X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;ds=sidebyside;f=youtube_dl%2Fextractor%2Fwistia.py;h=8b14840a2dba606951f1f7d80694f1e7f0cca8d6;hb=b24ab3e341b9082774785332a1aa6405764f7202;hp=e6bfa9e147a2b62e6dbcdd52343675bb9ca52a65;hpb=00558d94145f97c644e66ec086fa9b9d8c58280f;p=youtube-dl

diff --git a/youtube_dl/extractor/wistia.py b/youtube_dl/extractor/wistia.py
index e6bfa9e14..8b14840a2 100644
--- a/youtube_dl/extractor/wistia.py
+++ b/youtube_dl/extractor/wistia.py
@@ -1,13 +1,16 @@
 from __future__ import unicode_literals
 
-import json
-import re
-
 from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    sanitized_Request,
+    int_or_none,
+)
 
 
 class WistiaIE(InfoExtractor):
     _VALID_URL = r'https?://(?:fast\.)?wistia\.net/embed/iframe/(?P<id>[a-z0-9]+)'
+    _API_URL = 'http://fast.wistia.com/embed/medias/{0:}.json'
 
     _TEST = {
         'url': 'http://fast.wistia.net/embed/iframe/sh7fpupwlt',
@@ -16,47 +19,60 @@ class WistiaIE(InfoExtractor):
             'id': 'sh7fpupwlt',
             'ext': 'mov',
             'title': 'Being Resourceful',
+            'description': 'a Clients From Hell Video Series video from worldwidewebhosting',
+            'upload_date': '20131204',
+            'timestamp': 1386185018,
             'duration': 117,
         },
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-
-        webpage = self._download_webpage(url, video_id)
-        data_json = self._html_search_regex(
-            r'Wistia\.iframeInit\((.*?), {}\);', webpage, 'video data')
+        video_id = self._match_id(url)
 
-        data = json.loads(data_json)
+        request = sanitized_Request(self._API_URL.format(video_id))
+        request.add_header('Referer', url)  # Some videos require this.
+        data_json = self._download_json(request, video_id)
+        if data_json.get('error'):
+            raise ExtractorError('Error while getting the playlist',
+                                 expected=True)
+        data = data_json['media']
+        title = data['name']
 
         formats = []
         thumbnails = []
-        for atype, a in data['assets'].items():
-            if atype == 'still':
+        for a in data['assets']:
+            astatus = a.get('status')
+            atype = a.get('type')
+            if (astatus is not None and astatus != 2) or atype == 'preview':
+                continue
+            elif atype in ('still', 'still_image'):
                 thumbnails.append({
                     'url': a['url'],
                     'resolution': '%dx%d' % (a['width'], a['height']),
                 })
-                continue
-            if atype == 'preview':
-                continue
-            formats.append({
-                'format_id': atype,
-                'url': a['url'],
-                'width': a['width'],
-                'height': a['height'],
-                'filesize': a['size'],
-                'ext': a['ext'],
-                'preference': 1 if atype == 'original' else None,
-            })
+            else:
+                formats.append({
+                    'format_id': atype,
+                    'url': a['url'],
+                    'tbr': int_or_none(a.get('bitrate')),
+                    'vbr': int_or_none(a.get('opt_vbitrate')),
+                    'width': int_or_none(a.get('width')),
+                    'height': int_or_none(a.get('height')),
+                    'filesize': int_or_none(a.get('size')),
+                    'vcodec': a.get('codec'),
+                    'container': a.get('container'),
+                    'ext': a.get('ext'),
+                    'preference': 1 if atype == 'original' else None,
+                })
 
         self._sort_formats(formats)
 
         return {
             'id': video_id,
-            'title': data['name'],
+            'title': title,
+            'description': data.get('seoDescription'),
             'formats': formats,
             'thumbnails': thumbnails,
-            'duration': data.get('duration'),
+            'duration': int_or_none(data.get('duration')),
+            'timestamp': int_or_none(data.get('createdAt')),
         }