yt-dlp/yt_dlp/extractor/neteasemusic.py

477 lines
17 KiB
Python
Raw Normal View History

# coding: utf-8
from __future__ import unicode_literals
from hashlib import md5
2015-07-10 01:29:57 -04:00
from base64 import b64encode
from datetime import datetime
import re
from .common import InfoExtractor
from ..compat import (
compat_urllib_parse_urlencode,
compat_str,
compat_itertools_count,
)
2016-01-19 15:21:51 -05:00
from ..utils import (
sanitized_Request,
float_or_none,
)
class NetEaseMusicBaseIE(InfoExtractor):
_FORMATS = ['bMusic', 'mMusic', 'hMusic']
_NETEASE_SALT = '3go8&$8*3*3h0k(2)2'
_API_BASE = 'http://music.163.com/api/'
@classmethod
def _encrypt(cls, dfsid):
salt_bytes = bytearray(cls._NETEASE_SALT.encode('utf-8'))
string_bytes = bytearray(compat_str(dfsid).encode('ascii'))
salt_len = len(salt_bytes)
2015-07-10 01:29:57 -04:00
for i in range(len(string_bytes)):
string_bytes[i] = string_bytes[i] ^ salt_bytes[i % salt_len]
m = md5()
m.update(bytes(string_bytes))
result = b64encode(m.digest()).decode('ascii')
return result.replace('/', '_').replace('+', '-')
def extract_formats(self, info):
formats = []
for song_format in self._FORMATS:
details = info.get(song_format)
if not details:
continue
song_file_path = '/%s/%s.%s' % (
self._encrypt(details['dfsId']), details['dfsId'], details['extension'])
# 203.130.59.9, 124.40.233.182, 115.231.74.139, etc is a reverse proxy-like feature
# from NetEase's CDN provider that can be used if m5.music.126.net does not
# work, especially for users outside of Mainland China
# via: https://github.com/JixunMoe/unblock-163/issues/3#issuecomment-163115880
for host in ('http://m5.music.126.net', 'http://115.231.74.139/m1.music.126.net',
'http://124.40.233.182/m1.music.126.net', 'http://203.130.59.9/m1.music.126.net'):
song_url = host + song_file_path
if self._is_valid_url(song_url, info['id'], 'song'):
formats.append({
'url': song_url,
'ext': details.get('extension'),
2016-01-19 15:21:51 -05:00
'abr': float_or_none(details.get('bitrate'), scale=1000),
'format_id': song_format,
'filesize': details.get('size'),
'asr': details.get('sr')
})
break
return formats
@classmethod
def convert_milliseconds(cls, ms):
2015-07-10 06:30:23 -04:00
return int(round(ms / 1000.0))
def query_api(self, endpoint, video_id, note):
Switch codebase to use sanitized_Request instead of compat_urllib_request.Request [downloader/dash] Use sanitized_Request [downloader/http] Use sanitized_Request [atresplayer] Use sanitized_Request [bambuser] Use sanitized_Request [bliptv] Use sanitized_Request [brightcove] Use sanitized_Request [cbs] Use sanitized_Request [ceskatelevize] Use sanitized_Request [collegerama] Use sanitized_Request [extractor/common] Use sanitized_Request [crunchyroll] Use sanitized_Request [dailymotion] Use sanitized_Request [dcn] Use sanitized_Request [dramafever] Use sanitized_Request [dumpert] Use sanitized_Request [eitb] Use sanitized_Request [escapist] Use sanitized_Request [everyonesmixtape] Use sanitized_Request [extremetube] Use sanitized_Request [facebook] Use sanitized_Request [fc2] Use sanitized_Request [flickr] Use sanitized_Request [4tube] Use sanitized_Request [gdcvault] Use sanitized_Request [extractor/generic] Use sanitized_Request [hearthisat] Use sanitized_Request [hotnewhiphop] Use sanitized_Request [hypem] Use sanitized_Request [iprima] Use sanitized_Request [ivi] Use sanitized_Request [keezmovies] Use sanitized_Request [letv] Use sanitized_Request [lynda] Use sanitized_Request [metacafe] Use sanitized_Request [minhateca] Use sanitized_Request [miomio] Use sanitized_Request [meovideo] Use sanitized_Request [mofosex] Use sanitized_Request [moniker] Use sanitized_Request [mooshare] Use sanitized_Request [movieclips] Use sanitized_Request [mtv] Use sanitized_Request [myvideo] Use sanitized_Request [neteasemusic] Use sanitized_Request [nfb] Use sanitized_Request [niconico] Use sanitized_Request [noco] Use sanitized_Request [nosvideo] Use sanitized_Request [novamov] Use sanitized_Request [nowness] Use sanitized_Request [nuvid] Use sanitized_Request [played] Use sanitized_Request [pluralsight] Use sanitized_Request [pornhub] Use sanitized_Request [pornotube] Use sanitized_Request [primesharetv] Use sanitized_Request [promptfile] Use sanitized_Request [qqmusic] Use sanitized_Request [rtve] Use sanitized_Request [safari] Use sanitized_Request [sandia] Use sanitized_Request [shared] Use sanitized_Request [sharesix] Use sanitized_Request [sina] Use sanitized_Request [smotri] Use sanitized_Request [sohu] Use sanitized_Request [spankwire] Use sanitized_Request [sportdeutschland] Use sanitized_Request [streamcloud] Use sanitized_Request [streamcz] Use sanitized_Request [tapely] Use sanitized_Request [tube8] Use sanitized_Request [tubitv] Use sanitized_Request [twitch] Use sanitized_Request [twitter] Use sanitized_Request [udemy] Use sanitized_Request [vbox7] Use sanitized_Request [veoh] Use sanitized_Request [vessel] Use sanitized_Request [vevo] Use sanitized_Request [viddler] Use sanitized_Request [videomega] Use sanitized_Request [viewvster] Use sanitized_Request [viki] Use sanitized_Request [vk] Use sanitized_Request [vodlocker] Use sanitized_Request [voicerepublic] Use sanitized_Request [wistia] Use sanitized_Request [xfileshare] Use sanitized_Request [xtube] Use sanitized_Request [xvideos] Use sanitized_Request [yandexmusic] Use sanitized_Request [youku] Use sanitized_Request [youporn] Use sanitized_Request [youtube] Use sanitized_Request [patreon] Use sanitized_Request [extractor/common] Remove unused import [nfb] PEP 8
2015-11-21 11:18:17 -05:00
req = sanitized_Request('%s%s' % (self._API_BASE, endpoint))
req.add_header('Referer', self._API_BASE)
return self._download_json(req, video_id, note)
class NetEaseMusicIE(NetEaseMusicBaseIE):
IE_NAME = 'netease:song'
2015-07-10 12:51:35 -04:00
IE_DESC = '网易云音乐'
_VALID_URL = r'https?://music\.163\.com/(#/)?song\?id=(?P<id>[0-9]+)'
_TESTS = [{
'url': 'http://music.163.com/#/song?id=32102397',
'md5': 'f2e97280e6345c74ba9d5677dd5dcb45',
'info_dict': {
'id': '32102397',
'ext': 'mp3',
'title': 'Bad Blood (feat. Kendrick Lamar)',
'creator': 'Taylor Swift / Kendrick Lamar',
'upload_date': '20150517',
'timestamp': 1431878400,
'description': 'md5:a10a54589c2860300d02e1de821eb2ef',
},
'skip': 'Blocked outside Mainland China',
}, {
'note': 'No lyrics translation.',
'url': 'http://music.163.com/#/song?id=29822014',
'info_dict': {
'id': '29822014',
'ext': 'mp3',
'title': '听见下雨的声音',
'creator': '周杰伦',
'upload_date': '20141225',
'timestamp': 1419523200,
'description': 'md5:a4d8d89f44656af206b7b2555c0bce6c',
},
'skip': 'Blocked outside Mainland China',
}, {
'note': 'No lyrics.',
'url': 'http://music.163.com/song?id=17241424',
'info_dict': {
'id': '17241424',
'ext': 'mp3',
'title': 'Opus 28',
'creator': 'Dustin O\'Halloran',
'upload_date': '20080211',
'timestamp': 1202745600,
},
'skip': 'Blocked outside Mainland China',
2015-07-10 01:29:57 -04:00
}, {
'note': 'Has translated name.',
'url': 'http://music.163.com/#/song?id=22735043',
'info_dict': {
'id': '22735043',
'ext': 'mp3',
'title': '소원을 말해봐 (Genie)',
'creator': '少女时代',
'description': 'md5:79d99cc560e4ca97e0c4d86800ee4184',
'upload_date': '20100127',
'timestamp': 1264608000,
'alt_title': '说出愿望吧(Genie)',
},
'skip': 'Blocked outside Mainland China',
}]
def _process_lyrics(self, lyrics_info):
original = lyrics_info.get('lrc', {}).get('lyric')
translated = lyrics_info.get('tlyric', {}).get('lyric')
if not translated:
return original
lyrics_expr = r'(\[[0-9]{2}:[0-9]{2}\.[0-9]{2,}\])([^\n]+)'
original_ts_texts = re.findall(lyrics_expr, original)
2015-07-10 01:29:57 -04:00
translation_ts_dict = dict(
(time_stamp, text) for time_stamp, text in re.findall(lyrics_expr, translated)
)
lyrics = '\n'.join([
'%s%s / %s' % (time_stamp, text, translation_ts_dict.get(time_stamp, ''))
for time_stamp, text in original_ts_texts
])
return lyrics
def _real_extract(self, url):
song_id = self._match_id(url)
params = {
'id': song_id,
'ids': '[%s]' % song_id
}
info = self.query_api(
'song/detail?' + compat_urllib_parse_urlencode(params),
song_id, 'Downloading song info')['songs'][0]
formats = self.extract_formats(info)
self._sort_formats(formats)
lyrics_info = self.query_api(
2015-07-10 06:30:23 -04:00
'song/lyric?id=%s&lv=-1&tv=-1' % song_id,
song_id, 'Downloading lyrics data')
lyrics = self._process_lyrics(lyrics_info)
alt_title = None
2015-07-10 01:29:57 -04:00
if info.get('transNames'):
alt_title = '/'.join(info.get('transNames'))
return {
'id': song_id,
'title': info['name'],
'alt_title': alt_title,
'creator': ' / '.join([artist['name'] for artist in info.get('artists', [])]),
'timestamp': self.convert_milliseconds(info.get('album', {}).get('publishTime')),
'thumbnail': info.get('album', {}).get('picUrl'),
'duration': self.convert_milliseconds(info.get('duration', 0)),
'description': lyrics,
'formats': formats,
}
class NetEaseMusicAlbumIE(NetEaseMusicBaseIE):
IE_NAME = 'netease:album'
IE_DESC = '网易云音乐 - 专辑'
_VALID_URL = r'https?://music\.163\.com/(#/)?album\?id=(?P<id>[0-9]+)'
_TEST = {
'url': 'http://music.163.com/#/album?id=220780',
'info_dict': {
'id': '220780',
'title': 'B\'day',
},
'playlist_count': 23,
'skip': 'Blocked outside Mainland China',
}
def _real_extract(self, url):
album_id = self._match_id(url)
info = self.query_api(
2015-07-10 06:30:23 -04:00
'album/%s?id=%s' % (album_id, album_id),
album_id, 'Downloading album data')['album']
name = info['name']
desc = info.get('description')
entries = [
self.url_result('http://music.163.com/#/song?id=%s' % song['id'],
'NetEaseMusic', song['id'])
for song in info['songs']
]
return self.playlist_result(entries, album_id, name, desc)
class NetEaseMusicSingerIE(NetEaseMusicBaseIE):
IE_NAME = 'netease:singer'
IE_DESC = '网易云音乐 - 歌手'
_VALID_URL = r'https?://music\.163\.com/(#/)?artist\?id=(?P<id>[0-9]+)'
_TESTS = [{
'note': 'Singer has aliases.',
'url': 'http://music.163.com/#/artist?id=10559',
'info_dict': {
'id': '10559',
'title': '张惠妹 - aMEI;阿密特',
},
'playlist_count': 50,
'skip': 'Blocked outside Mainland China',
}, {
'note': 'Singer has translated name.',
'url': 'http://music.163.com/#/artist?id=124098',
'info_dict': {
'id': '124098',
'title': '李昇基 - 이승기',
},
'playlist_count': 50,
'skip': 'Blocked outside Mainland China',
}]
def _real_extract(self, url):
singer_id = self._match_id(url)
info = self.query_api(
2015-07-10 06:30:23 -04:00
'artist/%s?id=%s' % (singer_id, singer_id),
singer_id, 'Downloading singer data')
name = info['artist']['name']
if info['artist']['trans']:
name = '%s - %s' % (name, info['artist']['trans'])
if info['artist']['alias']:
name = '%s - %s' % (name, ';'.join(info['artist']['alias']))
entries = [
self.url_result('http://music.163.com/#/song?id=%s' % song['id'],
'NetEaseMusic', song['id'])
for song in info['hotSongs']
]
return self.playlist_result(entries, singer_id, name)
class NetEaseMusicListIE(NetEaseMusicBaseIE):
IE_NAME = 'netease:playlist'
IE_DESC = '网易云音乐 - 歌单'
_VALID_URL = r'https?://music\.163\.com/(#/)?(playlist|discover/toplist)\?id=(?P<id>[0-9]+)'
_TESTS = [{
'url': 'http://music.163.com/#/playlist?id=79177352',
'info_dict': {
'id': '79177352',
'title': 'Billboard 2007 Top 100',
'description': 'md5:12fd0819cab2965b9583ace0f8b7b022'
},
'playlist_count': 99,
'skip': 'Blocked outside Mainland China',
}, {
'note': 'Toplist/Charts sample',
'url': 'http://music.163.com/#/discover/toplist?id=3733003',
'info_dict': {
'id': '3733003',
'title': 're:韩国Melon排行榜周榜 [0-9]{4}-[0-9]{2}-[0-9]{2}',
'description': 'md5:73ec782a612711cadc7872d9c1e134fc',
},
'playlist_count': 50,
'skip': 'Blocked outside Mainland China',
}]
def _real_extract(self, url):
list_id = self._match_id(url)
info = self.query_api(
2015-07-10 06:30:23 -04:00
'playlist/detail?id=%s&lv=-1&tv=-1' % list_id,
list_id, 'Downloading playlist data')['result']
name = info['name']
desc = info.get('description')
if info.get('specialType') == 10: # is a chart/toplist
datestamp = datetime.fromtimestamp(
self.convert_milliseconds(info['updateTime'])).strftime('%Y-%m-%d')
name = '%s %s' % (name, datestamp)
entries = [
self.url_result('http://music.163.com/#/song?id=%s' % song['id'],
'NetEaseMusic', song['id'])
for song in info['tracks']
]
return self.playlist_result(entries, list_id, name, desc)
class NetEaseMusicMvIE(NetEaseMusicBaseIE):
IE_NAME = 'netease:mv'
IE_DESC = '网易云音乐 - MV'
_VALID_URL = r'https?://music\.163\.com/(#/)?mv\?id=(?P<id>[0-9]+)'
_TEST = {
'url': 'http://music.163.com/#/mv?id=415350',
'info_dict': {
'id': '415350',
'ext': 'mp4',
'title': '이럴거면 그러지말지',
'description': '白雅言自作曲唱甜蜜爱情',
'creator': '白雅言',
'upload_date': '20150520',
},
'skip': 'Blocked outside Mainland China',
}
def _real_extract(self, url):
mv_id = self._match_id(url)
info = self.query_api(
'mv/detail?id=%s&type=mp4' % mv_id,
mv_id, 'Downloading mv info')['data']
formats = [
2015-07-10 01:29:57 -04:00
{'url': mv_url, 'ext': 'mp4', 'format_id': '%sp' % brs, 'height': int(brs)}
for brs, mv_url in info['brs'].items()
]
self._sort_formats(formats)
return {
'id': mv_id,
'title': info['name'],
'description': info.get('desc') or info.get('briefDesc'),
'creator': info['artistName'],
'upload_date': info['publishTime'].replace('-', ''),
'formats': formats,
'thumbnail': info.get('cover'),
'duration': self.convert_milliseconds(info.get('duration', 0)),
}
class NetEaseMusicProgramIE(NetEaseMusicBaseIE):
IE_NAME = 'netease:program'
IE_DESC = '网易云音乐 - 电台节目'
_VALID_URL = r'https?://music\.163\.com/(#/?)program\?id=(?P<id>[0-9]+)'
_TESTS = [{
'url': 'http://music.163.com/#/program?id=10109055',
'info_dict': {
'id': '10109055',
'ext': 'mp3',
'title': '不丹足球背后的故事',
'description': '喜马拉雅人的足球梦 ...',
'creator': '大话西藏',
'timestamp': 1434179342,
'upload_date': '20150613',
'duration': 900,
},
'skip': 'Blocked outside Mainland China',
}, {
'note': 'This program has accompanying songs.',
'url': 'http://music.163.com/#/program?id=10141022',
'info_dict': {
'id': '10141022',
'title': '25岁你是自在如风的少年<27°C>',
'description': 'md5:8d594db46cc3e6509107ede70a4aaa3b',
},
'playlist_count': 4,
'skip': 'Blocked outside Mainland China',
}, {
'note': 'This program has accompanying songs.',
'url': 'http://music.163.com/#/program?id=10141022',
'info_dict': {
'id': '10141022',
'ext': 'mp3',
'title': '25岁你是自在如风的少年<27°C>',
'description': 'md5:8d594db46cc3e6509107ede70a4aaa3b',
'timestamp': 1434450841,
'upload_date': '20150616',
},
'params': {
'noplaylist': True
},
'skip': 'Blocked outside Mainland China',
}]
def _real_extract(self, url):
program_id = self._match_id(url)
info = self.query_api(
'dj/program/detail?id=%s' % program_id,
program_id, 'Downloading program info')['program']
name = info['name']
description = info['description']
if not self._yes_playlist(info['songs'] and program_id, info['mainSong']['id']):
formats = self.extract_formats(info['mainSong'])
self._sort_formats(formats)
return {
'id': info['mainSong']['id'],
'title': name,
'description': description,
'creator': info['dj']['brand'],
'timestamp': self.convert_milliseconds(info['createTime']),
'thumbnail': info['coverUrl'],
'duration': self.convert_milliseconds(info.get('duration', 0)),
'formats': formats,
}
song_ids = [info['mainSong']['id']]
song_ids.extend([song['id'] for song in info['songs']])
entries = [
self.url_result('http://music.163.com/#/song?id=%s' % song_id,
'NetEaseMusic', song_id)
for song_id in song_ids
]
return self.playlist_result(entries, program_id, name, description)
class NetEaseMusicDjRadioIE(NetEaseMusicBaseIE):
IE_NAME = 'netease:djradio'
IE_DESC = '网易云音乐 - 电台'
_VALID_URL = r'https?://music\.163\.com/(#/)?djradio\?id=(?P<id>[0-9]+)'
_TEST = {
'url': 'http://music.163.com/#/djradio?id=42',
'info_dict': {
'id': '42',
'title': '声音蔓延',
'description': 'md5:766220985cbd16fdd552f64c578a6b15'
},
'playlist_mincount': 40,
'skip': 'Blocked outside Mainland China',
}
_PAGE_SIZE = 1000
def _real_extract(self, url):
dj_id = self._match_id(url)
name = None
desc = None
entries = []
for offset in compat_itertools_count(start=0, step=self._PAGE_SIZE):
info = self.query_api(
2015-07-10 06:30:23 -04:00
'dj/program/byradio?asc=false&limit=%d&radioId=%s&offset=%d'
% (self._PAGE_SIZE, dj_id, offset),
dj_id, 'Downloading dj programs - %d' % offset)
entries.extend([
self.url_result(
'http://music.163.com/#/program?id=%s' % program['id'],
'NetEaseMusicProgram', program['id'])
for program in info['programs']
])
if name is None:
radio = info['programs'][0]['radio']
name = radio['name']
desc = radio['desc']
if not info['more']:
break
return self.playlist_result(entries, dj_id, name, desc)