[youtube] Fix extraction.

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

diff --git a/youtube_dl/extractor/cracked.py b/youtube_dl/extractor/cracked.py
index 74b880ffce9b0a8dbf7ab273063253a2743dbbdb..f77a68ecedf74bc5192fbc475848f6b1a1234315 100644 (file)
--- a/youtube_dl/extractor/cracked.py
+++ b/youtube_dl/extractor/cracked.py
@@ -3,6 +3,7 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from .youtube import YoutubeIE
 from ..utils import (
     parse_iso8601,
     str_to_int,
@@ -11,39 +12,63 @@ from ..utils import (
 
 class CrackedIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?cracked\.com/video_(?P<id>\d+)_[\da-z-]+\.html'
-    _TEST = {
+    _TESTS = [{
+        'url': 'http://www.cracked.com/video_19070_if-animal-actors-got-e21-true-hollywood-stories.html',
+        'md5': '89b90b9824e3806ca95072c4d78f13f7',
+        'info_dict': {
+            'id': '19070',
+            'ext': 'mp4',
+            'title': 'If Animal Actors Got E! True Hollywood Stories',
+            'timestamp': 1404954000,
+            'upload_date': '20140710',
+        }
+    }, {
+        # youtube embed
         'url': 'http://www.cracked.com/video_19006_4-plot-holes-you-didnt-notice-in-your-favorite-movies.html',
-        'md5': '4b29a5eeec292cd5eca6388c7558db9e',
+        'md5': 'ccd52866b50bde63a6ef3b35016ba8c7',
         'info_dict': {
-            'id': '19006',
+            'id': 'EjI00A3rZD0',
             'ext': 'mp4',
-            'title': '4 Plot Holes You Didn\'t Notice in Your Favorite Movies',
-            'description': 'md5:3b909e752661db86007d10e5ec2df769',
-            'timestamp': 1405659600,
-            'upload_date': '20140718',
+            'title': "4 Plot Holes You Didn't Notice in Your Favorite Movies - The Spit Take",
+            'description': 'md5:c603708c718b796fe6079e2b3351ffc7',
+            'upload_date': '20140725',
+            'uploader_id': 'Cracked',
+            'uploader': 'Cracked',
         }
-    }
+    }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = self._match_id(url)
 
         webpage = self._download_webpage(url, video_id)
 
+        youtube_url = YoutubeIE._extract_url(webpage)
+        if youtube_url:
+            return self.url_result(youtube_url, ie=YoutubeIE.ie_key())
+
         video_url = self._html_search_regex(
-            [r'var\s+CK_vidSrc\s*=\s*"([^"]+)"', r'<video\s+src="([^"]+)"'], webpage, 'video URL')
+            [r'var\s+CK_vidSrc\s*=\s*"([^"]+)"', r'<video\s+src="([^"]+)"'],
+            webpage, 'video URL')
+
+        title = self._search_regex(
+            [r'property="?og:title"?\s+content="([^"]+)"', r'class="?title"?>([^<]+)'],
+            webpage, 'title')
 
-        title = self._og_search_title(webpage)
-        description = self._og_search_description(webpage)
+        description = self._search_regex(
+            r'name="?(?:og:)?description"?\s+content="([^"]+)"',
+            webpage, 'description', default=None)
 
-        timestamp = self._html_search_regex(r'<time datetime="([^"]+)"', webpage, 'upload date', fatal=False)
+        timestamp = self._html_search_regex(
+            r'"date"\s*:\s*"([^"]+)"', webpage, 'upload date', fatal=False)
         if timestamp:
             timestamp = parse_iso8601(timestamp[:-6])
 
         view_count = str_to_int(self._html_search_regex(
-            r'<span class="views" id="viewCounts">([\d,\.]+) Views</span>', webpage, 'view count', fatal=False))
+            r'<span\s+class="?views"? id="?viewCounts"?>([\d,\.]+) Views</span>',
+            webpage, 'view count', fatal=False))
         comment_count = str_to_int(self._html_search_regex(
-            r'<span id="commentCounts">([\d,\.]+)</span>', webpage, 'comment count', fatal=False))
+            r'<span\s+id="?commentCounts"?>([\d,\.]+)</span>',
+            webpage, 'comment count', fatal=False))
 
         m = re.search(r'_(?P<width>\d+)X(?P<height>\d+)\.mp4$', video_url)
         if m:
@@ -54,7 +79,7 @@ class CrackedIE(InfoExtractor):
 
         return {
             'id': video_id,
-            'url':video_url,
+            'url': video_url,
             'title': title,
             'description': description,
             'timestamp': timestamp,
@@ -62,4 +87,4 @@ class CrackedIE(InfoExtractor):
             'comment_count': comment_count,
             'height': height,
             'width': width,
-        }
\ No newline at end of file
+        }