Update upstream source from tag 'upstream/2017.11.06'

[youtubedl] / youtube_dl / extractor / qqmusic.py

diff --git a/youtube_dl/extractor/qqmusic.py b/youtube_dl/extractor/qqmusic.py
index 13113820bc4c820012bd618104c1b192897a6985..084308aeb8922a1632eafe5a8395381710952c60 100644 (file)
--- a/youtube_dl/extractor/qqmusic.py
+++ b/youtube_dl/extractor/qqmusic.py
@@ -2,34 +2,70 @@
 from __future__ import unicode_literals
 
 import random
 from __future__ import unicode_literals
 
 import random

-import time

 import re
 import re

+import time

 
 from .common import InfoExtractor
 from ..utils import (
 
 from .common import InfoExtractor
 from ..utils import (

+    clean_html,
+    ExtractorError,

     strip_jsonp,
     unescapeHTML,
     strip_jsonp,
     unescapeHTML,

-    js_to_json,

 )
 )

-from ..compat import compat_urllib_request

 
 
 class QQMusicIE(InfoExtractor):
     IE_NAME = 'qqmusic'
 
 
 class QQMusicIE(InfoExtractor):
     IE_NAME = 'qqmusic'

-    _VALID_URL = r'http://y.qq.com/#type=song&mid=(?P<id>[0-9A-Za-z]+)'
+    IE_DESC = 'QQ音乐'
+    _VALID_URL = r'https?://y\.qq\.com/n/yqq/song/(?P<id>[0-9A-Za-z]+)\.html'

     _TESTS = [{
     _TESTS = [{

-        'url': 'http://y.qq.com/#type=song&mid=004295Et37taLD',
-        'md5': 'bed90b6db2a7a7a7e11bc585f471f63a',
+        'url': 'https://y.qq.com/n/yqq/song/004295Et37taLD.html',
+        'md5': '5f1e6cea39e182857da7ffc5ef5e6bb8',

         'info_dict': {
             'id': '004295Et37taLD',
         'info_dict': {
             'id': '004295Et37taLD',

-            'ext': 'm4a',
+            'ext': 'mp3',

             'title': '可惜没如果',
             'title': '可惜没如果',

-            'upload_date': '20141227',
+            'release_date': '20141227',

             'creator': '林俊杰',
             'creator': '林俊杰',

-            'description': 'md5:4348ff1dd24036906baa7b6f973f8d30',
+            'description': 'md5:d85afb3051952ecc50a1ee8a286d1eac',
+            'thumbnail': r're:^https?://.*\.jpg$',
+        }
+    }, {
+        'note': 'There is no mp3-320 version of this song.',
+        'url': 'https://y.qq.com/n/yqq/song/004MsGEo3DdNxV.html',
+        'md5': 'fa3926f0c585cda0af8fa4f796482e3e',
+        'info_dict': {
+            'id': '004MsGEo3DdNxV',
+            'ext': 'mp3',
+            'title': '如果',
+            'release_date': '20050626',
+            'creator': '李季美',
+            'description': 'md5:46857d5ed62bc4ba84607a805dccf437',
+            'thumbnail': r're:^https?://.*\.jpg$',

         }
         }

+    }, {
+        'note': 'lyrics not in .lrc format',
+        'url': 'https://y.qq.com/n/yqq/song/001JyApY11tIp6.html',
+        'info_dict': {
+            'id': '001JyApY11tIp6',
+            'ext': 'mp3',
+            'title': 'Shadows Over Transylvania',
+            'release_date': '19970225',
+            'creator': 'Dark Funeral',
+            'description': 'md5:c9b20210587cbcd6836a1c597bab4525',
+            'thumbnail': r're:^https?://.*\.jpg$',
+        },
+        'params': {
+            'skip_download': True,
+        },

     }]
 
     }]
 

+    _FORMATS = {
+        'mp3-320': {'prefix': 'M800', 'ext': 'mp3', 'preference': 40, 'abr': 320},
+        'mp3-128': {'prefix': 'M500', 'ext': 'mp3', 'preference': 30, 'abr': 128},
+        '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
     # Reference: m_r_GetRUin() in top_player.js
     # http://imgcache.gtimg.cn/music/portal_v3/y/top_player.js
     @staticmethod


@@ -60,6 +96,16 @@ class QQMusicIE(InfoExtractor):
         lrc_content = self._html_search_regex(
             r'<div class="content" id="lrc_content"[^<>]*>([^<>]+)</div>',
             detail_info_page, 'LRC lyrics', 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')
+
+        thumbnail_url = None
+        albummid = self._search_regex(
+            [r'albummid:\'([0-9a-zA-Z]+)\'', r'"albummid":"([0-9a-zA-Z]+)"'],
+            detail_info_page, 'album mid', default=None)
+        if albummid:
+            thumbnail_url = 'http://i.gtimg.cn/music/photo/mid_album_500/%s/%s/%s.jpg' \
+                            % (albummid[-2:-1], albummid[-1], albummid)

 
         guid = self.m_r_get_ruin()
 
 
         guid = self.m_r_get_ruin()
 


@@ -67,16 +113,41 @@ class QQMusicIE(InfoExtractor):
             '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']
             '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']

-        song_url = 'http://cc.stream.qqmusic.qq.com/C200%s.m4a?vkey=%s&guid=%s&fromtag=0' % (mid, vkey, guid)

 
 

-        return {
+        formats = []
+        for format_id, details in self._FORMATS.items():
+            formats.append({
+                'url': 'http://cc.stream.qqmusic.qq.com/%s%s.%s?vkey=%s&guid=%s&fromtag=0'
+                       % (details['prefix'], mid, details['ext'], vkey, guid),
+                'format': format_id,
+                'format_id': format_id,
+                'preference': details['preference'],
+                'abr': details.get('abr'),
+            })
+        self._check_formats(formats, mid)
+        self._sort_formats(formats)
+
+        actual_lrc_lyrics = ''.join(
+            line + '\n' for line in re.findall(
+                r'(?m)^(\[[0-9]{2}:[0-9]{2}(?:\.[0-9]{2,})?\][^\n]*|\[[^\]]*\])', lrc_content))
+
+        info_dict = {

             'id': mid,
             'id': mid,

-            'url': song_url,
+            'formats': formats,

             'title': song_name,
             'title': song_name,

-            'upload_date': publish_time,
+            'release_date': publish_time,

             'creator': singer,
             'description': lrc_content,
             'creator': singer,
             'description': lrc_content,

+            'thumbnail': thumbnail_url

         }
         }

+        if actual_lrc_lyrics:
+            info_dict['subtitles'] = {
+                'origin': [{
+                    'ext': 'lrc',
+                    'data': actual_lrc_lyrics,
+                }]
+            }
+        return info_dict

 
 
 class QQPlaylistBaseIE(InfoExtractor):
 
 
 class QQPlaylistBaseIE(InfoExtractor):


@@ -84,57 +155,72 @@ class QQPlaylistBaseIE(InfoExtractor):
     def qq_static_url(category, mid):
         return 'http://y.qq.com/y/static/%s/%s/%s/%s.html' % (category, mid[-2], mid[-1], mid)
 
     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):
+    def get_singer_all_songs(self, singmid, num):
+        return self._download_webpage(
+            r'https://c.y.qq.com/v8/fcg-bin/fcg_v8_singer_track_cp.fcg', singmid,
+            query={
+                'format': 'json',
+                'inCharset': 'utf8',
+                'outCharset': 'utf-8',
+                'platform': 'yqq',
+                'needNewCode': 0,
+                'singermid': singmid,
+                'order': 'listen',
+                'begin': 0,
+                'num': num,
+                'songstatus': 1,
+            })
+
+    def get_entries_from_page(self, singmid):

         entries = []
 
         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))
+        default_num = 1
+        json_text = self.get_singer_all_songs(singmid, default_num)
+        json_obj_all_songs = self._parse_json(json_text, singmid)
+
+        if json_obj_all_songs['code'] == 0:
+            total = json_obj_all_songs['data']['total']
+            json_text = self.get_singer_all_songs(singmid, total)
+            json_obj_all_songs = self._parse_json(json_text, singmid)
+
+        for item in json_obj_all_songs['data']['list']:
+            if item['musicData'].get('songmid') is not None:
+                songmid = item['musicData']['songmid']
+                entries.append(self.url_result(
+                    r'https://y.qq.com/n/yqq/song/%s.html' % songmid, 'QQMusic', songmid))

 
         return entries
 
 
 class QQMusicSingerIE(QQPlaylistBaseIE):
     IE_NAME = 'qqmusic:singer'
 
         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]+)'
+    IE_DESC = 'QQ音乐 - 歌手'
+    _VALID_URL = r'https?://y\.qq\.com/n/yqq/singer/(?P<id>[0-9A-Za-z]+)\.html'

     _TEST = {
     _TEST = {

-        'url': 'http://y.qq.com/#type=singer&mid=001BLpXF2DyJe2',
+        'url': 'https://y.qq.com/n/yqq/singer/001BLpXF2DyJe2.html',

         'info_dict': {
             'id': '001BLpXF2DyJe2',
             'title': '林俊杰',
         'info_dict': {
             'id': '001BLpXF2DyJe2',
             'title': '林俊杰',

-            'description': 'md5:2a222d89ba4455a3af19940c0481bb78',
+            'description': 'md5:870ec08f7d8547c29c93010899103751',

         },
         },

-        'playlist_count': 12,
+        'playlist_mincount': 12,

     }
 
     def _real_extract(self, url):
         mid = self._match_id(url)
 
     }
 
     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)
-
+        entries = self.get_entries_from_page(mid)
+        singer_page = self._download_webpage(url, mid, 'Download singer page')

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

         singer_desc = 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')
+        if mid:

             singer_desc_page = self._download_xml(
             singer_desc_page = self._download_xml(

-                req, mid, 'Donwload singer description XML')
+                'http://s.plcloud.music.qq.com/fcgi-bin/fcg_get_singer_desc.fcg', mid,
+                'Donwload singer description XML',
+                query={'utf8': 1, 'outCharset': 'utf-8', 'format': 'xml', 'singermid': mid},
+                headers={'Referer': 'https://y.qq.com/n/yqq/singer/'})

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


@@ -143,96 +229,141 @@ class QQMusicSingerIE(QQPlaylistBaseIE):
 
 class QQMusicAlbumIE(QQPlaylistBaseIE):
     IE_NAME = 'qqmusic:album'
 
 class QQMusicAlbumIE(QQPlaylistBaseIE):
     IE_NAME = 'qqmusic:album'

-    _VALID_URL = r'http://y.qq.com/#type=album&mid=(?P<id>[0-9A-Za-z]+)'
+    IE_DESC = 'QQ音乐 - 专辑'
+    _VALID_URL = r'https?://y\.qq\.com/n/yqq/album/(?P<id>[0-9A-Za-z]+)\.html'

 
 

-    _TEST = {
-        'url': 'http://y.qq.com/#type=album&mid=000gXCTb2AhRR1&play=0',
+    _TESTS = [{
+        'url': 'https://y.qq.com/n/yqq/album/000gXCTb2AhRR1.html',

         'info_dict': {
             'id': '000gXCTb2AhRR1',
             'title': '我们都是这样长大的',
         'info_dict': {
             'id': '000gXCTb2AhRR1',
             'title': '我们都是这样长大的',

-            'description': 'md5:d216c55a2d4b3537fe4415b8767d74d6',
+            'description': 'md5:179c5dce203a5931970d306aa9607ea6',

         },
         'playlist_count': 4,
         },
         'playlist_count': 4,

-    }
+    }, {
+        'url': 'https://y.qq.com/n/yqq/album/002Y5a3b3AlCu3.html',
+        'info_dict': {
+            'id': '002Y5a3b3AlCu3',
+            'title': '그리고...',
+            'description': 'md5:a48823755615508a95080e81b51ba729',
+        },
+        'playlist_count': 8,
+    }]

 
     def _real_extract(self, url):
         mid = self._match_id(url)
 
 
     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')
+        album = self._download_json(
+            'http://i.y.qq.com/v8/fcg-bin/fcg_v8_album_info_cp.fcg?albummid=%s&format=json' % mid,
+            mid, 'Download album page')['data']

 
 

-        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)
+        entries = [
+            self.url_result(
+                'https://y.qq.com/n/yqq/song/' + song['songmid'] + '.html', 'QQMusic', song['songmid']
+            ) for song in album['list']
+        ]
+        album_name = album.get('name')
+        album_detail = album.get('desc')
+        if album_detail is not None:
+            album_detail = album_detail.strip()

 
         return self.playlist_result(entries, mid, album_name, album_detail)
 
 
 class QQMusicToplistIE(QQPlaylistBaseIE):
     IE_NAME = 'qqmusic:toplist'
 
         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]+)'
+    IE_DESC = 'QQ音乐 - 排行榜'
+    _VALID_URL = r'https?://y\.qq\.com/n/yqq/toplist/(?P<id>[0-9]+)\.html'

 
     _TESTS = [{
 
     _TESTS = [{

-        'url': 'http://y.qq.com/#type=toplist&p=global_12',
+        'url': 'https://y.qq.com/n/yqq/toplist/123.html',

         'info_dict': {
         'info_dict': {

-            'id': 'global_12',
-            'title': 'itunes榜',
+            'id': '123',
+            'title': '美国iTunes榜',
+            'description': 'md5:89db2335fdbb10678dee2d43fe9aba08',

         },
         },

-        'playlist_count': 10,
+        'playlist_count': 100,

     }, {
     }, {

-        'url': 'http://y.qq.com/#type=toplist&p=top_6',
+        'url': 'https://y.qq.com/n/yqq/toplist/3.html',

         'info_dict': {
         'info_dict': {

-            'id': 'top_6',
-            'title': 'QQ音乐巅峰榜·欧美',
+            'id': '3',
+            'title': '巅峰榜·欧美',
+            'description': 'md5:5a600d42c01696b26b71f8c4d43407da',

         },
         'playlist_count': 100,
     }, {
         },
         'playlist_count': 100,
     }, {

-        'url': 'http://y.qq.com/#type=toplist&p=global_5',
+        'url': 'https://y.qq.com/n/yqq/toplist/106.html',

         'info_dict': {
         'info_dict': {

-            'id': 'global_5',
-            'title': '韩国mnet排行榜',
+            'id': '106',
+            'title': '韩国Mnet榜',
+            'description': 'md5:cb84b325215e1d21708c615cac82a6e7',

         },
         'playlist_count': 50,
     }]
 
         },
         '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)
 
     def _real_extract(self, url):
         list_id = self._match_id(url)
 

-        list_type, num_id = list_id.split("_")
+        toplist_json = self._download_json(
+            'http://i.y.qq.com/v8/fcg-bin/fcg_v8_toplist_cp.fcg', list_id,
+            note='Download toplist page',
+            query={'type': 'toplist', 'topid': list_id, 'format': 'json'})

 
 

-        list_page = self._download_webpage(
-            "http://y.qq.com/y/static/toplist/index/%s.html" % list_id,
-            list_id, 'Download toplist page')
+        entries = [self.url_result(
+            'https://y.qq.com/n/yqq/song/' + song['data']['songmid'] + '.html', 'QQMusic',
+            song['data']['songmid'])
+            for song in toplist_json['songlist']]

 
 

-        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
+        topinfo = toplist_json.get('topinfo', {})
+        list_name = topinfo.get('ListName')
+        list_description = topinfo.get('info')
+        return self.playlist_result(entries, list_id, list_name, list_description)

 
 

-        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))
+class QQMusicPlaylistIE(QQPlaylistBaseIE):
+    IE_NAME = 'qqmusic:playlist'
+    IE_DESC = 'QQ音乐 - 歌单'
+    _VALID_URL = r'https?://y\.qq\.com/n/yqq/playlist/(?P<id>[0-9]+)\.html'

 
 

-        list_name = self._html_search_regex(
-            r'<h2 id="top_name">([^\']+)</h2>', list_page, 'top list name',
-            default=None)
+    _TESTS = [{
+        'url': 'http://y.qq.com/n/yqq/playlist/3462654915.html',
+        'info_dict': {
+            'id': '3462654915',
+            'title': '韩国5月新歌精选下旬',
+            'description': 'md5:d2c9d758a96b9888cf4fe82f603121d4',
+        },
+        'playlist_count': 40,
+        'skip': 'playlist gone',
+    }, {
+        'url': 'https://y.qq.com/n/yqq/playlist/1374105607.html',
+        'info_dict': {
+            'id': '1374105607',
+            'title': '易入人心的华语民谣',
+            'description': '民谣的歌曲易于传唱、、歌词朗朗伤口、旋律简单温馨。属于那种才入耳孔。却上心头的感觉。没有太多的复杂情绪。简单而直接地表达乐者的情绪，就是这样的简单才易入人心。',
+        },
+        'playlist_count': 20,
+    }]
+
+    def _real_extract(self, url):
+        list_id = self._match_id(url)

 
 

-        return self.playlist_result(entries, list_id, list_name)
+        list_json = self._download_json(
+            'http://i.y.qq.com/qzone-music/fcg-bin/fcg_ucc_getcdinfo_byids_cp.fcg',
+            list_id, 'Download list page',
+            query={'type': 1, 'json': 1, 'utf8': 1, 'onlysong': 0, 'disstid': list_id},
+            transform_source=strip_jsonp)
+        if not len(list_json.get('cdlist', [])):
+            if list_json.get('code'):
+                raise ExtractorError(
+                    'QQ Music said: error %d in fetching playlist info' % list_json['code'],
+                    expected=True)
+            raise ExtractorError('Unable to get playlist info')
+
+        cdlist = list_json['cdlist'][0]
+        entries = [self.url_result(
+            'https://y.qq.com/n/yqq/song/' + song['songmid'] + '.html', 'QQMusic', song['songmid'])
+            for song in cdlist['songlist']]
+
+        list_name = cdlist.get('dissname')
+        list_description = clean_html(unescapeHTML(cdlist.get('desc')))
+        return self.playlist_result(entries, list_id, list_name, list_description)