]> Raphaël G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/vidzi.py
New upstream version 2018.04.25
[youtubedl] / youtube_dl / extractor / vidzi.py
index 7c6e98026823e0df415664ad0b4d6995c9e2f673..9026e778cfb108462c5fdfa5ca2a42d166dc37a7 100644 (file)
@@ -1,39 +1,64 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
-from ..utils import smuggle_url
+from ..utils import (
+    decode_packed_codes,
+    js_to_json,
+    NO_DEFAULT,
+    PACKED_CODES_RE,
+)
 
 
 class VidziIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?vidzi\.tv/(?P<id>\w+)'
-    _TEST = {
+    _VALID_URL = r'https?://(?:www\.)?vidzi\.(?:tv|cc|si)/(?: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',
+        'only_matching': True,
+    }, {
+        'url': 'http://vidzi.cc/cghql9yq6emu.html',
+        'only_matching': True,
+    }, {
+        'url': 'https://vidzi.si/rph9gztxj1et.html',
+        'only_matching': 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',
-        }
+        codes = [webpage]
+        codes.extend([
+            decode_packed_codes(mobj.group(0)).replace('\\\'', '\'')
+            for mobj in re.finditer(PACKED_CODES_RE, webpage)])
+        for num, code in enumerate(codes, 1):
+            jwplayer_data = self._parse_json(
+                self._search_regex(
+                    r'setup\(([^)]+)\)', code, 'jwplayer data',
+                    default=NO_DEFAULT if num == len(codes) else '{}'),
+                video_id, transform_source=js_to_json)
+            if jwplayer_data:
+                break
+
+        info_dict = self._parse_jwplayer_data(jwplayer_data, video_id, require_title=False)
+        info_dict['title'] = title
+
+        return info_dict