[vidzi] Fix extraction

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

diff --git a/youtube_dl/extractor/vidzi.py b/youtube_dl/extractor/vidzi.py
index 82d2179e9a72d2527aacda2b781881b1e8fdcc1b..4ec07db3a9fbfa468bb79f52c0e964d769dbf152 100644 (file)
--- a/youtube_dl/extractor/vidzi.py
+++ b/youtube_dl/extractor/vidzi.py
@@ -1,33 +1,55 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
-from .common import InfoExtractor
+import re
 
+from .jwplatform import JWPlatformBaseIE
+from ..utils import (
+    base36,
+    js_to_json,
+)
 
-class VidziIE(InfoExtractor):
+
+class VidziIE(JWPlatformBaseIE):
     _VALID_URL = r'https?://(?:www\.)?vidzi\.tv/(?P<id>\w+)'
     _TEST = {
-        'url': 'http://vidzi.tv/m1chxrwq7bx9',
-        'md5': '5c4c4a8ca2281a199c8eefe8f411d630',
+        'url': 'http://vidzi.tv/cghql9yq6emu.html',
+        'md5': '4f16c71ca0c8c8635ab6932b5f3f1660',
         'info_dict': {
-            'id': 'm1chxrwq7bx9',
+            'id': 'cghql9yq6emu',
             'ext': 'mp4',
-            'title': 'Watch Cadbury Dream Factory S01E04 HDTV x264 FiHTV mp4',
+            'title': 'youtube-dl test video  1\\\\2\'3/4<5\\\\6ä7↭',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
         },
     }
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        
+
         webpage = self._download_webpage(url, video_id)
-        video_url = self._html_search_regex(
-            r'{\s*file\s*:\s*"([^"]+)"\s*}', webpage, 'video url')
         title = self._html_search_regex(
-            r'<Title>([^<]+)<\/Title>', webpage, 'title')
-        
-        return {
-            'id': video_id,
-            'title': title,
-            'url': video_url,
-            'ext': 'mp4',
-        }
-        
\ No newline at end of file
+            r'(?s)<h2 class="video-title">(.*?)</h2>', webpage, 'title')
+
+        mobj = re.search(r"}\('(.+)',36,(\d+),'([^']+)'\.split\('\|'\)", webpage)
+        code, count, symbols = mobj.groups()
+
+        count = int(count)
+        symbols = symbols.split('|')
+
+        while count:
+            count -= 1
+            if symbols[count]:
+                code = re.sub(r'\b%s\b' % base36(count), symbols[count], code)
+
+        code = code.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