From 6c3828cc32625f3caab6e82bb291a40f2416ff7f Mon Sep 17 00:00:00 2001
From: David Zhuang <i+david.zhuang-mail.utoronto.ca@cnbeining.com>
Date: Fri, 6 May 2016 16:30:08 -0400
Subject: [PATCH] [MangGuoTV]Add support

---
 src/you_get/common.py              |   1 +
 src/you_get/extractors/__init__.py |   1 +
 src/you_get/extractors/mgtv.py     | 110 +++++++++++++++++++++++++++++
 3 files changed, 112 insertions(+)
 create mode 100644 src/you_get/extractors/mgtv.py

diff --git a/src/you_get/common.py b/src/you_get/common.py
index e20be32b..284b1660 100755
--- a/src/you_get/common.py
+++ b/src/you_get/common.py
@@ -47,6 +47,7 @@ SITES = {
     'lizhi'            : 'lizhi',
     'magisto'          : 'magisto',
     'metacafe'         : 'metacafe',
+    'mgtv'             : 'mgtv',
     'miomio'           : 'miomio',
     'mixcloud'         : 'mixcloud',
     'mtv81'            : 'mtv81',
diff --git a/src/you_get/extractors/__init__.py b/src/you_get/extractors/__init__.py
index 5af9cdd3..6c3864fd 100755
--- a/src/you_get/extractors/__init__.py
+++ b/src/you_get/extractors/__init__.py
@@ -38,6 +38,7 @@ from .le import *
 from .lizhi import *
 from .magisto import *
 from .metacafe import *
+from .mgtv import *
 from .miaopai import *
 from .miomio import *
 from .mixcloud import *
diff --git a/src/you_get/extractors/mgtv.py b/src/you_get/extractors/mgtv.py
new file mode 100644
index 00000000..a3b7a06e
--- /dev/null
+++ b/src/you_get/extractors/mgtv.py
@@ -0,0 +1,110 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+
+from ..common import *
+from ..extractor import VideoExtractor
+
+from json import loads
+from urllib.parse import urlsplit
+from os.path import dirname
+
+class MGTV(VideoExtractor):
+    name = "芒果 (MGTV)"
+
+    # Last updated: 2015-11-24
+    stream_types = [
+        {'id': 'hd', 'container': 'flv', 'video_profile': '超清'},
+        {'id': 'sd', 'container': 'flv', 'video_profile': '高清'},
+        {'id': 'ld', 'container': 'flv', 'video_profile': '标清'},
+    ]
+    
+    id_dic = {i['video_profile']:(i['id']) for i in stream_types}
+    
+    api_endpoint = 'http://v.api.mgtv.com/player/video?video_id={video_id}'
+
+    @staticmethod
+    def get_vid_from_url(url):
+        """Extracts video ID from URL.
+        """
+        return match1(url, 'http://www.mgtv.com/v/\d/\d+/\w+/(\d+).html')
+    
+    #----------------------------------------------------------------------
+    @staticmethod
+    def get_mgtv_real_url(url):
+        """str->list of str
+        Give you the real URLs."""
+        content = loads(get_content(url))
+        m3u_url = content['info']
+        split = urlsplit(m3u_url)
+        
+        base_url = "{scheme}://{netloc}{path}/".format(scheme = split[0],
+                                                      netloc = split[1],
+                                                      path = dirname(split[2]))
+
+        content = get_content(content['info'])  #get the REAL M3U url, maybe to be changed later?
+        segment_list = []
+        for i in content.split():
+            if not i.startswith('#'):  #not the best way, better we use the m3u8 package
+                segment_list.append(base_url + i)
+        return segment_list
+
+    def download_playlist_by_url(self, url, **kwargs):
+        pass
+
+    def prepare(self, **kwargs):
+        if self.url:
+            self.vid = self.get_vid_from_url(self.url)
+        content = get_content(self.api_endpoint.format(video_id = self.vid))
+        content = loads(content)
+        self.title = content['data']['info']['title']
+        
+        #stream_avalable = [i['name'] for i in content['data']['stream']]
+        stream_available = {}
+        for i in content['data']['stream']:
+            stream_available[i['name']] = i['url']
+
+        for s in self.stream_types:
+            if s['video_profile'] in stream_available.keys():
+                quality_id = self.id_dic[s['video_profile']]
+                url = stream_available[s['video_profile']]
+                segment_list_this = self.get_mgtv_real_url(url)
+                
+                container_this_stream = ''
+                size_this_stream = 0
+                stream_fileid_list = []
+                for i in segment_list_this:
+                    _, container_this_stream, size_this_seg = url_info(i)
+                    size_this_stream += size_this_seg
+                    stream_fileid_list.append(os.path.basename(i).split('.')[0])
+                    
+            #make pieces
+            pieces = []
+            for i in zip(stream_fileid_list, segment_list_this):
+                pieces.append({'fileid': i[0], 'segs': i[1],})
+
+                self.streams[quality_id] = {
+                        'container': 'flv',
+                        'video_profile': s['video_profile'],
+                        'size': size_this_stream,
+                        'pieces': pieces
+                    }
+                
+            if not kwargs['info_only']:
+                self.streams[quality_id]['src'] = segment_list_this
+
+    def extract(self, **kwargs):
+        if 'stream_id' in kwargs and kwargs['stream_id']:
+            # Extract the stream
+            stream_id = kwargs['stream_id']
+
+            if stream_id not in self.streams:
+                log.e('[Error] Invalid video format.')
+                log.e('Run \'-i\' command with no specific video format to view all available formats.')
+                exit(2)
+        else:
+            # Extract stream with the best quality
+            stream_id = self.streams_sorted[0]['id']
+
+site = MGTV()
+download = site.download_by_url
+download_playlist = site.download_playlist_by_url
\ No newline at end of file