]> Raphaël G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/vidzi.py
Prepare to upload
[youtubedl] / youtube_dl / extractor / vidzi.py
index 669979e13660b55d8c6f426aaab2b888007b8c5b..ac35d55a9505808144fed105d535d53779afb1cb 100644 (file)
@@ -1,12 +1,20 @@
-#coding: utf-8
+# coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
+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)/(?:embed-)?(?P<id>[0-9a-zA-Z]+)'
+    _TESTS = [{
         'url': 'http://vidzi.tv/cghql9yq6emu.html',
         'md5': '4f16c71ca0c8c8635ab6932b5f3f1660',
         'info_dict': {
@@ -14,20 +22,40 @@ class VidziIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'youtube-dl test video  1\\\\2\'3/4<5\\\\6ä7↭',
         },
-    }
+        '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,
+    }]
 
     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')
+
+        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')
-        
-        return {
-            'id': video_id,
-            'title': title,
-            'url': video_url,
-        }
-        
\ No newline at end of file
+
+        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