Switch codebase to use sanitized_Request instead of

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

diff --git a/youtube_dl/extractor/vk.py b/youtube_dl/extractor/vk.py
index 93c5b18c9bb8b8c3c16c6fbf849f28bdbaf349c2..d99a42a9f1dd982b50cc2b324714d6a75cb15d9b 100644 (file)
--- a/youtube_dl/extractor/vk.py
+++ b/youtube_dl/extractor/vk.py
@@ -8,19 +8,21 @@ from .common import InfoExtractor
 from ..compat import (
     compat_str,
     compat_urllib_parse,
-    compat_urllib_request,
 )
 from ..utils import (
     ExtractorError,
     orderedSet,
+    sanitized_Request,
     str_to_int,
     unescapeHTML,
     unified_strdate,
 )
+from .vimeo import VimeoIE
 
 
 class VKIE(InfoExtractor):
-    IE_NAME = 'vk.com'
+    IE_NAME = 'vk'
+    IE_DESC = 'VK'
     _VALID_URL = r'''(?x)
                     https?://
                         (?:
@@ -153,6 +155,11 @@ class VKIE(InfoExtractor):
             'url': 'http://vk.com/feed?z=video-43215063_166094326%2Fbb50cacd3177146d7a',
             'only_matching': True,
         },
+        {
+            # age restricted video, requires vk account credentials
+            'url': 'https://vk.com/video205387401_164765225',
+            'only_matching': True,
+        },
         {
             # vk wrapper
             'url': 'http://www.biqle.ru/watch/847655_160197695',
@@ -175,7 +182,7 @@ class VKIE(InfoExtractor):
             'pass': password.encode('cp1251'),
         })
 
-        request = compat_urllib_request.Request(
+        request = sanitized_Request(
             'https://login.vk.com/?act=login',
             compat_urllib_parse.urlencode(login_form).encode('utf-8'))
         login_page = self._download_webpage(
@@ -204,6 +211,12 @@ class VKIE(InfoExtractor):
 
         info_page = self._download_webpage(info_url, video_id)
 
+        error_message = self._html_search_regex(
+            r'(?s)<!><div[^>]+class="video_layer_message"[^>]*>(.+?)</div>',
+            info_page, 'error message', default=None)
+        if error_message:
+            raise ExtractorError(error_message, expected=True)
+
         if re.search(r'<!>/login\.php\?.*\bact=security_check', info_page):
             raise ExtractorError(
                 'You are trying to log in from an unusual location. You should confirm ownership at vk.com to log in with this IP.',
@@ -237,6 +250,10 @@ class VKIE(InfoExtractor):
         if youtube_url:
             return self.url_result(youtube_url, 'Youtube')
 
+        vimeo_url = VimeoIE._extract_vimeo_url(url, info_page)
+        if vimeo_url is not None:
+            return self.url_result(vimeo_url)
+
         m_rutube = re.search(
             r'\ssrc="((?:https?:)?//rutube\.ru\\?/video\\?/embed(?:.*?))\\?"', info_page)
         if m_rutube is not None:
@@ -264,9 +281,13 @@ class VKIE(InfoExtractor):
             mobj.group(1) + ' ' + mobj.group(2)
             upload_date = unified_strdate(mobj.group(1) + ' ' + mobj.group(2))
 
-        view_count = str_to_int(self._search_regex(
-            r'"mv_views_count_number"[^>]*>([\d,.]+) views<',
-            info_page, 'view count', fatal=False))
+        view_count = None
+        views = self._html_search_regex(
+            r'"mv_views_count_number"[^>]*>(.+?\bviews?)<',
+            info_page, 'view count', fatal=False)
+        if views:
+            view_count = str_to_int(self._search_regex(
+                r'([\d,.]+)', views, 'view count', fatal=False))
 
         formats = [{
             'format_id': k,
@@ -289,8 +310,8 @@ class VKIE(InfoExtractor):
 
 
 class VKUserVideosIE(InfoExtractor):
-    IE_NAME = 'vk.com:user-videos'
-    IE_DESC = 'vk.com:All of a user\'s videos'
+    IE_NAME = 'vk:uservideos'
+    IE_DESC = "VK - User's Videos"
     _VALID_URL = r'https?://vk\.com/videos(?P<id>-?[0-9]+)$'
     _TEMPLATE_URL = 'https://vk.com/videos'
     _TESTS = [{
@@ -313,7 +334,7 @@ class VKUserVideosIE(InfoExtractor):
         entries = [
             self.url_result(
                 'http://vk.com/video' + video_id, 'VK', video_id=video_id)
-            for video_id in set(re.findall(r'href="/video(-?[0-9_]+)"', webpage))]
+            for video_id in orderedSet(re.findall(r'href="/video(-?[0-9_]+)"', webpage))]
 
         title = unescapeHTML(self._search_regex(
             r'<title>\s*([^<]+?)\s+\|\s+\d+\s+videos',