X-Git-Url: http://git.cielonegro.org/gitweb.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fpornhd.py;h=33faf5e583df034aa33c2b47ebcf1683b8ddef0f;hb=d0d93f76ea0dd1dae15bdba6059815d9cc467b05;hp=954dfccb75954d50a9a46bc14bdb1d0dcbd5588c;hpb=cf372f0778e82cdc181a6173909589e640ac29fb;p=youtube-dl.git

diff --git a/youtube_dl/extractor/pornhd.py b/youtube_dl/extractor/pornhd.py
index 954dfccb7..33faf5e58 100644
--- 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',
@@ -36,26 +36,29 @@ class PornHdIE(InfoExtractor):
         webpage = self._download_webpage(url, display_id or video_id)
 
         title = self._html_search_regex(
-            r'<title>(.+) porn HD.+?</title>', webpage, 'title')
+            [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)
 
         quality = qualities(['sd', 'hd'])
         sources = json.loads(js_to_json(self._search_regex(
-            r"(?s)'sources'\s*:\s*(\{.+?\})\s*\}\);", webpage, 'sources')))
+            r"(?s)'sources'\s*:\s*(\{.+?\})\s*\}[;,)]",
+            webpage, 'sources')))
         formats = []
-        for container, s in sources.items():
-            for qname, video_url in s.items():
-                formats.append({
-                    'url': video_url,
-                    'container': container,
-                    'format_id': '%s-%s' % (container, qname),
-                    'quality': quality(qname),
-                })
+        for qname, video_url in sources.items():
+            if not video_url:
+                continue
+            formats.append({
+                'url': video_url,
+                'format_id': qname,
+                'quality': quality(qname),
+            })
         self._sort_formats(formats)
 
         return {