Merge tag 'upstream/2017.02.07'

[youtubedl] / youtube_dl / extractor / vidzi.py

diff --git a/youtube_dl/extractor/vidzi.py b/youtube_dl/extractor/vidzi.py
index 08a5a7b8ddbace34da0a4c6bb751562eb51a95ba..9950c62ad636ee4f03389bef627da4318f019c22 100644 (file)
--- a/youtube_dl/extractor/vidzi.py
+++ b/youtube_dl/extractor/vidzi.py
@@ -1,12 +1,20 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
 # coding: utf-8
 from __future__ import unicode_literals
 

-from .common import InfoExtractor
+import re

 
 

+from .jwplatform import JWPlatformBaseIE
+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 = {
+
+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': {
         'url': 'http://vidzi.tv/cghql9yq6emu.html',
         'md5': '4f16c71ca0c8c8635ab6932b5f3f1660',
         'info_dict': {


@@ -14,19 +22,36 @@ class VidziIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'youtube-dl test video  1\\\\2\'3/4<5\\\\6ä7↭',
         },
             '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',
+        'skip_download': True,
+    }]

 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
 
     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')
 
         title = self._html_search_regex(
             r'(?s)<h2 class="video-title">(.*?)</h2>', webpage, 'title')
 

-        return {
-            'id': video_id,
-            'title': title,
-            'url': video_url,
-        }
+        packed_codes = [mobj.group(0) for mobj in re.finditer(
+            PACKED_CODES_RE, webpage)]
+        for num, pc in enumerate(packed_codes, 1):
+            code = decode_packed_codes(pc).replace('\\\'', '\'')
+            jwplayer_data = self._parse_json(
+                self._search_regex(
+                    r'setup\(([^)]+)\)', code, 'jwplayer data',
+                    default=NO_DEFAULT if num == len(packed_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