X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/9ed7fe4fe4c445eb7d9f3197bb300d0db8f1807a..81657e6063f6dd0b3ee6243d446b8df1b13adfbc:/youtube_dl/extractor/wistia.py

diff --git a/youtube_dl/extractor/wistia.py b/youtube_dl/extractor/wistia.py
index fdb16d9..2182d6f 100644
--- a/youtube_dl/extractor/wistia.py
+++ b/youtube_dl/extractor/wistia.py
@@ -1,65 +1,120 @@
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
-    sanitized_Request,
+    int_or_none,
+    float_or_none,
+    unescapeHTML,
 )
 
 
 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'
+    _VALID_URL = r'(?:wistia:|https?://(?:fast\.)?wistia\.net/embed/iframe/)(?P<id>[a-z0-9]+)'
+    _API_URL = 'http://fast.wistia.com/embed/medias/%s.json'
+    _IFRAME_URL = 'http://fast.wistia.net/embed/iframe/%s'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://fast.wistia.net/embed/iframe/sh7fpupwlt',
         'md5': 'cafeb56ec0c53c18c97405eecb3133df',
         'info_dict': {
             'id': 'sh7fpupwlt',
             'ext': 'mov',
             'title': 'Being Resourceful',
+            'description': 'a Clients From Hell Video Series video from worldwidewebhosting',
+            'upload_date': '20131204',
+            'timestamp': 1386185018,
             'duration': 117,
         },
-    }
+    }, {
+        'url': 'wistia:sh7fpupwlt',
+        'only_matching': True,
+    }, {
+        # with hls video
+        'url': 'wistia:807fafadvk',
+        'only_matching': True,
+    }]
+
+    @staticmethod
+    def _extract_url(webpage):
+        match = re.search(
+            r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
+        if match:
+            return unescapeHTML(match.group('url'))
+
+        match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
+        if match:
+            return 'wistia:%s' % match.group('id')
+
+        match = re.search(
+            r'''(?sx)
+                <script[^>]+src=(["'])(?:https?:)?//fast\.wistia\.com/assets/external/E-v1\.js\1[^>]*>.*?
+                <div[^>]+class=(["']).*?\bwistia_async_(?P<id>[a-z0-9]+)\b.*?\2
+            ''', webpage)
+        if match:
+            return 'wistia:%s' % match.group('id')
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        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)
+        data_json = self._download_json(
+            self._API_URL % video_id, video_id,
+            # Some videos require this.
+            headers={
+                'Referer': url if url.startswith('http') else self._IFRAME_URL % video_id,
+            })
+
         if data_json.get('error'):
-            raise ExtractorError('Error while getting the playlist',
-                                 expected=True)
+            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':
-                thumbnails.append({
-                    'url': a['url'],
-                    'resolution': '%dx%d' % (a['width'], a['height']),
-                })
+        for a in data['assets']:
+            aurl = a.get('url')
+            if not aurl:
                 continue
-            if atype == 'preview':
+            astatus = a.get('status')
+            atype = a.get('type')
+            if (astatus is not None and astatus != 2) or atype in ('preview', 'storyboard'):
                 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,
-            })
+            elif atype in ('still', 'still_image'):
+                thumbnails.append({
+                    'url': aurl,
+                    'width': int_or_none(a.get('width')),
+                    'height': int_or_none(a.get('height')),
+                })
+            else:
+                aext = a.get('ext')
+                is_m3u8 = a.get('container') == 'm3u8' or aext == 'm3u8'
+                formats.append({
+                    'format_id': atype,
+                    'url': aurl,
+                    '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': 'mp4' if is_m3u8 else aext,
+                    'protocol': 'm3u8' if is_m3u8 else None,
+                    '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': float_or_none(data.get('duration')),
+            'timestamp': int_or_none(data.get('createdAt')),
         }