Merge branch 'jackyzy823-develop' into develop

* iQIYI: fix #334 and close #150
* QQ: fix #381

src/you_get/extractors/iqiyi.py

@@ -3,8 +3,62 @@
 __all__ = ['iqiyi_download']
 
 from ..common import *
+from uuid import uuid4
+from random import random,randint
+import json
+from math import floor
+import hashlib
+
+'''
+com.qiyi.player.core.model.def.DefinitonEnum
+bid meaning for quality
+0 none
+1 standard
+2 high
+3 super
+4 suprt-high
+5 fullhd
+10 4k
+96 topspeed
+
+'''
+
+
+def getVRSXORCode(arg1,arg2):
+    loc3=arg2 %3
+    if loc3 == 1:
+        return arg1^121
+    if loc3 == 2:
+        return arg1^72
+    return arg1^103
+
+
+def getVrsEncodeCode(vlink):
+    loc6=0
+    loc2=''
+    loc3=vlink.split("-")
+    loc4=len(loc3)
+    # loc5=loc4-1
+    for i in range(loc4-1,-1,-1):
+        loc6=getVRSXORCode(int(loc3[loc4-i-1],16),i)
+        loc2+=chr(loc6)
+    return loc2[::-1]
+
+def getVMS(tvid,vid,uid):
+    tm=randint(1000,2000)
+    vmsreq='http://cache.video.qiyi.com/vms?key=fvip&src=p'+"&tvId="+tvid+"&vid="+vid+"&vinfo=1&tm="+str(tm)+"&enc="+hashlib.new('md5',bytes('ts56gh'+str(tm)+tvid,"utf-8")).hexdigest()+"&qyid="+uid+"&tn="+str(random())
+    return json.loads(get_content(vmsreq))
+
+def getDispathKey(rid):
+    tp=")(*&^flash@#$%a"  #magic from swf
+    time=json.loads(get_content("http://data.video.qiyi.com/t?tn="+str(random())))["t"]
+    t=str(int(floor(int(time)/(10*60.0))))
+    return hashlib.new("md5",bytes(t+tp+rid,"utf-8")).hexdigest()
+
 
 def iqiyi_download(url, output_dir = '.', merge = True, info_only = False):
+    gen_uid=uuid4().hex
+
     html = get_html(url)
 
     tvid = r1(r'data-player-tvid="([^"]+)"', html)
@@ -12,26 +66,46 @@ def iqiyi_download(url, output_dir = '.', merge = True, info_only = False):
     assert tvid
     assert videoid
 
-    info_url = 'http://cache.video.qiyi.com/vj/%s/%s/' % (tvid, videoid)
-    info = get_html(info_url)
-    raise NotImplementedError('iqiyi')
-
-    from xml.dom.minidom import parseString
-    doc = parseString(info_xml)
-    title = doc.getElementsByTagName('title')[0].firstChild.nodeValue
-    size = int(doc.getElementsByTagName('totalBytes')[0].firstChild.nodeValue)
-    urls = [n.firstChild.nodeValue for n in doc.getElementsByTagName('file')]
-    assert urls[0].endswith('.f4v'), urls[0]
-
-    for i in range(len(urls)):
-        temp_url = "http://data.video.qiyi.com/%s" % urls[i].split("/")[-1].split(".")[0] + ".ts"
-        try:
-            response = request.urlopen(temp_url)
-        except request.HTTPError as e:
-            key = r1(r'key=(.*)', e.geturl())
-        assert key
-        urls[i] += "?key=%s" % key
+    info = getVMS(tvid,videoid,gen_uid)
+
+    title = info["data"]["vi"]["vn"]
+
+    #for highest qualities
+    #for http://www.iqiyi.com/v_19rrmmz5yw.html  not vp -> np
+    try:
+        if not info["data"]["vp"] and info["data"]['vp']["tkl"]=='' :
+            raise ValueError
+    except:
+        log.e("[Error] Do not support for iQIYI VIP video.")
+        exit(-1)
+
+    # assert info["data"]['vp']["tkl"]!=''
+    bid=0
+    for i in info["data"]["vp"]["tkl"][0]["vs"]:
+        if int(i["bid"])<=10 and int(i["bid"])>=bid:
+            bid=int(i["bid"])
+            video_links=i["fs"] 
+    #todo support choose quality  with cmdline
+
+    urls=[]
+    size=0
+    for i in video_links:
+        vlink=i["l"]
+        # print(vlink)
+        if not vlink.startswith("/"):
+            #vlink is encode
+            vlink=getVrsEncodeCode(vlink)
+        assert vlink.endswith(".f4v")
+        size+=i["b"]
+        key=getDispathKey(vlink.split("/")[-1].split(".")[0])
+        baseurl=info["data"]["vp"]["du"].split("/")
+        baseurl.insert(-1,key)
+        url="/".join(baseurl)+vlink+'?su='+gen_uid+'&client=&z=&bt=&ct=&tn='+str(randint(10000,20000))
+        urls.append(json.loads(get_content(url))["l"])
 
+    #download should be complete in 10 minutes
+    #because the url is generated before start downloading
+    #and the key may be expired after 10 minutes
     print_info(site_info, title, 'flv', size)
     if not info_only:
         download_urls(urls, title, 'flv', size, output_dir = output_dir, merge = merge)

src/you_get/extractors/qq.py

@@ -3,7 +3,7 @@
 __all__ = ['qq_download']
 
 from ..common import *
-
+import uuid
 #QQMUSIC
 #SINGLE
 #1. http://y.qq.com/#type=song&mid=000A9lMb0iEqwN
@@ -25,19 +25,36 @@ def qq_download_by_id(id, title=None, output_dir='.', merge=True, info_only=Fals
     doc_vl = doc_root.getElementsByTagName('vl')[0]
     doc_vi = doc_vl.getElementsByTagName('vi')[0]
     fn = doc_vi.getElementsByTagName('fn')[0].firstChild.data
-    fclip = doc_vi.getElementsByTagName('fclip')[0].firstChild.data
-    if int(fclip) > 0:
-        fn = fn[:-4] + "." + fclip + fn[-4:]
+    # fclip = doc_vi.getElementsByTagName('fclip')[0].firstChild.data
+    # fc=doc_vi.getElementsByTagName('fc')[0].firstChild.data
     fvkey = doc_vi.getElementsByTagName('fvkey')[0].firstChild.data
     doc_ul = doc_vi.getElementsByTagName('ul')
-    url = doc_ul[0].getElementsByTagName('url')[0].firstChild.data
-    url = url + fn + '?vkey=' + fvkey
 
-    _, ext, size = url_info(url)
+
+    url = doc_ul[0].getElementsByTagName('url')[1].firstChild.data
+
+    # print(i.firstChild.data)
+    urls=[]
+    ext=fn[-3:]
+    size=0
+    for i in doc.getElementsByTagName("cs"):
+        size+=int(i.firstChild.data)
+
+    # size=sum(map(int,doc.getElementsByTagName("cs")))
+    locid=str(uuid.uuid4())
+    for i in doc.getElementsByTagName("ci"):
+        urls.append(url+fn[:-4] + "." + i.getElementsByTagName("idx")[0].firstChild.data + fn[-4:] + '?vkey=' + fvkey+ '&sdtfrom=v1000&type='+ fn[-3:0] +'&locid=' + locid + "&&level=1&platform=11&br=133&fmt=hd&sp=0")
+        
+    print(urls)    
+    # if int(fclip) > 0:
+    #     fn = fn[:-4] + "." + fclip + fn[-4:]
+    # url = url + fn + '?vkey=' + fvkey
+
+    # _, ext, size = url_info(url)
 
     print_info(site_info, title, ext, size)
     if not info_only:
-        download_urls([url], title, ext, size, output_dir=output_dir, merge=merge)
+        download_urls(urls, title, ext, size, output_dir=output_dir, merge=merge)
 
 def qq_download(url, output_dir = '.', merge = True, info_only = False):
     if re.match(r'http://v.qq.com/([^\?]+)\?vid', url):