[youtube] Fix extraction (closes #20758, closes #20759, closes #20761, closes #20762...

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

diff --git a/youtube_dl/extractor/criterion.py b/youtube_dl/extractor/criterion.py
index d1fe7fc3c10485fa167eb4c556955afb05b3bbfe..f7815b905d13910e0a931f2609fa015c9ac3f00a 100644 (file)
--- a/youtube_dl/extractor/criterion.py
+++ b/youtube_dl/extractor/criterion.py
@@ -1,31 +1,39 @@
-# -*- coding: utf-8 -*-
-
-import re
+# coding: utf-8
+from __future__ import unicode_literals
 
 from .common import InfoExtractor
 
+
 class CriterionIE(InfoExtractor):
-    _VALID_URL = r'http://www.criterion.com/films/(.*)'
+    _VALID_URL = r'https?://(?:www\.)?criterion\.com/films/(?P<id>[0-9]+)-.+'
+    _TEST = {
+        'url': 'http://www.criterion.com/films/184-le-samourai',
+        'md5': 'bc51beba55685509883a9a7830919ec3',
+        'info_dict': {
+            'id': '184',
+            'ext': 'mp4',
+            'title': 'Le Samouraï',
+            'description': 'md5:a2b4b116326558149bef81f76dcbb93f',
+            'thumbnail': r're:^https?://.*\.jpg$',
+        }
+    }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group(1).split('-')[0]
+        video_id = self._match_id(url)
         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" />',
-                                webpage, 'video title')
-        description = self._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]
+        final_url = self._search_regex(
+            r'so\.addVariable\("videoURL", "(.+?)"\)\;', webpage, 'video url')
+        title = self._og_search_title(webpage)
+        description = self._html_search_meta('description', webpage)
+        thumbnail = self._search_regex(
+            r'so\.addVariable\("thumbnailURL", "(.+?)"\)\;',
+            webpage, 'thumbnail url')
 
-        return {'id': video_id,
-                'url' : final_url,
-                'title': title,
-                'ext': ext,
-                'description': description,
-                'thumbnail': thumbnail,
-                }
+        return {
+            'id': video_id,
+            'url': final_url,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+        }