From 8663ef14bdde7cda3f0bc77ed112c2dd545d5270 Mon Sep 17 00:00:00 2001
From: Zhang Ning <zhangn1985@gmail.com>
Date: Fri, 4 Sep 2015 18:45:35 +0800
Subject: [PATCH] support my.tv.sohu.com

try tv.sohu.com first
then my.tv.sohu.com

Signed-off-by: Zhang Ning <zhangn1985@gmail.com>
---
 src/you_get/extractors/sohu.py | 61 ++++++++++++++++++++--------------
 1 file changed, 36 insertions(+), 25 deletions(-)

diff --git a/src/you_get/extractors/sohu.py b/src/you_get/extractors/sohu.py
index 39be6026..df894e02 100644
--- a/src/you_get/extractors/sohu.py
+++ b/src/you_get/extractors/sohu.py
@@ -23,47 +23,58 @@ class Sohu(VideoExtractor):
         {'id': 'superVid', 'container': 'mp4', 'video_profile': '超清'},
         {'id': 'highVid', 'container': 'mp4', 'video_profile': '高清'},
         {'id': 'norVid', 'container': 'mp4', 'video_profile': '标清'},
-        {'id': 'relativeId', 'container': 'mp4', 'video_profile': '当前'},
     ]
 
     realurls = { 'oriVid': [], 'superVid': [], 'highVid': [], 'norVid': [], 'relativeId': []}
-    vids = {}
+    vids = { 'oriVid': 0, 'superVid': 0, 'highVid': 0, 'norVid': 0, 'relativeId': 0}
 
     def real_url(host, vid, tvid, new, clipURL, ck):
         return 'http://'+host+'/?prot=9&prod=flash&pt=1&file='+clipURL+'&new='+new +'&key='+ ck+'&vid='+str(vid)+'&uid='+str(int(time.time()*1000))+'&t='+str(random())
 
 
-    def get_vid_from_url(url):
-        return match1(url, 'id=(\d+)')
-
     def get_vid_from_content(content):
-        return match1(content, '\/([0-9]+)\/v\.swf')
+        return match1(content, '\/([0-9]+)\/v\.swf') or \
+                match1(content, '\&id=(\d+)')
+
+    def parser_info(self, info, stream, lvid):
+        host = info['allot']
+        prot = info['prot']
+        tvid = info['tvid']
+        data = info['data']
+        size = sum(map(int,data['clipsBytes']))
+        assert len(data['clipsURL']) == len(data['clipsBytes']) == len(data['su'])
+        for new, clip, ck, in zip(data['su'], data['clipsURL'], data['ck']):
+            clipURL = urlparse(clip).path
+            self.realurls[stream['id']].append(self.__class__.real_url(host, lvid, tvid, new, clipURL, ck))
+        self.streams[stream['id']] = {'container': 'mp4', 'video_profile': stream['video_profile'], 'size' : size}
+        self.vids[stream['id']] = lvid
+
 
     def prepare(self, **kwargs):
         assert self.url or self.vid
 
         if self.url and not self.vid:
-            self.vid = self.__class__.get_vid_from_url(self.url) or \
-                       self.__class__.get_vid_from_content(str(get_decoded_html(self.url)))
+            self.vid = self.__class__.get_vid_from_content(str(get_decoded_html(self.url)))
 
         info = json.loads(get_decoded_html('http://hot.vrs.sohu.com/vrs_flash.action?vid=%s' % self.vid))
-        data = info['data']
-        self.title = data['tvName']
-        for stream in self.stream_types:
-            lvid = data[stream['id']]
-            if lvid != 0 and lvid != self.vid :
-                info = json.loads(get_decoded_html('http://hot.vrs.sohu.com/vrs_flash.action?vid=%s' % lvid))
-                data = info['data']
-            host = info['allot']
-            prot = info['prot']
-            tvid = info['tvid']
-            size = sum(data['clipsBytes'])
-            assert len(data['clipsURL']) == len(data['clipsBytes']) == len(data['su'])
-            for new, clip, ck, in zip(data['su'], data['clipsURL'], data['ck']):
-                clipURL = urlparse(clip).path
-                self.realurls[stream['id']].append(self.__class__.real_url(host, lvid, tvid, new, clipURL, ck))
-            self.streams[stream['id']] = {'container': 'mp4', 'video_profile': stream['video_profile'], 'size' : size}
-            self.vids[stream['id']] = lvid
+        if info['status'] == 1:
+            data = info['data']
+            self.title = data['tvName']
+            for stream in self.stream_types:
+                lvid = data[stream['id']]
+                if lvid != 0 and lvid != self.vid :
+                    info = json.loads(get_decoded_html('http://hot.vrs.sohu.com/vrs_flash.action?vid=%s' % lvid))
+                self.parser_info(info, stream, lvid)
+            return
+        info = json.loads(get_decoded_html('http://my.tv.sohu.com/play/videonew.do?vid=%s&referer=http://my.tv.sohu.com' % self.vid))
+        if info['status'] == 1:
+            data = info['data']
+            self.title = data['tvName']
+            for stream in self.stream_types:
+                lvid = data[stream['id']]
+                info = json.loads(get_decoded_html('http://my.tv.sohu.com/play/videonew.do?vid=%s&referer=http://my.tv.sohu.com' % lvid))
+                self.parser_info(info, stream, lvid)
+            return
 
     def extract(self, **kwargs):
         if 'stream_id' in kwargs and kwargs['stream_id']: