CriterionIE: simplify some parts and use _html_search_regex
authorJaime Marquínez Ferrándiz <jaime.marquinez.ferrandiz@gmail.com>
Sat, 13 Jul 2013 10:26:05 +0000 (12:26 +0200)
committerJaime Marquínez Ferrándiz <jaime.marquinez.ferrandiz@gmail.com>
Sat, 13 Jul 2013 10:26:05 +0000 (12:26 +0200)
youtube_dl/extractor/criterion.py

index a149d29007982e60463ec3c85bf8e239dbc882c0..31fe3d57b31ccddccadbbcbb66244ef6aea63cd2 100644 (file)
@@ -3,38 +3,38 @@
 import re
 
 from .common import InfoExtractor
+from ..utils import determine_ext
 
 class CriterionIE(InfoExtractor):
-    _VALID_URL = r'http://www.criterion.com/films/(.*)'
+    _VALID_URL = r'https?://www\.criterion\.com/films/(\d*)-.+'
     _TEST = {
         u'url': u'http://www.criterion.com/films/184-le-samourai',
         u'file': u'184.mp4',
         u'md5': u'bc51beba55685509883a9a7830919ec3',
         u'info_dict': {
             u"title": u"Le Samouraï",
-            u"description" : u"In a career-defining performance, Alain Delon plays a contract killer with samurai instincts. A razor-sharp cocktail of 1940s American gangster cinema and 1960s French pop culture, maverick director Jean-Pierre Melville&#x27;s masterpiece _Le Samouraï_ defines cool. "
+            u"description" : u'md5:a2b4b116326558149bef81f76dcbb93f',
         }
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group(1).split('-')[0]
+        video_id = mobj.group(1)
         webpage = self._download_webpage(url, video_id)
 
         final_url = self._search_regex(r'so.addVariable\("videoURL", "(.+?)"\)\;',
                                 webpage, 'video url')
-        title = self._search_regex(r'<meta content="(.+?)" property="og:title" />',
+        title = self._html_search_regex(r'<meta content="(.+?)" property="og:title" />',
                                 webpage, 'video title')
-        description = self._search_regex(r'<meta name="description" content="(.+?)" />',
+        description = self._html_search_regex(r'<meta name="description" content="(.+?)" />',
                                 webpage, 'video description')
         thumbnail = self._search_regex(r'so.addVariable\("thumbnailURL", "(.+?)"\)\;',
                                 webpage, 'thumbnail url')
-        ext = final_url.split('.')[-1]
 
         return {'id': video_id,
                 'url' : final_url,
                 'title': title,
-                'ext': ext,
+                'ext': determine_ext(final_url),
                 'description': description,
                 'thumbnail': thumbnail,
                 }