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 725e01ac3ab6c158c7ad4ee9e5dcf413181322c6..b3547174dd92beffafaf8f220b50b94a25f2fa2b 100644 (file)
--- a/youtube_dl/extractor/xhamster.py
+++ b/youtube_dl/extractor/xhamster.py
@@ -4,11 +4,10 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
-    ExtractorError,
-    unified_strdate,
-    str_to_int,
+    dict_get,
+    float_or_none,
     int_or_none,
-    parse_duration,
+    unified_strdate,
 )
 
 
@@ -22,8 +21,8 @@ class XHamsterIE(InfoExtractor):
                 'ext': 'mp4',
                 'title': 'FemaleAgent Shy beauty takes the bait',
                 'upload_date': '20121014',
-                'uploader_id': 'Ruseful2011',
-                'duration': 893,
+                'uploader': 'Ruseful2011',
+                'duration': 893.52,
                 'age_limit': 18,
             }
         },
@@ -34,8 +33,8 @@ class XHamsterIE(InfoExtractor):
                 'ext': 'mp4',
                 'title': 'Britney Spears  Sexy Booty',
                 'upload_date': '20130914',
-                'uploader_id': 'jojo747400',
-                'duration': 200,
+                'uploader': 'jojo747400',
+                'duration': 200.48,
                 'age_limit': 18,
             }
         },
@@ -46,12 +45,12 @@ class XHamsterIE(InfoExtractor):
     ]
 
     def _real_extract(self, url):
-        def extract_video_url(webpage):
-            mp4 = re.search(r'<video\s+.*?file="([^"]+)".*?>', webpage)
-            if mp4 is None:
-                raise ExtractorError('Unable to extract media URL')
-            else:
-                return mp4.group(1)
+        def extract_video_url(webpage, name):
+            return self._search_regex(
+                [r'''file\s*:\s*(?P<q>["'])(?P<mp4>.+?)(?P=q)''',
+                 r'''<a\s+href=(?P<q>["'])(?P<mp4>.+?)(?P=q)\s+class=["']mp4Thumb''',
+                 r'''<video[^>]+file=(?P<q>["'])(?P<mp4>.+?)(?P=q)[^>]*>'''],
+                webpage, name, group='mp4')
 
         def is_hd(webpage):
             return '<div class=\'icon iconHD\'' in webpage
@@ -64,28 +63,36 @@ class XHamsterIE(InfoExtractor):
         mrss_url = '%s://xhamster.com/movies/%s/%s.html' % (proto, video_id, seo)
         webpage = self._download_webpage(mrss_url, video_id)
 
-        title = self._html_search_regex(r'<title>(?P<title>.+?) - xHamster\.com</title>', webpage, 'title')
+        title = self._html_search_regex(
+            [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)
         description = mobj.group(1) if mobj else None
 
-        upload_date = self._html_search_regex(r'hint=\'(\d{4}-\d{2}-\d{2}) \d{2}:\d{2}:\d{2} [A-Z]{3,4}\'',
-                                              webpage, 'upload date', fatal=False)
-        if upload_date:
-            upload_date = unified_strdate(upload_date)
+        upload_date = unified_strdate(self._search_regex(
+            r'hint=["\'](\d{4}-\d{2}-\d{2}) \d{2}:\d{2}:\d{2} [A-Z]{3,4}',
+            webpage, 'upload date', fatal=False))
 
-        uploader_id = self._html_search_regex(r'<a href=\'/user/[^>]+>(?P<uploader_id>[^<]+)',
-                                              webpage, 'uploader id', default='anonymous')
+        uploader = self._html_search_regex(
+            r'<span[^>]+itemprop=["\']author[^>]+><a[^>]+href=["\'].+?xhamster\.com/user/[^>]+>(?P<uploader>.+?)</a>',
+            webpage, 'uploader', default='anonymous')
 
-        thumbnail = self._html_search_regex(r'<video\s+.*?poster="([^"]+)".*?>', webpage, 'thumbnail', fatal=False)
+        thumbnail = self._search_regex(
+            [r'''thumb\s*:\s*(?P<q>["'])(?P<thumbnail>.+?)(?P=q)''',
+             r'''<video[^>]+poster=(?P<q>["'])(?P<thumbnail>.+?)(?P=q)[^>]*>'''],
+            webpage, 'thumbnail', fatal=False, group='thumbnail')
 
-        duration = parse_duration(self._html_search_regex(r'<span>Runtime:</span> (\d+:\d+)</div>',
-                                                          webpage, 'duration', fatal=False))
+        duration = float_or_none(self._search_regex(
+            r'(["\'])duration\1\s*:\s*(["\'])(?P<duration>.+?)\2',
+            webpage, 'duration', fatal=False, group='duration'))
 
-        view_count = self._html_search_regex(r'<span>Views:</span> ([^<]+)</div>', webpage, 'view count', fatal=False)
-        if view_count:
-            view_count = str_to_int(view_count)
+        view_count = int_or_none(self._search_regex(
+            r'content=["\']User(?:View|Play)s:(\d+)',
+            webpage, 'view count', fatal=False))
 
         mobj = re.search(r"hint='(?P<likecount>\d+) Likes / (?P<dislikecount>\d+) Dislikes'", webpage)
         (like_count, dislike_count) = (mobj.group('likecount'), mobj.group('dislikecount')) if mobj else (None, None)
@@ -97,7 +104,9 @@ class XHamsterIE(InfoExtractor):
 
         hd = is_hd(webpage)
 
-        video_url = extract_video_url(webpage)
+        format_id = 'hd' if hd else 'sd'
+
+        video_url = extract_video_url(webpage, format_id)
         formats = [{
             'url': video_url,
             'format_id': 'hd' if hd else 'sd',
@@ -108,7 +117,7 @@ class XHamsterIE(InfoExtractor):
             mrss_url = self._search_regex(r'<link rel="canonical" href="([^"]+)', webpage, 'mrss_url')
             webpage = self._download_webpage(mrss_url + '?hd', video_id, note='Downloading HD webpage')
             if is_hd(webpage):
-                video_url = extract_video_url(webpage)
+                video_url = extract_video_url(webpage, 'hd')
                 formats.append({
                     'url': video_url,
                     'format_id': 'hd',
@@ -122,7 +131,7 @@ class XHamsterIE(InfoExtractor):
             'title': title,
             'description': description,
             'upload_date': upload_date,
-            'uploader_id': uploader_id,
+            'uploader': uploader,
             'thumbnail': thumbnail,
             'duration': duration,
             'view_count': view_count,
@@ -162,6 +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');
+        return self.url_result(video_url, 'XHamster')