[sunporno] Fix metadata extraction (Closes #10316)

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

diff --git a/youtube_dl/extractor/sunporno.py b/youtube_dl/extractor/sunporno.py
index 263f09b4645fa8b6255f1216e99cab27afce2bee..4269f2a30409a2a6536457def53b39369c522ef0 100644 (file)
--- a/youtube_dl/extractor/sunporno.py
+++ b/youtube_dl/extractor/sunporno.py
@@ -15,10 +15,10 @@ class SunPornoIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?sunporno\.com/videos/(?P<id>\d+)'
     _TEST = {
         'url': 'http://www.sunporno.com/videos/807778/',
-        'md5': '6457d3c165fd6de062b99ef6c2ff4c86',
+        'md5': '507887e29033502f29dba69affeebfc9',
         'info_dict': {
             'id': '807778',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'md5:0a400058e8105d39e35c35e7c5184164',
             'description': 'md5:a31241990e1bd3a64e72ae99afb325fb',
             'thumbnail': 're:^https?://.*\.jpg$',
@@ -28,27 +28,32 @@ class SunPornoIE(InfoExtractor):
     }
 
     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)
 
-        title = self._html_search_regex(r'<title>([^<]+)</title>', webpage, 'title')
-        description = self._html_search_meta('description', webpage, 'description')
+        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)
 
         duration = parse_duration(self._search_regex(
-            r'Duration:\s*(\d+:\d+)\s*<', webpage, 'duration', fatal=False))
+            (r'itemprop="duration"[^>]*>\s*(\d+:\d+)\s*<',
+             r'>Duration:\s*<span[^>]+>\s*(\d+:\d+)\s*<'),
+            webpage, 'duration', fatal=False))
 
         view_count = int_or_none(self._html_search_regex(
-            r'class="views">\s*(\d+)\s*<', 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))
+            r'(\d+)</b> Comments?',
+            webpage, 'comment count', fatal=False, default=None))
 
         formats = []
         quality = qualities(['mp4', 'flv'])
-        for video_url in re.findall(r'<source src="([^"]+)"', webpage):
+        for video_url in re.findall(r'<(?:source|video) src="([^"]+)"', webpage):
             video_ext = determine_ext(video_url)
             formats.append({
                 'url': video_url,