Fix extraction from youtube.

[youtubedl] / youtube_dl / extractor / dumpert.py

diff --git a/youtube_dl/extractor/dumpert.py b/youtube_dl/extractor/dumpert.py
index 1f00386feae15d00a4421b3166335c15f3b01aa9..be2e3d37841b48ce7cf187b01ddd5236f85d7683 100644 (file)
--- a/youtube_dl/extractor/dumpert.py
+++ b/youtube_dl/extractor/dumpert.py
@@ -1,15 +1,18 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import base64
+import re
 
 from .common import InfoExtractor
-from ..compat import compat_urllib_request
-from ..utils import qualities
+from ..compat import compat_b64decode
+from ..utils import (
+    qualities,
+    sanitized_Request,
+)
 
 
 class DumpertIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?dumpert\.nl/(?:mediabase|embed)/(?P<id>[0-9]+/[0-9a-zA-Z]+)'
+    _VALID_URL = r'(?P<protocol>https?)://(?:www\.)?dumpert\.nl/(?:mediabase|embed)/(?P<id>[0-9]+/[0-9a-zA-Z]+)'
     _TESTS = [{
         'url': 'http://www.dumpert.nl/mediabase/6646981/951bc60f/',
         'md5': '1b9318d7d5054e7dcb9dc7654f21d643',
@@ -18,7 +21,7 @@ class DumpertIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Ik heb nieuws voor je',
             'description': 'Niet schrikken hoor',
-            'thumbnail': 're:^https?://.*\.jpg$',
+            'thumbnail': r're:^https?://.*\.jpg$',
         }
     }, {
         'url': 'http://www.dumpert.nl/embed/6675421/dc440fe7/',
@@ -26,10 +29,12 @@ class DumpertIE(InfoExtractor):
     }]
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        protocol = mobj.group('protocol')
 
-        url = 'https://www.dumpert.nl/mediabase/' + video_id
-        req = compat_urllib_request.Request(url)
+        url = '%s://www.dumpert.nl/mediabase/%s' % (protocol, video_id)
+        req = sanitized_Request(url)
         req.add_header('Cookie', 'nsfw=1; cpc=10')
         webpage = self._download_webpage(req, video_id)
 
@@ -37,7 +42,7 @@ class DumpertIE(InfoExtractor):
             r'data-files="([^"]+)"', webpage, 'data files')
 
         files = self._parse_json(
-            base64.b64decode(files_base64.encode('utf-8')).decode('utf-8'),
+            compat_b64decode(files_base64).decode('utf-8'),
             video_id)
 
         quality = qualities(['flv', 'mobile', 'tablet', '720p'])