]> Raphaƫl G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/redtube.py
debian/copyright: Fix lintian's dep5-copyright-license-name-not-unique.
[youtubedl] / youtube_dl / extractor / redtube.py
index 4295cf93a75188844bfe0838789e2efcb4363229..d6054d7175fd49a22117dd357bea7905f6e739be 100644 (file)
@@ -1,43 +1,35 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
+from ..utils import ExtractorError
 
 
 class RedTubeIE(InfoExtractor):
     _VALID_URL = r'http://(?:www\.)?redtube\.com/(?P<id>[0-9]+)'
     _TEST = {
         'url': 'http://www.redtube.com/66418',
-        'file': '66418.mp4',
-        # md5 varies from time to time, as in
-        # https://travis-ci.org/rg3/youtube-dl/jobs/14052463#L295
-        #'md5': u'7b8c22b5e7098a3e1c09709df1126d2d',
+        'md5': '7b8c22b5e7098a3e1c09709df1126d2d',
         'info_dict': {
-            "title": "Sucked on a toilet",
-            "age_limit": 18,
+            'id': '66418',
+            'ext': 'mp4',
+            'title': 'Sucked on a toilet',
+            'age_limit': 18,
         }
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-
-        video_id = mobj.group('id')
-        video_extension = 'mp4'
+        video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
-        self.report_extraction(video_id)
+        if any(s in webpage for s in ['video-deleted-info', '>This video has been removed']):
+            raise ExtractorError('Video %s has been removed' % video_id, expected=True)
 
         video_url = self._html_search_regex(
-            r'<source src="(.+?)" type="video/mp4">', webpage, u'video URL')
-
+            r'<source src="(.+?)" type="video/mp4">', webpage, 'video URL')
         video_title = self._html_search_regex(
             r'<h1 class="videoTitle[^"]*">(.+?)</h1>',
-            webpage, u'title')
-
-        video_thumbnail = self._html_search_regex(
-            r'playerInnerHTML.+?<img\s+src="(.+?)"',
-            webpage, u'thumbnail', fatal=False)
+            webpage, 'title')
+        video_thumbnail = self._og_search_thumbnail(webpage)
 
         # No self-labeling, but they describe themselves as
         # "Home of Videos Porno"
@@ -46,7 +38,7 @@ class RedTubeIE(InfoExtractor):
         return {
             'id': video_id,
             'url': video_url,
-            'ext': video_extension,
+            'ext': 'mp4',
             'title': video_title,
             'thumbnail': video_thumbnail,
             'age_limit': age_limit,