[youtube] fix extraction for embed restricted live streams(fixes #16433)

[youtube-dl] / youtube_dl / extractor / wistia.py

diff --git a/youtube_dl/extractor/wistia.py b/youtube_dl/extractor/wistia.py
index 6eb94fcabc1440e3069c9491b5971c3635eae48d..2182d6fd485bf4f1ed6ead7d78451b7c357ddc37 100644 (file)
--- a/youtube_dl/extractor/wistia.py
+++ b/youtube_dl/extractor/wistia.py
@@ -1,9 +1,13 @@
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
     int_or_none,
+    float_or_none,
+    unescapeHTML,
 )
 
 
@@ -27,8 +31,31 @@ class WistiaIE(InfoExtractor):
     }, {
         '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)
 
@@ -63,6 +90,8 @@ class WistiaIE(InfoExtractor):
                     '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,
@@ -73,7 +102,8 @@ class WistiaIE(InfoExtractor):
                     'filesize': int_or_none(a.get('size')),
                     'vcodec': a.get('codec'),
                     'container': a.get('container'),
-                    'ext': a.get('ext'),
+                    'ext': 'mp4' if is_m3u8 else aext,
+                    'protocol': 'm3u8' if is_m3u8 else None,
                     'preference': 1 if atype == 'original' else None,
                 })
 
@@ -85,6 +115,6 @@ class WistiaIE(InfoExtractor):
             'description': data.get('seoDescription'),
             'formats': formats,
             'thumbnails': thumbnails,
-            'duration': int_or_none(data.get('duration')),
+            'duration': float_or_none(data.get('duration')),
             'timestamp': int_or_none(data.get('createdAt')),
         }