[pornhd] Fix metadata extraction

[youtube-dl.git] / youtube_dl / extractor / pornhd.py

diff --git a/youtube_dl/extractor/pornhd.py b/youtube_dl/extractor/pornhd.py
index 57c78ba52a994a9c2aff224470b86b913702241f..33faf5e583df034aa33c2b47ebcf1683b8ddef0f 100644 (file)
--- a/youtube_dl/extractor/pornhd.py
+++ b/youtube_dl/extractor/pornhd.py
@@ -12,7 +12,7 @@ from ..utils import (
 
 
 class PornHdIE(InfoExtractor):
-    _VALID_URL = r'http://(?:www\.)?pornhd\.com/(?:[a-z]{2,4}/)?videos/(?P<id>\d+)(?:/(?P<display_id>.+))?'
+    _VALID_URL = r'https?://(?:www\.)?pornhd\.com/(?:[a-z]{2,4}/)?videos/(?P<id>\d+)(?:/(?P<display_id>.+))?'
     _TEST = {
         'url': 'http://www.pornhd.com/videos/1962/sierra-day-gets-his-cum-all-over-herself-hd-porn-video',
         'md5': '956b8ca569f7f4d8ec563e2c41598441',
@@ -39,9 +39,10 @@ class PornHdIE(InfoExtractor):
             [r'<span[^>]+class=["\']video-name["\'][^>]*>([^<]+)',
              r'<title>(.+?) - .*?[Pp]ornHD.*?</title>'], webpage, 'title')
         description = self._html_search_regex(
-            r'<div class="description">([^<]+)</div>', webpage, 'description', fatal=False)
+            r'<(div|p)[^>]+class="description"[^>]*>(?P<value>[^<]+)</\1',
+            webpage, 'description', fatal=False, group='value')
         view_count = int_or_none(self._html_search_regex(
-            r'(\d+) views\s*</span>', webpage, 'view count', fatal=False))
+            r'(\d+) views\s*<', webpage, 'view count', fatal=False))
         thumbnail = self._search_regex(
             r"'poster'\s*:\s*'([^']+)'", webpage, 'thumbnail', fatal=False)