From ec40feeced8ba2f095573605548016f6d3b5697e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?D=C3=A9stin=20Reed?= <trox1972@users.noreply.github.com>
Date: Sat, 1 Oct 2016 14:51:22 +0200
Subject: [PATCH] [anysex] Improve metadata extraction

---
 youtube_dl/extractor/anysex.py | 26 +++++++++++++-------------
 1 file changed, 13 insertions(+), 13 deletions(-)

diff --git a/youtube_dl/extractor/anysex.py b/youtube_dl/extractor/anysex.py
index ad86d6e58..628b13e86 100644
--- a/youtube_dl/extractor/anysex.py
+++ b/youtube_dl/extractor/anysex.py
@@ -4,8 +4,11 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
-    parse_duration,
+    get_element_by_attribute,
+    get_element_by_class,
     int_or_none,
+    parse_duration,
+    js_to_json,
 )
 
 
@@ -19,6 +22,7 @@ class AnySexIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Busty and sexy blondie in her bikini strips for you',
             'description': 'md5:de9e418178e2931c10b62966474e1383',
+            'thumbnail': 're:^https?://.*\.jpg$',
             'categories': ['Erotic'],
             'duration': 270,
             'age_limit': 18,
@@ -26,24 +30,20 @@ class AnySexIE(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)
 
-        video_url = self._html_search_regex(r"video_url\s*:\s*'([^']+)'", webpage, 'video URL')
+        video_data = self._parse_json(self._search_regex(
+            r'var\s+flashvars\s*=\s*({[^}]+});', webpage, 'video data'),
+        video_id, transform_source=js_to_json)
 
+        video_url = video_data['video_url']
         title = self._html_search_regex(r'<title>(.*?)</title>', webpage, 'title')
-        description = self._html_search_regex(
-            r'<div class="description"[^>]*>([^<]+)</div>', webpage, 'description', fatal=False)
-        thumbnail = self._html_search_regex(
-            r'preview_url\s*:\s*\'(.*?)\'', webpage, 'thumbnail', fatal=False)
 
         categories = re.findall(
             r'<a href="http://anysex\.com/categories/[^"]+" title="[^"]*">([^<]+)</a>', webpage)
 
-        duration = parse_duration(self._search_regex(
-            r'<b>Duration:</b> (?:<q itemprop="duration">)?(\d+:\d+)', webpage, 'duration', fatal=False))
+        duration = parse_duration(get_element_by_attribute('itemprop', 'duration', webpage))
         view_count = int_or_none(self._html_search_regex(
             r'<b>Views:</b> (\d+)', webpage, 'view count', fatal=False))
 
@@ -52,8 +52,8 @@ class AnySexIE(InfoExtractor):
             'url': video_url,
             'ext': 'mp4',
             'title': title,
-            'description': description,
-            'thumbnail': thumbnail,
+            'description': get_element_by_class('description', webpage),
+            'thumbnail': video_data.get('preview_url'),
             'categories': categories,
             'duration': duration,
             'view_count': view_count,