[yt-dlp.git] / youtube_dl / extractor / qqmusic.py

# coding: utf-8
from __future__ import unicode_literals

import random
import time
import re

from .common import InfoExtractor
from ..utils import (
    strip_jsonp,
    unescapeHTML,
    js_to_json,
)
from ..compat import compat_urllib_request


class QQMusicIE(InfoExtractor):
    IE_NAME = 'qqmusic'
    _VALID_URL = r'http://y.qq.com/#type=song&mid=(?P<id>[0-9A-Za-z]+)'
    _TESTS = [{
        'url': 'http://y.qq.com/#type=song&mid=004295Et37taLD',
        'md5': '9ce1c1c8445f561506d2e3cfb0255705',
        'info_dict': {
            'id': '004295Et37taLD',
            'ext': 'mp3',
            'title': '可惜没如果',
            'upload_date': '20141227',
            'creator': '林俊杰',
            'description': 'md5:d327722d0361576fde558f1ac68a7065',
        }
    }]

    _FORMATS = {
        'mp3-320': {'prefix': 'M800', 'ext': 'mp3', 'preference': 40},
        'mp3-128': {'prefix': 'M500', 'ext': 'mp3', 'preference': 30},
        'm4a': {'prefix': 'C200', 'ext': 'm4a', 'preference': 10}
    }

    # Reference: m_r_GetRUin() in top_player.js
    # http://imgcache.gtimg.cn/music/portal_v3/y/top_player.js
    @staticmethod
    def m_r_get_ruin():
        curMs = int(time.time() * 1000) % 1000
        return int(round(random.random() * 2147483647) * curMs % 1E10)

    def _real_extract(self, url):
        mid = self._match_id(url)

        detail_info_page = self._download_webpage(
            'http://s.plcloud.music.qq.com/fcgi-bin/fcg_yqq_song_detail_info.fcg?songmid=%s&play=0' % mid,
            mid, note='Download song detail info',
            errnote='Unable to get song detail info', encoding='gbk')

        song_name = self._html_search_regex(
            r"songname:\s*'([^']+)'", detail_info_page, 'song name')

        publish_time = self._html_search_regex(
            r'发行时间：(\d{4}-\d{2}-\d{2})', detail_info_page,
            'publish time', default=None)
        if publish_time:
            publish_time = publish_time.replace('-', '')

        singer = self._html_search_regex(
            r"singer:\s*'([^']+)", detail_info_page, 'singer', default=None)

        lrc_content = self._html_search_regex(
            r'<div class="content" id="lrc_content"[^<>]*>([^<>]+)</div>',
            detail_info_page, 'LRC lyrics', default=None)
        if lrc_content:
            lrc_content = lrc_content.replace('\\n', '\n')

        guid = self.m_r_get_ruin()

        vkey = self._download_json(
            'http://base.music.qq.com/fcgi-bin/fcg_musicexpress.fcg?json=3&guid=%s' % guid,
            mid, note='Retrieve vkey', errnote='Unable to get vkey',
            transform_source=strip_jsonp)['key']

        formats = []
        for k, sf in self._FORMATS.items():
            formats.append({
                'url': 'http://cc.stream.qqmusic.qq.com/%s%s.%s?vkey=%s&guid=%s&fromtag=0'
                       % (sf['prefix'], mid, sf['ext'], vkey, guid),
                'format': k, 'format_id': k, 'preference': sf['preference']
            })
        self._sort_formats(formats)

        return {
            'id': mid,
            'formats': formats,
            'title': song_name,
            'upload_date': publish_time,
            'creator': singer,
            'description': lrc_content,
        }


class QQPlaylistBaseIE(InfoExtractor):
    @staticmethod
    def qq_static_url(category, mid):
        return 'http://y.qq.com/y/static/%s/%s/%s/%s.html' % (category, mid[-2], mid[-1], mid)

    @classmethod
    def get_entries_from_page(cls, page):
        entries = []

        for item in re.findall(r'class="data"[^<>]*>([^<>]+)</', page):
            song_mid = unescapeHTML(item).split('|')[-5]
            entries.append(cls.url_result(
                'http://y.qq.com/#type=song&mid=' + song_mid, 'QQMusic',
                song_mid))

        return entries


class QQMusicSingerIE(QQPlaylistBaseIE):
    IE_NAME = 'qqmusic:singer'
    _VALID_URL = r'http://y.qq.com/#type=singer&mid=(?P<id>[0-9A-Za-z]+)'
    _TEST = {
        'url': 'http://y.qq.com/#type=singer&mid=001BLpXF2DyJe2',
        'info_dict': {
            'id': '001BLpXF2DyJe2',
            'title': '林俊杰',
            'description': 'md5:2a222d89ba4455a3af19940c0481bb78',
        },
        'playlist_count': 12,
    }

    def _real_extract(self, url):
        mid = self._match_id(url)

        singer_page = self._download_webpage(
            self.qq_static_url('singer', mid), mid, 'Download singer page')

        entries = self.get_entries_from_page(singer_page)

        singer_name = self._html_search_regex(
            r"singername\s*:\s*'([^']+)'", singer_page, 'singer name',
            default=None)

        singer_id = self._html_search_regex(
            r"singerid\s*:\s*'([0-9]+)'", singer_page, 'singer id',
            default=None)

        singer_desc = None

        if singer_id:
            req = compat_urllib_request.Request(
                'http://s.plcloud.music.qq.com/fcgi-bin/fcg_get_singer_desc.fcg?utf8=1&outCharset=utf-8&format=xml&singerid=%s' % singer_id)
            req.add_header(
                'Referer', 'http://s.plcloud.music.qq.com/xhr_proxy_utf8.html')
            singer_desc_page = self._download_xml(
                req, mid, 'Donwload singer description XML')

            singer_desc = singer_desc_page.find('./data/info/desc').text

        return self.playlist_result(entries, mid, singer_name, singer_desc)


class QQMusicAlbumIE(QQPlaylistBaseIE):
    IE_NAME = 'qqmusic:album'
    _VALID_URL = r'http://y.qq.com/#type=album&mid=(?P<id>[0-9A-Za-z]+)'

    _TEST = {
        'url': 'http://y.qq.com/#type=album&mid=000gXCTb2AhRR1&play=0',
        'info_dict': {
            'id': '000gXCTb2AhRR1',
            'title': '我们都是这样长大的',
            'description': 'md5:d216c55a2d4b3537fe4415b8767d74d6',
        },
        'playlist_count': 4,
    }

    def _real_extract(self, url):
        mid = self._match_id(url)

        album_page = self._download_webpage(
            self.qq_static_url('album', mid), mid, 'Download album page')

        entries = self.get_entries_from_page(album_page)

        album_name = self._html_search_regex(
            r"albumname\s*:\s*'([^']+)',", album_page, 'album name',
            default=None)

        album_detail = self._html_search_regex(
            r'<div class="album_detail close_detail">\s*<p>((?:[^<>]+(?:<br />)?)+)</p>',
            album_page, 'album details', default=None)

        return self.playlist_result(entries, mid, album_name, album_detail)


class QQMusicToplistIE(QQPlaylistBaseIE):
    IE_NAME = 'qqmusic:toplist'
    _VALID_URL = r'http://y\.qq\.com/#type=toplist&p=(?P<id>(top|global)_[0-9]+)'

    _TESTS = [{
        'url': 'http://y.qq.com/#type=toplist&p=global_12',
        'info_dict': {
            'id': 'global_12',
            'title': 'itunes榜',
        },
        'playlist_count': 10,
    }, {
        'url': 'http://y.qq.com/#type=toplist&p=top_6',
        'info_dict': {
            'id': 'top_6',
            'title': 'QQ音乐巅峰榜·欧美',
        },
        'playlist_count': 100,
    }, {
        'url': 'http://y.qq.com/#type=toplist&p=global_5',
        'info_dict': {
            'id': 'global_5',
            'title': '韩国mnet排行榜',
        },
        'playlist_count': 50,
    }]

    @staticmethod
    def strip_qq_jsonp(code):
        return js_to_json(re.sub(r'^MusicJsonCallback\((.*?)\)/\*.+?\*/$', r'\1', code))

    def _real_extract(self, url):
        list_id = self._match_id(url)

        list_type, num_id = list_id.split("_")

        list_page = self._download_webpage(
            "http://y.qq.com/y/static/toplist/index/%s.html" % list_id,
            list_id, 'Download toplist page')

        entries = []
        if list_type == 'top':
            jsonp_url = "http://y.qq.com/y/static/toplist/json/top/%s/1.js" % num_id
        else:
            jsonp_url = "http://y.qq.com/y/static/toplist/json/global/%s/1_1.js" % num_id

        toplist_json = self._download_json(
            jsonp_url, list_id, note='Retrieve toplist json',
            errnote='Unable to get toplist json', transform_source=self.strip_qq_jsonp)

        for song in toplist_json['l']:
            s = song['s']
            song_mid = s.split("|")[20]
            entries.append(self.url_result(
                'http://y.qq.com/#type=song&mid=' + song_mid, 'QQMusic',
                song_mid))

        list_name = self._html_search_regex(
            r'<h2 id="top_name">([^\']+)</h2>', list_page, 'top list name',
            default=None)

        return self.playlist_result(entries, list_id, list_name)
Commit	Line	Data
5d98908b YCH	1	# coding: utf-8
	2	from __future__ import unicode_literals
	3
a2043572 YCH	4	import random
a2043572 YCH	5	import time
8afff9f8	6	import re
a2043572	7
5d98908b	8	from .common import InfoExtractor
5edea45f YCH	9	from ..utils import (
	10	strip_jsonp,
	11	unescapeHTML,
41333b97	12	js_to_json,
5edea45f	13	)
8afff9f8	14	from ..compat import compat_urllib_request
5d98908b	15
5d98908b YCH	16
5d98908b YCH	17	class QQMusicIE(InfoExtractor):
7ec676bb	18	IE_NAME = 'qqmusic'
5d98908b YCH	19	_VALID_URL = r'http://y.qq.com/#type=song&mid=(?P<id>[0-9A-Za-z]+)'
	20	_TESTS = [{
	21	'url': 'http://y.qq.com/#type=song&mid=004295Et37taLD',
55e5841f	22	'md5': '9ce1c1c8445f561506d2e3cfb0255705',
5d98908b YCH	23	'info_dict': {
5d98908b YCH	24	'id': '004295Et37taLD',
55e5841f	25	'ext': 'mp3',
5d98908b YCH	26	'title': '可惜没如果',
	27	'upload_date': '20141227',
	28	'creator': '林俊杰',
b813d8ca	29	'description': 'md5:d327722d0361576fde558f1ac68a7065',
5d98908b YCH	30	}
	31	}]
	32
55e5841f	33	_FORMATS = {
	34	'mp3-320': {'prefix': 'M800', 'ext': 'mp3', 'preference': 40},
	35	'mp3-128': {'prefix': 'M500', 'ext': 'mp3', 'preference': 30},
	36	'm4a': {'prefix': 'C200', 'ext': 'm4a', 'preference': 10}
	37	}
	38
a2043572 YCH	39	# Reference: m_r_GetRUin() in top_player.js
	40	# http://imgcache.gtimg.cn/music/portal_v3/y/top_player.js
	41	@staticmethod
	42	def m_r_get_ruin():
	43	curMs = int(time.time() * 1000) % 1000
	44	return int(round(random.random() * 2147483647) * curMs % 1E10)
	45
5d98908b YCH	46	def _real_extract(self, url):
	47	mid = self._match_id(url)
	48
	49	detail_info_page = self._download_webpage(
	50	'http://s.plcloud.music.qq.com/fcgi-bin/fcg_yqq_song_detail_info.fcg?songmid=%s&play=0' % mid,
8afff9f8	51	mid, note='Download song detail info',
c9a77969	52	errnote='Unable to get song detail info', encoding='gbk')
5d98908b YCH	53
	54	song_name = self._html_search_regex(
	55	r"songname:\s*'([^']+)'", detail_info_page, 'song name')
	56
	57	publish_time = self._html_search_regex(
	58	r'发行时间：(\d{4}-\d{2}-\d{2})', detail_info_page,
a685ae51 YCH	59	'publish time', default=None)
	60	if publish_time:
	61	publish_time = publish_time.replace('-', '')
5d98908b YCH	62
5d98908b YCH	63	singer = self._html_search_regex(
a685ae51 YCH	64	r"singer:\s*'([^']+)", detail_info_page, 'singer', default=None)
	65
	66	lrc_content = self._html_search_regex(
	67	r'<div class="content" id="lrc_content"[^<>]*>([^<>]+)</div>',
	68	detail_info_page, 'LRC lyrics', default=None)
b813d8ca YCH	69	if lrc_content:
b813d8ca YCH	70	lrc_content = lrc_content.replace('\\n', '\n')
5d98908b	71
a2043572 YCH	72	guid = self.m_r_get_ruin()
a2043572 YCH	73
5d98908b YCH	74	vkey = self._download_json(
	75	'http://base.music.qq.com/fcgi-bin/fcg_musicexpress.fcg?json=3&guid=%s' % guid,
	76	mid, note='Retrieve vkey', errnote='Unable to get vkey',
	77	transform_source=strip_jsonp)['key']
55e5841f	78
	79	formats = []
	80	for k, sf in self._FORMATS.items():
	81	formats.append({
	82	'url': 'http://cc.stream.qqmusic.qq.com/%s%s.%s?vkey=%s&guid=%s&fromtag=0'
	83	% (sf['prefix'], mid, sf['ext'], vkey, guid),
	84	'format': k, 'format_id': k, 'preference': sf['preference']
	85	})
	86	self._sort_formats(formats)
5d98908b YCH	87
	88	return {
	89	'id': mid,
55e5841f	90	'formats': formats,
5d98908b YCH	91	'title': song_name,
	92	'upload_date': publish_time,
	93	'creator': singer,
a685ae51	94	'description': lrc_content,
5d98908b	95	}
8afff9f8 YCH	96
8afff9f8 YCH	97
5edea45f YCH	98	class QQPlaylistBaseIE(InfoExtractor):
	99	@staticmethod
	100	def qq_static_url(category, mid):
	101	return 'http://y.qq.com/y/static/%s/%s/%s/%s.html' % (category, mid[-2], mid[-1], mid)
	102
5edea45f YCH	103	@classmethod
	104	def get_entries_from_page(cls, page):
	105	entries = []
	106
	107	for item in re.findall(r'class="data"[^<>]*>([^<>]+)</', page):
	108	song_mid = unescapeHTML(item).split('\|')[-5]
	109	entries.append(cls.url_result(
a685ae51 YCH	110	'http://y.qq.com/#type=song&mid=' + song_mid, 'QQMusic',
a685ae51 YCH	111	song_mid))
5edea45f YCH	112
	113	return entries
	114
	115
	116	class QQMusicSingerIE(QQPlaylistBaseIE):
7ec676bb	117	IE_NAME = 'qqmusic:singer'
8afff9f8 YCH	118	_VALID_URL = r'http://y.qq.com/#type=singer&mid=(?P<id>[0-9A-Za-z]+)'
	119	_TEST = {
	120	'url': 'http://y.qq.com/#type=singer&mid=001BLpXF2DyJe2',
	121	'info_dict': {
	122	'id': '001BLpXF2DyJe2',
	123	'title': '林俊杰',
	124	'description': 'md5:2a222d89ba4455a3af19940c0481bb78',
	125	},
	126	'playlist_count': 12,
	127	}
	128
	129	def _real_extract(self, url):
	130	mid = self._match_id(url)
	131
	132	singer_page = self._download_webpage(
5edea45f	133	self.qq_static_url('singer', mid), mid, 'Download singer page')
8afff9f8	134
5edea45f	135	entries = self.get_entries_from_page(singer_page)
8afff9f8 YCH	136
	137	singer_name = self._html_search_regex(
	138	r"singername\s:\s'([^']+)'", singer_page, 'singer name',
	139	default=None)
	140
	141	singer_id = self._html_search_regex(
	142	r"singerid\s:\s'([0-9]+)'", singer_page, 'singer id',
	143	default=None)
	144
	145	singer_desc = None
	146
	147	if singer_id:
	148	req = compat_urllib_request.Request(
	149	'http://s.plcloud.music.qq.com/fcgi-bin/fcg_get_singer_desc.fcg?utf8=1&outCharset=utf-8&format=xml&singerid=%s' % singer_id)
	150	req.add_header(
	151	'Referer', 'http://s.plcloud.music.qq.com/xhr_proxy_utf8.html')
	152	singer_desc_page = self._download_xml(
5edea45f	153	req, mid, 'Donwload singer description XML')
8afff9f8 YCH	154
	155	singer_desc = singer_desc_page.find('./data/info/desc').text
	156
	157	return self.playlist_result(entries, mid, singer_name, singer_desc)
5edea45f YCH	158
	159
	160	class QQMusicAlbumIE(QQPlaylistBaseIE):
7ec676bb	161	IE_NAME = 'qqmusic:album'
5edea45f YCH	162	_VALID_URL = r'http://y.qq.com/#type=album&mid=(?P<id>[0-9A-Za-z]+)'
	163
	164	_TEST = {
	165	'url': 'http://y.qq.com/#type=album&mid=000gXCTb2AhRR1&play=0',
	166	'info_dict': {
	167	'id': '000gXCTb2AhRR1',
	168	'title': '我们都是这样长大的',
	169	'description': 'md5:d216c55a2d4b3537fe4415b8767d74d6',
	170	},
	171	'playlist_count': 4,
	172	}
	173
	174	def _real_extract(self, url):
	175	mid = self._match_id(url)
	176
	177	album_page = self._download_webpage(
	178	self.qq_static_url('album', mid), mid, 'Download album page')
	179
	180	entries = self.get_entries_from_page(album_page)
	181
	182	album_name = self._html_search_regex(
	183	r"albumname\s:\s'([^']+)',", album_page, 'album name',
	184	default=None)
	185
	186	album_detail = self._html_search_regex(
	187	r'<div class="album_detail close_detail">\s*<p>((?:[^<>]+(?:<br />)?)+)</p>',
	188	album_page, 'album details', default=None)
	189
	190	return self.playlist_result(entries, mid, album_name, album_detail)
41333b97	191
	192
	193	class QQMusicToplistIE(QQPlaylistBaseIE):
7ec676bb	194	IE_NAME = 'qqmusic:toplist'
41333b97	195	_VALID_URL = r'http://y\.qq\.com/#type=toplist&p=(?P<id>(top\|global)_[0-9]+)'
54889739	196
41333b97	197	_TESTS = [{
	198	'url': 'http://y.qq.com/#type=toplist&p=global_12',
	199	'info_dict': {
	200	'id': 'global_12',
	201	'title': 'itunes榜',
	202	},
	203	'playlist_count': 10,
	204	}, {
	205	'url': 'http://y.qq.com/#type=toplist&p=top_6',
	206	'info_dict': {
	207	'id': 'top_6',
	208	'title': 'QQ音乐巅峰榜·欧美',
	209	},
	210	'playlist_count': 100,
fd4eefed	211	}, {
	212	'url': 'http://y.qq.com/#type=toplist&p=global_5',
	213	'info_dict': {
	214	'id': 'global_5',
	215	'title': '韩国mnet排行榜',
	216	},
	217	'playlist_count': 50,
41333b97	218	}]
	219
	220	@staticmethod
	221	def strip_qq_jsonp(code):
	222	return js_to_json(re.sub(r'^MusicJsonCallback\((.?)\)/\.+?\*/$', r'\1', code))
54889739	223
41333b97	224	def _real_extract(self, url):
	225	list_id = self._match_id(url)
	226
29ea5728	227	list_type, num_id = list_id.split("_")
41333b97	228
b480e787	229	list_page = self._download_webpage(
54889739	230	"http://y.qq.com/y/static/toplist/index/%s.html" % list_id,
b480e787	231	list_id, 'Download toplist page')
fd4eefed	232
41333b97	233	entries = []
41333b97	234	if list_type == 'top':
fd4eefed	235	jsonp_url = "http://y.qq.com/y/static/toplist/json/top/%s/1.js" % num_id
86ec1e48	236	else:
fd4eefed	237	jsonp_url = "http://y.qq.com/y/static/toplist/json/global/%s/1_1.js" % num_id
54889739	238
29ea5728	239	toplist_json = self._download_json(
54889739	240	jsonp_url, list_id, note='Retrieve toplist json',
fd4eefed	241	errnote='Unable to get toplist json', transform_source=self.strip_qq_jsonp)
54889739	242
29ea5728	243	for song in toplist_json['l']:
fd4eefed	244	s = song['s']
	245	song_mid = s.split("\|")[20]
	246	entries.append(self.url_result(
	247	'http://y.qq.com/#type=song&mid=' + song_mid, 'QQMusic',
	248	song_mid))
41333b97	249
	250	list_name = self._html_search_regex(
	251	r'<h2 id="top_name">([^\']+)</h2>', list_page, 'top list name',
	252	default=None)
41333b97	253
29ea5728	254	return self.playlist_result(entries, list_id, list_name)