[extractor/generic] Add support for async wistia embeds (Closes #9549)

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

diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py
index 95d23325900e8ed0d61bcec5d09ff3ba3d5e7a82..632d7b5f01a612d1fb5e33ddbfc845d09ca6e034 100644 (file)
--- a/youtube_dl/extractor/generic.py
+++ b/youtube_dl/extractor/generic.py
@@ -51,7 +51,7 @@ from .tnaflix import TNAFlixNetworkEmbedIE
 from .vimeo import VimeoIE
 from .dailymotion import DailymotionCloudIE
 from .onionstudios import OnionStudiosIE
-from .snagfilms import SnagFilmsEmbedIE
+from .viewlift import ViewLiftEmbedIE
 from .screenwavemedia import ScreenwaveMediaIE
 from .mtv import MTVServicesEmbeddedIE
 from .pladform import PladformIE
@@ -61,6 +61,7 @@ from .jwplatform import JWPlatformIE
 from .digiteka import DigitekaIE
 from .instagram import InstagramIE
 from .liveleak import LiveLeakIE
+from .threeqsdn import ThreeQSDNIE
 
 
 class GenericIE(InfoExtractor):
@@ -237,6 +238,7 @@ class GenericIE(InfoExtractor):
                 'ext': 'mp4',
                 'title': 'car-20120827-manifest',
                 'formats': 'mincount:9',
+                'upload_date': '20130904',
             },
             'params': {
                 'format': 'bestvideo',
@@ -596,7 +598,11 @@ class GenericIE(InfoExtractor):
                 'id': 'k2mm4bCdJ6CQ2i7c8o2',
                 'ext': 'mp4',
                 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
+                'description': 'md5:faf028e48a461b8b7fad38f1e104b119',
                 'uploader': 'Spi0n',
+                'uploader_id': 'xgditw',
+                'upload_date': '20140425',
+                'timestamp': 1398441542,
             },
             'add_ie': ['Dailymotion'],
         },
@@ -729,8 +735,11 @@ class GenericIE(InfoExtractor):
                 'id': 'uxjb0lwrcz',
                 'ext': 'mp4',
                 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks',
+                'description': 'a Martin Fowler video from ThoughtWorks',
                 'duration': 1715.0,
                 'uploader': 'thoughtworks.wistia.com',
+                'upload_date': '20140603',
+                'timestamp': 1401832161,
             },
         },
         # Soundcloud embed
@@ -879,6 +888,7 @@ class GenericIE(InfoExtractor):
         # Eagle.Platform embed (generic URL)
         {
             'url': 'http://lenta.ru/news/2015/03/06/navalny/',
+            # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used
             'info_dict': {
                 'id': '227304',
                 'ext': 'mp4',
@@ -893,6 +903,7 @@ class GenericIE(InfoExtractor):
         # ClipYou (Eagle.Platform) embed (custom URL)
         {
             'url': 'http://muz-tv.ru/play/7129/',
+            # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used
             'info_dict': {
                 'id': '12820',
                 'ext': 'mp4',
@@ -981,6 +992,9 @@ class GenericIE(InfoExtractor):
                 'ext': 'flv',
                 'title': "PFT Live: New leader in the 'new-look' defense",
                 'description': 'md5:65a19b4bbfb3b0c0c5768bed1dfad74e',
+                'uploader': 'NBCU-SPORTS',
+                'upload_date': '20140107',
+                'timestamp': 1389118457,
             },
         },
         # UDN embed
@@ -1033,6 +1047,9 @@ class GenericIE(InfoExtractor):
                 'title': 'SN Presents: Russell Martin, World Citizen',
                 'description': 'To understand why he was the Toronto Blue Jays’ top off-season priority is to appreciate his background and upbringing in Montreal, where he first developed his baseball skills. Written and narrated by Stephen Brunt.',
                 'uploader': 'Rogers Sportsnet',
+                'uploader_id': '1704050871',
+                'upload_date': '20150525',
+                'timestamp': 1432570283,
             },
         },
         # Dailymotion Cloud video
@@ -1124,6 +1141,9 @@ class GenericIE(InfoExtractor):
                 'title': 'The Cardinal Pell Interview',
                 'description': 'Sky News Contributor Andrew Bolt interviews George Pell in Rome, following the Cardinal\'s evidence before the Royal Commission into Child Abuse. ',
                 'uploader': 'GlobeCast Australia - GlobeStream',
+                'uploader_id': '2733773828001',
+                'upload_date': '20160304',
+                'timestamp': 1457083087,
             },
             'params': {
                 # m3u8 downloads
@@ -1408,7 +1428,8 @@ class GenericIE(InfoExtractor):
         #   Site Name | Video Title
         #   Video Title - Tagline | Site Name
         # and so on and so forth; it's just not practical
-        video_title = self._html_search_regex(
+        video_title = self._og_search_title(
+            webpage, default=None) or self._html_search_regex(
             r'(?s)<title>(.*?)</title>', webpage, 'video title',
             default='video')
 
@@ -1426,6 +1447,9 @@ class GenericIE(InfoExtractor):
         video_uploader = self._search_regex(
             r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
 
+        video_description = self._og_search_description(webpage, default=None)
+        video_thumbnail = self._og_search_thumbnail(webpage, default=None)
+
         # Helper method
         def _playlist_from_matches(matches, getter=None, ie=None):
             urlrs = orderedSet(
@@ -1539,6 +1563,15 @@ class GenericIE(InfoExtractor):
                 'id': 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 self.url_result(self._proto_relative_url(
+                'wistia:%s' % match.group('id')), 'Wistia')
+
         # Look for SVT player
         svt_url = SVTIE._extract_url(webpage)
         if svt_url:
@@ -1905,10 +1938,10 @@ class GenericIE(InfoExtractor):
         if onionstudios_url:
             return self.url_result(onionstudios_url)
 
-        # Look for SnagFilms embeds
-        snagfilms_url = SnagFilmsEmbedIE._extract_url(webpage)
-        if snagfilms_url:
-            return self.url_result(snagfilms_url)
+        # Look for ViewLift embeds
+        viewlift_url = ViewLiftEmbedIE._extract_url(webpage)
+        if viewlift_url:
+            return self.url_result(viewlift_url)
 
         # Look for JWPlatform embeds
         jwplatform_url = JWPlatformIE._extract_url(webpage)
@@ -1964,6 +1997,19 @@ class GenericIE(InfoExtractor):
         if liveleak_url:
             return self.url_result(liveleak_url, 'LiveLeak')
 
+        # Look for 3Q SDN embeds
+        threeqsdn_url = ThreeQSDNIE._extract_url(webpage)
+        if threeqsdn_url:
+            return {
+                '_type': 'url_transparent',
+                'ie_key': ThreeQSDNIE.ie_key(),
+                'url': self._proto_relative_url(threeqsdn_url),
+                'title': video_title,
+                'description': video_description,
+                'thumbnail': video_thumbnail,
+                'uploader': video_uploader,
+            }
+
         def check_video(vurl):
             if YoutubeIE.suitable(vurl):
                 return True
@@ -2045,6 +2091,7 @@ class GenericIE(InfoExtractor):
 
         entries = []
         for video_url in found:
+            video_url = unescapeHTML(video_url)
             video_url = video_url.replace('\\/', '/')
             video_url = compat_urlparse.urljoin(url, video_url)
             video_id = compat_urllib_parse_unquote(os.path.basename(video_url))