Merge branch 'master' of https://github.com/ytdl-org/youtube-dl into bugfix/youtube/chapters-fix-extractor

youtube_dl/extractor/biqle.py

@@ -3,10 +3,11 @@ from __future__ import unicode_literals
 
 from .common import InfoExtractor
 from .vk import VKIE
-from ..utils import (
-    HEADRequest,
-    int_or_none,
+from ..compat import (
+    compat_b64decode,
+    compat_urllib_parse_unquote,
 )
+from ..utils import int_or_none
 
 
 class BIQLEIE(InfoExtractor):
@@ -47,9 +48,16 @@ class BIQLEIE(InfoExtractor):
         if VKIE.suitable(embed_url):
             return self.url_result(embed_url, VKIE.ie_key(), video_id)
 
-        self._request_webpage(
-            HEADRequest(embed_url), video_id, headers={'Referer': url})
-        video_id, sig, _, access_token = self._get_cookies(embed_url)['video_ext'].value.split('%3A')
+        embed_page = self._download_webpage(
+            embed_url, video_id, headers={'Referer': url})
+        video_ext = self._get_cookies(embed_url).get('video_ext')
+        if video_ext:
+            video_ext = compat_urllib_parse_unquote(video_ext.value)
+        if not video_ext:
+            video_ext = compat_b64decode(self._search_regex(
+                r'video_ext\s*:\s*[\'"]([A-Za-z0-9+/=]+)',
+                embed_page, 'video_ext')).decode()
+        video_id, sig, _, access_token = video_ext.split(':')
         item = self._download_json(
             'https://api.vk.com/method/video.get', video_id,
             headers={'User-Agent': 'okhttp/3.4.1'}, query={

youtube_dl/extractor/xhamster.py

@@ -20,13 +20,13 @@ from ..utils import (
 
 
 class XHamsterIE(InfoExtractor):
-    _DOMAINS = r'(?:xhamster\.(?:com|one|desi)|xhms\.pro|xhamster[27]\.com)'
+    _DOMAINS = r'(?:xhamster\.(?:com|one|desi)|xhms\.pro|xhamster\d+\.com)'
     _VALID_URL = r'''(?x)
                     https?://
                         (?:.+?\.)?%s/
                         (?:
-                            movies/(?P<id>\d+)/(?P<display_id>[^/]*)\.html|
-                            videos/(?P<display_id_2>[^/]*)-(?P<id_2>\d+)
+                            movies/(?P<id>[\dA-Za-z]+)/(?P<display_id>[^/]*)\.html|
+                            videos/(?P<display_id_2>[^/]*)-(?P<id_2>[\dA-Za-z]+)
                         )
                     ''' % _DOMAINS
     _TESTS = [{
@@ -99,12 +99,21 @@ class XHamsterIE(InfoExtractor):
     }, {
         'url': 'https://xhamster2.com/videos/femaleagent-shy-beauty-takes-the-bait-1509445',
         'only_matching': True,
+    }, {
+        'url': 'https://xhamster11.com/videos/femaleagent-shy-beauty-takes-the-bait-1509445',
+        'only_matching': True,
+    }, {
+        'url': 'https://xhamster26.com/videos/femaleagent-shy-beauty-takes-the-bait-1509445',
+        'only_matching': True,
     }, {
         'url': 'http://xhamster.com/movies/1509445/femaleagent_shy_beauty_takes_the_bait.html',
         'only_matching': True,
     }, {
         'url': 'http://xhamster.com/movies/2221348/britney_spears_sexy_booty.html?hd',
         'only_matching': True,
+    }, {
+        'url': 'http://de.xhamster.com/videos/skinny-girl-fucks-herself-hard-in-the-forest-xhnBJZx',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -129,7 +138,7 @@ class XHamsterIE(InfoExtractor):
 
         initials = self._parse_json(
             self._search_regex(
-                r'window\.initials\s*=\s*({.+?})\s*;\s*\n', webpage, 'initials',
+                r'window\.initials\s*=\s*({.+?})\s*;', webpage, 'initials',
                 default='{}'),
             video_id, fatal=False)
         if initials: