]> git.bitcoin.ninja Git - youtube-dl/commitdiff
[Dumpert] Add new extractor
authorJoram Schrijver <i@joram.io>
Sun, 29 Mar 2015 21:41:06 +0000 (23:41 +0200)
committerJoram Schrijver <i@joram.io>
Sun, 29 Mar 2015 21:41:06 +0000 (23:41 +0200)
Add support for the Dutch video site Dumpert. http://www.dumpert.nl/

youtube_dl/extractor/__init__.py
youtube_dl/extractor/dumpert.py [new file with mode: 0644]

index a65c0c25b8144addfe5d643f92827b25d220a6a8..43bac0252bf68b443231c4416320759b8e126c19 100644 (file)
@@ -115,6 +115,7 @@ from .drtuber import DrTuberIE
 from .drtv import DRTVIE
 from .dvtv import DVTVIE
 from .dump import DumpIE
+from .dumpert import DumpertIE
 from .defense import DefenseGouvFrIE
 from .discovery import DiscoveryIE
 from .divxstage import DivxStageIE
diff --git a/youtube_dl/extractor/dumpert.py b/youtube_dl/extractor/dumpert.py
new file mode 100644 (file)
index 0000000..52d07de
--- /dev/null
@@ -0,0 +1,47 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import base64
+
+from .common import InfoExtractor
+
+
+class DumpertIE(InfoExtractor):
+    _VALID_URL = (r'https?://(?:www\.)?dumpert\.nl/mediabase/'
+                  r'(?P<id>[0-9]+/[0-9a-zA-Z]+)/?.*')
+    _TEST = {
+        'url': 'http://www.dumpert.nl/mediabase/6646981/951bc60f/',
+        'md5': '1b9318d7d5054e7dcb9dc7654f21d643',
+        'info_dict': {
+            'id': '6646981/951bc60f',
+            'ext': 'mp4',
+            'title': 'Ik heb nieuws voor je',
+            'description': 'Niet schrikken hoor'
+        }
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        title = self._html_search_meta('title', webpage)
+        description = self._html_search_meta('description', webpage)
+
+        files_base64 = self._html_search_regex(r'data-files="(.*?)"',
+                                               webpage,
+                                               'files')
+        files_json = base64.b64decode(files_base64).decode('iso-8859-1')
+        files = self._parse_json(files_json, video_id)
+
+        format_names = ['flv', 'mobile', 'tablet', '720p']
+        formats = [{'format_id': name,
+                    'url': files[name].replace(r'\/', '/')}
+                   for name in format_names
+                   if name in files]
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'formats': formats
+        }