X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fsunporno.py;h=e527aa97188b1860e054f8af7c7bd7a33301729e;hb=c3124c3085e6a9a83ee31ace3a7d528a324c42da;hp=2a193dee3dc84ed83cafa7ecf339043f2cf51c0a;hpb=7eb21356f98315b15fc05d9e616951511253076f;p=youtube-dl

diff --git a/youtube_dl/extractor/sunporno.py b/youtube_dl/extractor/sunporno.py
index 2a193dee3..e527aa971 100644
--- a/youtube_dl/extractor/sunporno.py
+++ b/youtube_dl/extractor/sunporno.py
@@ -6,6 +6,8 @@ from .common import InfoExtractor
 from ..utils import (
     parse_duration,
     int_or_none,
+    qualities,
+    determine_ext,
 )
 
 
@@ -19,39 +21,54 @@ class SunPornoIE(InfoExtractor):
             'ext': 'flv',
             'title': 'md5:0a400058e8105d39e35c35e7c5184164',
             'description': 'md5:a31241990e1bd3a64e72ae99afb325fb',
+            'thumbnail': 're:^https?://.*\.jpg$',
             'duration': 302,
+            'age_limit': 18,
         }
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = self._match_id(url)
 
         webpage = self._download_webpage(url, video_id)
 
-        video_url = self._html_search_regex(
-            r'videoSource\s*=\s*\'<source\s*src="([^"]*)"', webpage, 'video URL')
-
-        title = self._html_search_regex(r'<title>([^<]*)</title>', webpage, 'title')
-
-        description = self._html_search_regex(
-            r'<meta name="description" content="([^"]*)"', webpage, 'description', fatal=False)
-        
+        title = self._html_search_regex(
+            r'<title>([^<]+)</title>', webpage, 'title')
+        description = self._html_search_meta(
+            'description', webpage, 'description')
         thumbnail = self._html_search_regex(
-            r'poster="([^"]*)"', webpage, 'thumbnail', fatal=False)
+            r'poster="([^"]+)"', webpage, 'thumbnail', fatal=False)
 
         duration = parse_duration(self._search_regex(
-            r'<span>Duration: (\d+:\d+)</span>', webpage, 'duration', fatal=False))
+            r'itemprop="duration">\s*(\d+:\d+)\s*<',
+            webpage, 'duration', fatal=False))
 
         view_count = int_or_none(self._html_search_regex(
-            r'<span class="views">(\d+)</span>', webpage, 'view count', fatal=False))
+            r'class="views">(?:<noscript>)?\s*(\d+)\s*<',
+            webpage, 'view count', fatal=False))
+        comment_count = int_or_none(self._html_search_regex(
+            r'(\d+)</b> Comments?',
+            webpage, 'comment count', fatal=False))
+
+        formats = []
+        quality = qualities(['mp4', 'flv'])
+        for video_url in re.findall(r'<(?:source|video) src="([^"]+)"', webpage):
+            video_ext = determine_ext(video_url)
+            formats.append({
+                'url': video_url,
+                'format_id': video_ext,
+                'quality': quality(video_ext),
+            })
+        self._sort_formats(formats)
 
         return {
             'id': video_id,
-            'url': video_url,
             'title': title,
             'description': description,
             'thumbnail': thumbnail,
             'duration': duration,
             'view_count': view_count,
+            'comment_count': comment_count,
+            'formats': formats,
+            'age_limit': 18,
         }