Fix extraction from youtube.

[youtubedl] / youtube_dl / extractor / wimp.py

diff --git a/youtube_dl/extractor/wimp.py b/youtube_dl/extractor/wimp.py
index 82a626e0eb866e4924f32f7809e33d59d9261168..ea234e3c5088f7c7c99e049ab95f1e69542dd38f 100644 (file)
--- a/youtube_dl/extractor/wimp.py
+++ b/youtube_dl/extractor/wimp.py
@@ -1,33 +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/([^/]+)/'
-    _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'https?://(?:www\.)?wimp\.com/(?P<id>[^/]+)'
+    _TESTS = [{
+        'url': 'http://www.wimp.com/maru-is-exhausted/',
+        'md5': 'ee21217ffd66d058e8b16be340b74883',
+        'info_dict': {
+            'id': 'maru-is-exhausted',
+            'ext': 'mp4',
+            'title': 'Maru is exhausted.',
+            'description': 'md5:57e099e857c0a4ea312542b684a869b8',
         }
-    }
+    }, {
+        'url': 'http://www.wimp.com/clowncar/',
+        'md5': '5c31ad862a90dc5b1f023956faec13fe',
+        'info_dict': {
+            'id': 'cG4CEr2aiSg',
+            'ext': 'webm',
+            'title': 'Basset hound clown car...incredible!',
+            'description': '5 of my Bassets crawled in this dog loo! www.bellinghambassets.com\n\nFor licensing/usage please contact: licensing(at)jukinmediadotcom',
+            'upload_date': '20140303',
+            'uploader': 'Gretchen Hoey',
+            'uploader_id': 'gretchenandjeff1',
+        },
+        'add_ie': ['Youtube'],
+    }]
 
     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')
 
-        return {
+        youtube_id = self._search_regex(
+            (r"videoId\s*:\s*[\"']([0-9A-Za-z_-]{11})[\"']",
+             r'data-id=["\']([0-9A-Za-z_-]{11})'),
+            webpage, 'video URL', default=None)
+        if youtube_id:
+            return self.url_result(youtube_id, YoutubeIE.ie_key())
+
+        info_dict = self._extract_jwplayer_data(
+            webpage, video_id, require_title=False)
+
+        info_dict.update({
             'id': video_id,
-            'url': final_url,
             'title': self._og_search_title(webpage),
-            'thumbnail': self._og_search_thumbnail(webpage),
             'description': self._og_search_description(webpage),
-        }
+        })
+
+        return info_dict