Merge pull request #7045 from remitamine/ign

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

diff --git a/youtube_dl/extractor/wimp.py b/youtube_dl/extractor/wimp.py
index 84f065a3df97c4d4b1c40c37cac5d4494d6ff8f6..041ff6c555123d44c97bc63810d2aa7903ec069e 100644 (file)
--- a/youtube_dl/extractor/wimp.py
+++ b/youtube_dl/extractor/wimp.py
@@ -1,28 +1,54 @@
-import re
-import base64
+from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from .youtube import YoutubeIE
 
 
 class WimpIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?(?:www\.)?wimp\.com/([^/]+)/'
+    _VALID_URL = r'http://(?:www\.)?wimp\.com/(?P<id>[^/]+)'
+    _TESTS = [{
+        'url': 'http://www.wimp.com/maruexhausted/',
+        'md5': 'ee21217ffd66d058e8b16be340b74883',
+        'info_dict': {
+            'id': 'maruexhausted',
+            'ext': 'mp4',
+            'title': 'Maru is exhausted.',
+            'description': 'md5:57e099e857c0a4ea312542b684a869b8',
+        }
+    }, {
+        'url': 'http://www.wimp.com/clowncar/',
+        'md5': '4e2986c793694b55b37cf92521d12bb4',
+        'info_dict': {
+            'id': 'clowncar',
+            'ext': 'mp4',
+            'title': 'It\'s like a clown car.',
+            'description': 'md5:0e56db1370a6e49c5c1d19124c0d2fb2',
+        },
+    }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group(1)
+        video_id = self._match_id(url)
+
         webpage = self._download_webpage(url, video_id)
-        title = self._search_regex(r'<meta name="description" content="(.+?)" />',webpage, 'video title')
-        thumbnail_url = self._search_regex(r'<meta property="og\:image" content="(.+?)" />', webpage,'video thumbnail')
-        googleString = self._search_regex("googleCode = '(.*?)'", webpage, 'file url')
-        googleString = base64.b64decode(googleString).decode('ascii')
-        final_url = self._search_regex('","(.*?)"', googleString,'final video url')
-        ext = final_url.rpartition(u'.')[2]
-
-        return [{
-            'id':        video_id,
-            'url':       final_url,
-            'ext':       ext,
-            'title':     title,
-            'thumbnail': thumbnail_url,
-        }]
 
+        youtube_id = self._search_regex(
+            r"videoId\s*:\s*[\"']([0-9A-Za-z_-]{11})[\"']",
+            webpage, 'video URL', default=None)
+        if youtube_id:
+            return {
+                '_type': 'url',
+                'url': youtube_id,
+                'ie_key': YoutubeIE.ie_key(),
+            }
+
+        video_url = self._search_regex(
+            r'<video[^>]+>\s*<source[^>]+src=(["\'])(?P<url>.+?)\1',
+            webpage, 'video URL', group='url')
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'title': self._og_search_title(webpage),
+            'thumbnail': self._og_search_thumbnail(webpage),
+            'description': self._og_search_description(webpage),
+        }