From 506c5b5bc12f3b3597354cb31212b305227462c5 Mon Sep 17 00:00:00 2001
From: ehds <grephale@gmail.com>
Date: Sun, 29 Jul 2018 14:03:10 +0800
Subject: [PATCH] upadate kugou api

---
 src/you_get/extractors/kugou.py | 63 ++++++++++++++++++++++++++-------
 1 file changed, 50 insertions(+), 13 deletions(-)

diff --git a/src/you_get/extractors/kugou.py b/src/you_get/extractors/kugou.py
index 925bdf1c..a98daac6 100644
--- a/src/you_get/extractors/kugou.py
+++ b/src/you_get/extractors/kugou.py
@@ -20,32 +20,69 @@ def kugou_download(url, output_dir=".", merge=True, info_only=False, **kwargs):
         print_info(site_info, title, songtype, size)
         if not info_only:
             download_urls([url], title, ext, size, output_dir, merge=merge)
+    elif url.lower().find("hash")!=-1:
+        return kugou_download_by_hash(url,output_dir,merge,info_only)
     else:
         #for the www.kugou.com/
         return kugou_download_playlist(url, output_dir=output_dir, merge=merge, info_only=info_only)
         # raise NotImplementedError(url)       
 
-def kugou_download_by_hash(title,hash_val,output_dir = '.', merge = True, info_only = False):
+
+def kugou_download_by_hash(url,output_dir = '.', merge = True, info_only = False):
     #sample
-    #url_sample:http://www.kugou.com/yy/album/single/536957.html
-    #hash ->key  md5(hash+kgcloud")->key  decompile swf
-    #cmd 4 for mp3 cmd 3 for m4a
-    key=hashlib.new('md5',(hash_val+"kgcloud").encode("utf-8")).hexdigest()
-    html=get_html("http://trackercdn.kugou.com/i/?pid=6&key=%s&acceptMp3=1&cmd=4&hash=%s"%(key,hash_val))
-    j=loads(html)
-    url=j['url']
+    #url_sample:http://www.kugou.com/song/#hash=93F7D2FC6E95424739448218B591AEAF&album_id=9019462
+    hash_val = match1(url,'hash=(\w+)')
+    album_id = match1(url,'album_id=(\d+)')
+    html = get_html("http://www.kugou.com/yy/index.php?r=play/getdata&hash={}&album_id={}".format(hash_val,album_id))
+    j =loads(html)
+    url = j['data']['play_url']
+    title = j['data']['audio_name']
+    # some songs cann't play because of copyright protection
+    if(url == ''):
+        return
     songtype, ext, size = url_info(url)
     print_info(site_info, title, songtype, size)
     if not info_only:
         download_urls([url], title, ext, size, output_dir, merge=merge)
 
 def kugou_download_playlist(url, output_dir = '.', merge = True, info_only = False, **kwargs):
-    html=get_html(url)
-    pattern=re.compile('title="(.*?)".* data="(\w*)\|.*?"')
-    pairs=pattern.findall(html)
-    for title,hash_val in pairs:
-        kugou_download_by_hash(title,hash_val,output_dir,merge,info_only)
+    urls=[]
+    
+    #download music leaderboard
+    #sample: http://www.kugou.com/yy/html/rank.html
+    if url.lower().find('rank') !=-1:
+        html=get_html(url)
+        pattern = re.compile('<a href="(http://.*?)" data-active=')
+        res = pattern.findall(html)
+        for song in res:
+            res = get_html(song)
+            pattern_url = re.compile('"hash":"(\w+)".*"album_id":(\d)+')
+            hash_val,album_id= res = pattern_url.findall(res)[0]
+            urls.append('http://www.kugou.com/song/#hash=%s&album_id=%s'%(hash_val,album_id))
+    
+    # download album
+    # album sample:   http://www.kugou.com/yy/album/single/1645030.html
+    elif url.lower().find('album')!=-1:
+        html = get_html(url)
+        pattern = re.compile('var data=(\[.*?\]);')
+        res = pattern.findall(html)[0]
+        for v in json.loads(res):
+            urls.append('http://www.kugou.com/song/#hash=%s&album_id=%s'%(v['hash'],v['album_id']))
 
+    # download the playlist        
+    # playlist sample:http://www.kugou.com/yy/special/single/487279.html
+    else:
+        html = get_html(url)
+        pattern = re.compile('data="(\w+)\|(\d+)"')
+        for v in pattern.findall(html):
+            urls.append('http://www.kugou.com/song/#hash=%s&album_id=%s'%(v[0],v[1]))
+            print('http://www.kugou.com/song/#hash=%s&album_id=%s'%(v[0],v[1]))
+
+    #download the list by hash
+    for url in urls:
+        kugou_download_by_hash(url,output_dir,merge,info_only)
+
+                
 
 site_info = "kugou.com"
 download = kugou_download