Merge branch 'ping-viki-shows'

[youtube-dl] / youtube_dl / extractor / sohu.py

diff --git a/youtube_dl/extractor/sohu.py b/youtube_dl/extractor/sohu.py
index 5adc734d9da1dc0b0aa8e50184f416403f5bd3d8..29bd9ce6f76247b5ac997050075c7e720d8a3b2b 100644 (file)
--- a/youtube_dl/extractor/sohu.py
+++ b/youtube_dl/extractor/sohu.py
@@ -8,7 +8,7 @@ from ..compat import (
     compat_str,
     compat_urllib_request
 )
-from ..utils import url_sanitize_consecutive_slashes
+from ..utils import ExtractorError
 
 
 class SohuIE(InfoExtractor):
@@ -23,12 +23,10 @@ class SohuIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'MV：Far East Movement《The Illest》',
         },
-        'params': {
-            'cn_verification_proxy': 'proxy.uku.im:8888'
-        }
+        'skip': 'On available in China',
     }, {
         'url': 'http://tv.sohu.com/20150305/n409385080.shtml',
-        'md5': '699060e75cf58858dd47fb9c03c42cfb',
+        'md5': 'ac9a5d322b4bf9ae184d53e4711e4f1a',
         'info_dict': {
             'id': '409385080',
             'ext': 'mp4',
@@ -36,12 +34,55 @@ class SohuIE(InfoExtractor):
         }
     }, {
         'url': 'http://my.tv.sohu.com/us/232799889/78693464.shtml',
-        'md5': '9bf34be48f2f4dadcb226c74127e203c',
+        'md5': '49308ff6dafde5ece51137d04aec311e',
         'info_dict': {
             'id': '78693464',
             'ext': 'mp4',
             'title': '【爱范品】第31期：MWC见不到的奇葩手机',
         }
+    }, {
+        'note': 'Multipart video',
+        'url': 'http://my.tv.sohu.com/pl/8384802/78910339.shtml',
+        'info_dict': {
+            'id': '78910339',
+            'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
+        },
+        'playlist': [{
+            'md5': '492923eac023ba2f13ff69617c32754a',
+            'info_dict': {
+                'id': '78910339_part1',
+                'ext': 'mp4',
+                'duration': 294,
+                'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
+            }
+        }, {
+            'md5': 'de604848c0e8e9c4a4dde7e1347c0637',
+            'info_dict': {
+                'id': '78910339_part2',
+                'ext': 'mp4',
+                'duration': 300,
+                'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
+            }
+        }, {
+            'md5': '93584716ee0657c0b205b8aa3d27aa13',
+            'info_dict': {
+                'id': '78910339_part3',
+                'ext': 'mp4',
+                'duration': 150,
+                'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
+            }
+        }]
+    }, {
+        'note': 'Video with title containing dash',
+        'url': 'http://my.tv.sohu.com/us/249884221/78932792.shtml',
+        'info_dict': {
+            'id': '78932792',
+            'ext': 'mp4',
+            'title': 'youtube-dl testing video',
+        },
+        'params': {
+            'skip_download': True
+        }
     }]
 
     def _real_extract(self, url):
@@ -58,23 +99,31 @@ class SohuIE(InfoExtractor):
             if cn_verification_proxy:
                 req.add_header('Ytdl-request-proxy', cn_verification_proxy)
 
-            return self._download_json(req, video_id,
-                                       'Downloading JSON data for %s' % vid_id)
+            return self._download_json(
+                req, video_id,
+                'Downloading JSON data for %s' % vid_id)
 
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
         mytv = mobj.group('mytv') is not None
 
         webpage = self._download_webpage(url, video_id)
-        raw_title = self._html_search_regex(
-            r'(?s)<title>(.+?)</title>',
-            webpage, 'video title')
-        title = raw_title.partition('-')[0].strip()
+
+        title = re.sub(r' - 搜狐视频$', '', self._og_search_title(webpage))
 
         vid = self._html_search_regex(
             r'var vid ?= ?["\'](\d+)["\']',
             webpage, 'video path')
         vid_data = _fetch_data(vid, mytv)
+        if vid_data['play'] != 1:
+            if vid_data.get('status') == 12:
+                raise ExtractorError(
+                    'Sohu said: There\'s something wrong in the video.',
+                    expected=True)
+            else:
+                raise ExtractorError(
+                    'Sohu said: The video is only licensed to users in Mainland China.',
+                    expected=True)
 
         formats_json = {}
         for format_id in ('nor', 'high', 'super', 'ori', 'h2644k', 'h2654k'):
@@ -90,24 +139,21 @@ class SohuIE(InfoExtractor):
         for i in range(part_count):
             formats = []
             for format_id, format_data in formats_json.items():
-                allot = format_data['allot']
-                prot = format_data['prot']
-
                 data = format_data['data']
-                clips_url = data['clipsURL']
-                su = data['su']
-
-                part_str = self._download_webpage(
-                    'http://%s/?prot=%s&file=%s&new=%s' %
-                    (allot, prot, clips_url[i], su[i]),
-                    video_id,
-                    'Downloading %s video URL part %d of %d'
-                    % (format_id, i + 1, part_count))
 
-                part_info = part_str.split('|')
-
-                video_url = url_sanitize_consecutive_slashes(
-                    '%s%s?key=%s' % (part_info[0], su[i], part_info[3]))
+                # URLs starts with http://newflv.sohu.ccgslb.net/ is not usable
+                # so retry until got a working URL
+                video_url = 'newflv.sohu.ccgslb.net'
+                retries = 0
+                while 'newflv.sohu.ccgslb.net' in video_url and retries < 5:
+                    download_note = 'Download information from CDN gateway for format ' + format_id
+                    if retries > 0:
+                        download_note += ' (retry #%d)' % retries
+                    retries += 1
+                    cdn_info = self._download_json(
+                        'http://data.vod.itc.cn/cdnList?new=' + data['su'][i],
+                        video_id, download_note)
+                    video_url = cdn_info['url']
 
                 formats.append({
                     'url': video_url,
@@ -131,9 +177,10 @@ class SohuIE(InfoExtractor):
             info['id'] = video_id
         else:
             info = {
-                '_type': 'playlist',
+                '_type': 'multi_video',
                 'entries': playlist,
                 'id': video_id,
+                'title': title,
             }
 
         return info