]> git.bitcoin.ninja Git - youtube-dl/commitdiff
[manyvids] Improve (closes #14059)
authorSergey M․ <dstftw@gmail.com>
Sun, 3 Sep 2017 10:30:02 +0000 (17:30 +0700)
committerSergey M․ <dstftw@gmail.com>
Sun, 3 Sep 2017 10:32:23 +0000 (17:32 +0700)
youtube_dl/extractor/manyvids.py

index ea739ce3ff7e29ccf679616c3bda36acbf420865..b94b3c2abe333a2ef194abe2186775b182145b9d 100644 (file)
@@ -2,35 +2,47 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..compat import compat_urllib_parse_unquote
+from ..utils import int_or_none
 
 
 class ManyVidsIE(InfoExtractor):
-    _VALID_URL = r'https?://www.manyvids\.com/Video/(?P<id>[0-9]+)'
+    _VALID_URL = r'(?i)https?://(?:www\.)?manyvids\.com/video/(?P<id>\d+)'
     _TEST = {
         'url': 'https://www.manyvids.com/Video/133957/everthing-about-me/',
         'md5': '03f11bb21c52dd12a05be21a5c7dcc97',
         'info_dict': {
             'id': '133957',
             'ext': 'mp4',
-            'title': 'everthing about me',
-
-        }
+            'title': 'everthing about me (Preview)',
+            'view_count': int,
+            'like_count': int,
+        },
     }
 
     def _real_extract(self, url):
-        formats = []
         video_id = self._match_id(url)
+
         webpage = self._download_webpage(url, video_id)
-        video_url = compat_urllib_parse_unquote(self._search_regex(
-            r'data-video-filepath=\"(.+?)\"', webpage, 'video URL', default=''))
 
-        title = self._html_search_regex(r'<h2[^>]+class="m-a-0"[^>]*>([^<]+)', webpage, 'title')
-        formats.append({
-            'url': video_url
-        })
+        video_url = self._search_regex(
+            r'data-(?:video-filepath|meta-video)\s*=s*(["\'])(?P<url>(?:(?!\1).)+)\1',
+            webpage, 'video URL', group='url')
+
+        title = '%s (Preview)' % self._html_search_regex(
+            r'<h2[^>]+class="m-a-0"[^>]*>([^<]+)', webpage, 'title')
+
+        like_count = int_or_none(self._search_regex(
+            r'data-likes=["\'](\d+)', webpage, 'like count', default=None))
+        view_count = int_or_none(self._html_search_regex(
+            r'(?s)<span[^>]+class="views-wrapper"[^>]*>(.+?)</span', webpage,
+            'view count', default=None))
+
         return {
             'id': video_id,
             'title': title,
-            'formats': formats,
+            'view_count': view_count,
+            'like_count': like_count,
+            'formats': [{
+                'url': video_url,
+            }],
         }