[crackled] Improve, fix invalid regexes and extract more metadata

youtube_dl/extractor/cracked.py

@@ -1,23 +1,26 @@
-# coding: utf-8
 from __future__ import unicode_literals
 
 import re
 
 from .common import InfoExtractor
+from ..utils import (
+    parse_iso8601,
+    str_to_int,
+)
+
 
 class CrackedIE(InfoExtractor):
-    _VALID_URL = r'http?://.*?\.cracked\.com/video_+(?P<id>.*)_.*'
+    _VALID_URL = r'https?://(?:www\.)?cracked\.com/video_(?P<id>\d+)_[\da-z-]+\.html'
     _TEST = {
-        'url': 'http://www.cracked.com/video_18803_4-social-criticisms-hidden-in-sonic-hedgehog-games.html',
+        'url': 'http://www.cracked.com/video_19006_4-plot-holes-you-didnt-notice-in-your-favorite-movies.html',
-
+        'md5': '4b29a5eeec292cd5eca6388c7558db9e',
         'info_dict': {
-            'id': '18803',
+            'id': '19006',
             'ext': 'mp4',
-            'title': "4 Social Criticisms Hidden in 'Sonic the Hedgehog' Games | Cracked.com",
+            'title': '4 Plot Holes You Didn\'t Notice in Your Favorite Movies',
-            'height': 375,
+            'description': 'md5:3b909e752661db86007d10e5ec2df769',
-            'width': 666,
+            'timestamp': 1405659600,
-
+            'upload_date': '20140718',
-
         }
     }
 
@@ -26,21 +29,37 @@ class CrackedIE(InfoExtractor):
         video_id = mobj.group('id')
 
         webpage = self._download_webpage(url, video_id)
-        title = self._search_regex(r'<title>(.*?)</title>',webpage,'title')
-        video_url = self._search_regex(r'var CK_vidSrc = "+(.*)"',webpage,'url')
-        width = self._search_regex(r'width="(.*?)"',webpage,'width')
-        height = re.findall(r'height="(.*?)"',webpage)[1]
 
+        video_url = self._html_search_regex(
+            [r'var\s+CK_vidSrc\s*=\s*"([^"]+)"', r'<video\s+src="([^"]+)"'], webpage, 'video URL')
 
+        title = self._og_search_title(webpage)
+        description = self._og_search_description(webpage)
 
+        timestamp = self._html_search_regex(r'<time datetime="([^"]+)"', 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))
+        comment_count = str_to_int(self._html_search_regex(
+            r'<span 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:
+            width = int(m.group('width'))
+            height = int(m.group('height'))
+        else:
+            width = height = None
 
         return {
-            'url':video_url,
             'id': video_id,
-            'ext':'mp4',
+            'url':video_url,
-            'title':title,
+            'title': title,
-            'height':int(height),
+            'description': description,
-            'width':int(width)
+            'timestamp': timestamp,
-
+            'view_count': view_count,
-
+            'comment_count': comment_count,
+            'height': height,
+            'width': width,
         }