Merge pull request #1 from e7appew/python3

[youtubedl] / youtube_dl / extractor / vidzi.py

diff --git a/youtube_dl/extractor/vidzi.py b/youtube_dl/extractor/vidzi.py
index 7c6e98026823e0df415664ad0b4d6995c9e2f673..d49cc6cbc567a8a0219f304a52707bd4129d1119 100644 (file)
--- a/youtube_dl/extractor/vidzi.py
+++ b/youtube_dl/extractor/vidzi.py
@@ -1,39 +1,46 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-from .common import InfoExtractor
-from ..utils import smuggle_url
+from .jwplatform import JWPlatformBaseIE
+from ..utils import (
+    decode_packed_codes,
+    js_to_json,
+)
 
 
-class VidziIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?vidzi\.tv/(?P<id>\w+)'
-    _TEST = {
+class VidziIE(JWPlatformBaseIE):
+    _VALID_URL = r'https?://(?:www\.)?vidzi\.tv/(?:embed-)?(?P<id>[0-9a-zA-Z]+)'
+    _TESTS = [{
         'url': 'http://vidzi.tv/cghql9yq6emu.html',
         'md5': '4f16c71ca0c8c8635ab6932b5f3f1660',
         'info_dict': {
             'id': 'cghql9yq6emu',
             'ext': 'mp4',
             'title': 'youtube-dl test video  1\\\\2\'3/4<5\\\\6ä7↭',
-            'uploader': 'vidzi.tv',
         },
         'params': {
             # m3u8 download
             'skip_download': True,
         },
-    }
+    }, {
+        'url': 'http://vidzi.tv/embed-4z2yb0rzphe9-600x338.html',
+        'skip_download': True,
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        webpage = self._download_webpage(url, video_id)
+        webpage = self._download_webpage(
+            'http://vidzi.tv/%s' % video_id, video_id)
         title = self._html_search_regex(
             r'(?s)<h2 class="video-title">(.*?)</h2>', webpage, 'title')
 
-        # Vidzi now uses jwplayer, which can be handled by GenericIE
-        return {
-            '_type': 'url_transparent',
-            'id': video_id,
-            'title': title,
-            'url': smuggle_url(url, {'to_generic': True}),
-            'ie_key': 'Generic',
-        }
+        code = decode_packed_codes(webpage).replace('\\\'', '\'')
+        jwplayer_data = self._parse_json(
+            self._search_regex(r'setup\(([^)]+)\)', code, 'jwplayer data'),
+            video_id, transform_source=js_to_json)
+
+        info_dict = self._parse_jwplayer_data(jwplayer_data, video_id, require_title=False)
+        info_dict['title'] = title
+
+        return info_dict