEST = {
518 'url': 'http://music.163.com/#/djradio?id=42',
519 'info_dict': {
520 'id': '42',
521 'title': '声音蔓延',
522 'description': 'md5:766220985cbd16fdd552f64c578a6b15'
523 },
524 'playlist_mincount': 40,
525 'skip': 'Blocked outside Mainland China',
526 }
527 _PAGE_SIZE = 1000
528
529 def _real_extract(self, url):
530 dj_id = self._match_id(url)
531
532 name = None
533 desc = None
534 entries = []
535 for offset in compat_itertools_count(start=0, step=self._PAGE_SIZE):
536 info = self.query_api(
537 'dj/program/byradio?asc=false&limit=%d&radioId=%s&offset=%d'
538 % (self._PAGE_SIZE, dj_id, offset),
539 dj_id, 'Downloading dj programs - %d' % offset)
540
541 entries.extend([
542 self.url_result(
543 'http://music.163.com/#/program?id=%s' % program['id'],
544 'NetEaseMusicProgram', program['id'])
545 for program in info['programs']
546 ])
547
548 if name is None:
549 radio = info['programs'][0]['radio']
550 name = radio['name']
551 desc = radio['desc']
552
553 if not info['more']:
554 break
555
556 return self.playlist_result(entries, dj_id, name, desc)
|
