Merge pull request #8092 from bpfoley/twitter-thumbnail

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

diff --git a/youtube_dl/extractor/wimp.py b/youtube_dl/extractor/wimp.py
index 82a626e0eb866e4924f32f7809e33d59d9261168..fb0accac744532625c04bb964c1fa031723ed8ff 100644 (file)
--- a/youtube_dl/extractor/wimp.py
+++ b/youtube_dl/extractor/wimp.py
@@ -1,32 +1,53 @@
-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/([^/]+)/'
-    _TEST = {
-        u'url': u'http://www.wimp.com/deerfence/',
-        u'file': u'deerfence.flv',
-        u'md5': u'8b215e2e0168c6081a1cf84b2846a2b5',
-        u'info_dict': {
-            u"title": u"Watch Till End: Herd of deer jump over a fence.",
-            u"description": u"These deer look as fluid as running water when they jump over this fence as a herd. This video is one that needs to be watched until the very end for the true majesty to be witnessed, but once it comes, it's sure to take your breath away.",
+    _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': 'webm',
+            '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)
-        googleString = self._search_regex("googleCode = '(.*?)'", webpage, 'file url')
-        googleString = base64.b64decode(googleString).decode('ascii')
-        final_url = self._search_regex('","(.*?)"', googleString, u'final video 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': final_url,
+            'url': video_url,
             'title': self._og_search_title(webpage),
             'thumbnail': self._og_search_thumbnail(webpage),
             'description': self._og_search_description(webpage),