Merge pull request #8898 from dstftw/fragment-retries

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

diff --git a/youtube_dl/extractor/xhamster.py b/youtube_dl/extractor/xhamster.py
index ccee77359761b5214eb4ff24f8051f3dda1995eb..b3547174dd92beffafaf8f220b50b94a25f2fa2b 100644 (file)
--- a/youtube_dl/extractor/xhamster.py
+++ b/youtube_dl/extractor/xhamster.py
@@ -4,9 +4,9 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
+    dict_get,
     float_or_none,
     int_or_none,
-    str_to_int,
     unified_strdate,
 )
 
@@ -22,7 +22,7 @@ class XHamsterIE(InfoExtractor):
                 'title': 'FemaleAgent Shy beauty takes the bait',
                 'upload_date': '20121014',
                 'uploader': 'Ruseful2011',
-                'duration': 893,
+                'duration': 893.52,
                 'age_limit': 18,
             }
         },
@@ -34,7 +34,7 @@ class XHamsterIE(InfoExtractor):
                 'title': 'Britney Spears  Sexy Booty',
                 'upload_date': '20130914',
                 'uploader': 'jojo747400',
-                'duration': 200,
+                'duration': 200.48,
                 'age_limit': 18,
             }
         },
@@ -64,9 +64,10 @@ class XHamsterIE(InfoExtractor):
         webpage = self._download_webpage(mrss_url, video_id)
 
         title = self._html_search_regex(
-            [r'<title[^>]*>(.+?)(?:,\s*[^,]*?\s*Porn\s*[^,]*?:\s*xHamster[^<]*| - xHamster\.com)</title>',
-             r'<h1[^>]*>([^<]+)</h1>',
-             r'<meta[^>]+itemprop=".*?caption.*?"[^>]+content="(.+?)"'], webpage, 'title')
+            [r'<h1[^>]*>([^<]+)</h1>',
+             r'<meta[^>]+itemprop=".*?caption.*?"[^>]+content="(.+?)"',
+             r'<title[^>]*>(.+?)(?:,\s*[^,]*?\s*Porn\s*[^,]*?:\s*xHamster[^<]*| - xHamster\.com)</title>'],
+            webpage, 'title')
 
         # Only a few videos have an description
         mobj = re.search(r'<span>Description: </span>([^<]+)', webpage)
@@ -77,7 +78,7 @@ class XHamsterIE(InfoExtractor):
             webpage, 'upload date', fatal=False))
 
         uploader = self._html_search_regex(
-            r"<a href='[^']+xhamster\.com/user/[^>]+>(?P<uploader>[^<]+)",
+            r'<span[^>]+itemprop=["\']author[^>]+><a[^>]+href=["\'].+?xhamster\.com/user/[^>]+>(?P<uploader>.+?)</a>',
             webpage, 'uploader', default='anonymous')
 
         thumbnail = self._search_regex(
@@ -170,6 +171,12 @@ class XHamsterEmbedIE(InfoExtractor):
 
         video_url = self._search_regex(
             r'href="(https?://xhamster\.com/movies/%s/[^"]+\.html[^"]*)"' % video_id,
-            webpage, 'xhamster url')
+            webpage, 'xhamster url', default=None)
+
+        if not video_url:
+            vars = self._parse_json(
+                self._search_regex(r'vars\s*:\s*({.+?})\s*,\s*\n', webpage, 'vars'),
+                video_id)
+            video_url = dict_get(vars, ('downloadLink', 'homepageLink', 'commentsLink', 'shareUrl'))
 
         return self.url_result(video_url, 'XHamster')