[qqmusic] Add support for charts / top lists

diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py
index de19dfd7a8cfe1b2f08b8099087fd0301df384c9..8ec0c1032a1538f87a17a66e4211deb2db88418d 100644 (file)
--- a/youtube_dl/extractor/__init__.py
+++ b/youtube_dl/extractor/__init__.py
@@ -414,6 +414,7 @@ from .qqmusic import (
     QQMusicIE,
     QQMusicSingerIE,
     QQMusicAlbumIE,
+    QQMusicToplistIE,
 )
 from .quickvid import QuickVidIE
 from .r7 import R7IE

diff --git a/youtube_dl/extractor/qqmusic.py b/youtube_dl/extractor/qqmusic.py
index 174c8e0ae751fa9d1a875c666faa906d9729330f..d4a85d8c33a64bab3c650693831d9ef7fc99413b 100644 (file)
--- a/youtube_dl/extractor/qqmusic.py
+++ b/youtube_dl/extractor/qqmusic.py
@@ -9,6 +9,7 @@ from .common import InfoExtractor
 from ..utils import (
     strip_jsonp,
     unescapeHTML,
+    js_to_json,
 )
 from ..compat import compat_urllib_request
 
@@ -168,3 +169,57 @@ class QQMusicAlbumIE(QQPlaylistBaseIE):
             album_page, 'album details', default=None)
 
         return self.playlist_result(entries, mid, album_name, album_detail)
+
+
+class QQMusicToplistIE(QQPlaylistBaseIE):
+    _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,
+    }]
+
+    @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 = list_id.split("_")[0]
+        num_id = list_id.split("_")[1]
+
+        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':
+            list = self._download_json(
+                "http://y.qq.com/y/static/toplist/json/top/%s/1.js" % num_id,
+                list_id, note='Retrieve toplist json', errnote='Unable to get toplist json', transform_source=self.strip_qq_jsonp)
+
+            for song in list['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))
+
+        elif list_type == 'global':
+            entries = self.get_entries_from_page(list_page)
+
+        list_name = self._html_search_regex(
+            r'<h2 id="top_name">([^\']+)</h2>', list_page, 'top list name',
+            default=None)
+        list_desc = None
+
+        return self.playlist_result(entries, list_id, list_name, list_desc)
\ No newline at end of file