projects / youtube-dl / blobdiff
? search:
summary | shortlog | log | commit | commitdiff | tree
raw | inline | side by side
Merge branch 'brightcove_in_page_embed' of https://github.com/remitamine/youtube...
[youtube-dl] / youtube_dl / extractor / generic.py
diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py
index 42e4e70353f041323631aacf0a4389f18a92bdad..34d930a2db82fe340d55efe681bd43ff00b32aa1 100644 (file)
--- a/youtube_dl/extractor/generic.py
+++ b/youtube_dl/extractor/generic.py
@@ -4,11 +4,12 @@ from __future__ import unicode_literals
 
 import os
 import re
 
 import os
 import re
+import sys
 
 from .common import InfoExtractor
 from .youtube import YoutubeIE
 from ..compat import (
 
 from .common import InfoExtractor
 from .youtube import YoutubeIE
 from ..compat import (
-    compat_urllib_parse,
+    compat_etree_fromstring,
     compat_urllib_parse_unquote,
     compat_urllib_request,
     compat_urlparse,
     compat_urllib_parse_unquote,
     compat_urllib_request,
     compat_urlparse,
@@ -21,7 +22,6 @@ from ..utils import (
     HEADRequest,
     is_html,
     orderedSet,
     HEADRequest,
     is_html,
     orderedSet,
-    parse_xml,
     smuggle_url,
     unescapeHTML,
     unified_strdate,
     smuggle_url,
     unescapeHTML,
     unified_strdate,
@@ -30,13 +30,17 @@ from ..utils import (
     url_basename,
     xpath_text,
 )
     url_basename,
     xpath_text,
 )
-from .brightcove import BrightcoveIE
+from .brightcove import (
+    BrightcoveIE,
+    BrightcoveInPageEmbedIE,
+)
 from .nbc import NBCSportsVPlayerIE
 from .ooyala import OoyalaIE
 from .rutv import RUTVIE
 from .tvc import TVCIE
 from .sportbox import SportBoxEmbedIE
 from .smotri import SmotriIE
 from .nbc import NBCSportsVPlayerIE
 from .ooyala import OoyalaIE
 from .rutv import RUTVIE
 from .tvc import TVCIE
 from .sportbox import SportBoxEmbedIE
 from .smotri import SmotriIE
+from .myvi import MyviIE
 from .condenast import CondeNastIE
 from .udn import UDNEmbedIE
 from .senateisvp import SenateISVPIE
 from .condenast import CondeNastIE
 from .udn import UDNEmbedIE
 from .senateisvp import SenateISVPIE
@@ -47,6 +51,9 @@ from .xhamster import XHamsterEmbedIE
 from .vimeo import VimeoIE
 from .dailymotion import DailymotionCloudIE
 from .onionstudios import OnionStudiosIE
 from .vimeo import VimeoIE
 from .dailymotion import DailymotionCloudIE
 from .onionstudios import OnionStudiosIE
+from .snagfilms import SnagFilmsEmbedIE
+from .screenwavemedia import ScreenwaveMediaIE
+from .mtv import MTVServicesEmbeddedIE
 
 
 class GenericIE(InfoExtractor):
 
 
 class GenericIE(InfoExtractor):
@@ -129,6 +136,90 @@ class GenericIE(InfoExtractor):
                 'title': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
             }
         },
                 'title': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
             }
         },
+        # SMIL from http://videolectures.net/promogram_igor_mekjavic_eng
+        {
+            'url': 'http://videolectures.net/promogram_igor_mekjavic_eng/video/1/smil.xml',
+            'info_dict': {
+                'id': 'smil',
+                'ext': 'mp4',
+                'title': 'Automatics, robotics and biocybernetics',
+                'description': 'md5:815fc1deb6b3a2bff99de2d5325be482',
+                'upload_date': '20130627',
+                'formats': 'mincount:16',
+                'subtitles': 'mincount:1',
+            },
+            'params': {
+                'force_generic_extractor': True,
+                'skip_download': True,
+            },
+        },
+        # SMIL from http://www1.wdr.de/mediathek/video/livestream/index.html
+        {
+            'url': 'http://metafilegenerator.de/WDR/WDR_FS/hds/hds.smil',
+            'info_dict': {
+                'id': 'hds',
+                'ext': 'flv',
+                'title': 'hds',
+                'formats': 'mincount:1',
+            },
+            'params': {
+                'skip_download': True,
+            },
+        },
+        # SMIL from https://www.restudy.dk/video/play/id/1637
+        {
+            'url': 'https://www.restudy.dk/awsmedia/SmilDirectory/video_1637.xml',
+            'info_dict': {
+                'id': 'video_1637',
+                'ext': 'flv',
+                'title': 'video_1637',
+                'formats': 'mincount:3',
+            },
+            'params': {
+                'skip_download': True,
+            },
+        },
+        # SMIL from http://adventure.howstuffworks.com/5266-cool-jobs-iditarod-musher-video.htm
+        {
+            'url': 'http://services.media.howstuffworks.com/videos/450221/smil-service.smil',
+            'info_dict': {
+                'id': 'smil-service',
+                'ext': 'flv',
+                'title': 'smil-service',
+                'formats': 'mincount:1',
+            },
+            'params': {
+                'skip_download': True,
+            },
+        },
+        # SMIL from http://new.livestream.com/CoheedandCambria/WebsterHall/videos/4719370
+        {
+            'url': 'http://api.new.livestream.com/accounts/1570303/events/1585861/videos/4719370.smil',
+            'info_dict': {
+                'id': '4719370',
+                'ext': 'mp4',
+                'title': '571de1fd-47bc-48db-abf9-238872a58d1f',
+                'formats': 'mincount:3',
+            },
+            'params': {
+                'skip_download': True,
+            },
+        },
+        # XSPF playlist from http://www.telegraaf.nl/tv/nieuws/binnenland/24353229/__Tikibad_ontruimd_wegens_brand__.html
+        {
+            'url': 'http://www.telegraaf.nl/xml/playlist/2015/8/7/mZlp2ctYIUEB.xspf',
+            'info_dict': {
+                'id': 'mZlp2ctYIUEB',
+                'ext': 'mp4',
+                'title': 'Tikibad ontruimd wegens brand',
+                'description': 'md5:05ca046ff47b931f9b04855015e163a4',
+                'thumbnail': 're:^https?://.*\.jpg$',
+                'duration': 33,
+            },
+            'params': {
+                'skip_download': True,
+            },
+        },
         # google redirect
         {
             'url': 'http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&ved=0CCUQtwIwAA&url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DcmQHVoWB5FY&ei=F-sNU-LLCaXk4QT52ICQBQ&usg=AFQjCNEw4hL29zgOohLXvpJ-Bdh2bils1Q&bvm=bv.61965928,d.bGE',
         # google redirect
         {
             'url': 'http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&ved=0CCUQtwIwAA&url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DcmQHVoWB5FY&ei=F-sNU-LLCaXk4QT52ICQBQ&usg=AFQjCNEw4hL29zgOohLXvpJ-Bdh2bils1Q&bvm=bv.61965928,d.bGE',
@@ -145,6 +236,22 @@ class GenericIE(InfoExtractor):
                 'skip_download': False,
             }
         },
                 'skip_download': False,
             }
         },
+        {
+            # redirect in Refresh HTTP header
+            'url': 'https://www.facebook.com/l.php?u=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DpO8h3EaFRdo&h=TAQHsoToz&enc=AZN16h-b6o4Zq9pZkCCdOLNKMN96BbGMNtcFwHSaazus4JHT_MFYkAA-WARTX2kvsCIdlAIyHZjl6d33ILIJU7Jzwk_K3mcenAXoAzBNoZDI_Q7EXGDJnIhrGkLXo_LJ_pAa2Jzbx17UHMd3jAs--6j2zaeto5w9RTn8T_1kKg3fdC5WPX9Dbb18vzH7YFX0eSJmoa6SP114rvlkw6pkS1-T&s=1',
+            'info_dict': {
+                'id': 'pO8h3EaFRdo',
+                'ext': 'mp4',
+                'title': 'Tripeo Boiler Room x Dekmantel Festival DJ Set',
+                'description': 'md5:6294cc1af09c4049e0652b51a2df10d5',
+                'upload_date': '20150917',
+                'uploader_id': 'brtvofficial',
+                'uploader': 'Boiler Room',
+            },
+            'params': {
+                'skip_download': False,
+            },
+        },
         {
             'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
             'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
         {
             'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
             'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
@@ -235,6 +342,19 @@ class GenericIE(InfoExtractor):
             },
             'add_ie': ['Ooyala'],
         },
             },
             'add_ie': ['Ooyala'],
         },
+        {
+            # ooyala video embedded with http://player.ooyala.com/iframe.js
+            'url': 'http://www.macrumors.com/2015/07/24/steve-jobs-the-man-in-the-machine-first-trailer/',
+            'info_dict': {
+                'id': 'p0MGJndjoG5SOKqO_hZJuZFPB-Tr5VgB',
+                'ext': 'mp4',
+                'title': '"Steve Jobs: Man in the Machine" trailer',
+                'description': 'The first trailer for the Alex Gibney documentary "Steve Jobs: Man in the Machine."',
+            },
+            'params': {
+                'skip_download': True,
+            },
+        },
         # multiple ooyala embeds on SBN network websites
         {
             'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
         # multiple ooyala embeds on SBN network websites
         {
             'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
@@ -275,14 +395,6 @@ class GenericIE(InfoExtractor):
                 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
             },
         },
                 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
             },
         },
-        # BBC iPlayer embeds
-        {
-            'url': 'http://www.bbc.co.uk/blogs/adamcurtis/posts/BUGGER',
-            'info_dict': {
-                'title': 'BBC - Blogs -  Adam Curtis - BUGGER',
-            },
-            'playlist_mincount': 18,
-        },
         # RUTV embed
         {
             'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
         # RUTV embed
         {
             'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
@@ -337,6 +449,17 @@ class GenericIE(InfoExtractor):
                 'skip_download': True,
             },
         },
                 'skip_download': True,
             },
         },
+        # Myvi.ru embed
+        {
+            'url': 'http://www.kinomyvi.tv/news/detail/Pervij-dublirovannij-trejler--Uzhastikov-_nOw1',
+            'info_dict': {
+                'id': 'f4dafcad-ff21-423d-89b5-146cfd89fa1e',
+                'ext': 'mp4',
+                'title': 'Ужастики, русский трейлер (2015)',
+                'thumbnail': 're:^https?://.*\.jpg$',
+                'duration': 153,
+            }
+        },
         # XHamster embed
         {
             'url': 'http://www.numisc.com/forum/showthread.php?11696-FM15-which-pumiscer-was-this-%28-vid-%29-%28-alfa-as-fuck-srx-%29&s=711f5db534502e22260dec8c5e2d66d8',
         # XHamster embed
         {
             'url': 'http://www.numisc.com/forum/showthread.php?11696-FM15-which-pumiscer-was-this-%28-vid-%29-%28-alfa-as-fuck-srx-%29&s=711f5db534502e22260dec8c5e2d66d8',
@@ -395,6 +518,26 @@ class GenericIE(InfoExtractor):
                 'skip_download': 'Requires rtmpdump'
             }
         },
                 'skip_download': 'Requires rtmpdump'
             }
         },
+        # francetv embed
+        {
+            'url': 'http://www.tsprod.com/replay-du-concert-alcaline-de-calogero',
+            'info_dict': {
+                'id': 'EV_30231',
+                'ext': 'mp4',
+                'title': 'Alcaline, le concert avec Calogero',
+                'description': 'md5:61f08036dcc8f47e9cfc33aed08ffaff',
+                'upload_date': '20150226',
+                'timestamp': 1424989860,
+                'duration': 5400,
+            },
+            'params': {
+                # m3u8 downloads
+                'skip_download': True,
+            },
+            'expected_warnings': [
+                'Forbidden'
+            ]
+        },
         # Condé Nast embed
         {
             'url': 'http://www.wired.com/2014/04/honda-asimo/',
         # Condé Nast embed
         {
             'url': 'http://www.wired.com/2014/04/honda-asimo/',
@@ -668,6 +811,18 @@ class GenericIE(InfoExtractor):
                 'title': 'John Carlson Postgame 2/25/15',
             },
         },
                 'title': 'John Carlson Postgame 2/25/15',
             },
         },
+        # Kaltura embed (different embed code)
+        {
+            'url': 'http://www.premierchristianradio.com/Shows/Saturday/Unbelievable/Conference-Videos/Os-Guinness-Is-It-Fools-Talk-Unbelievable-Conference-2014',
+            'info_dict': {
+                'id': '1_a52wc67y',
+                'ext': 'flv',
+                'upload_date': '20150127',
+                'uploader_id': 'PremierMedia',
+                'timestamp': int,
+                'title': 'Os Guinness // Is It Fools Talk? // Unbelievable? Conference 2014',
+            },
+        },
         # Eagle.Platform embed (generic URL)
         {
             'url': 'http://lenta.ru/news/2015/03/06/navalny/',
         # Eagle.Platform embed (generic URL)
         {
             'url': 'http://lenta.ru/news/2015/03/06/navalny/',
@@ -849,6 +1004,15 @@ class GenericIE(InfoExtractor):
                 'uploader_id': 'clickhole',
             }
         },
                 'uploader_id': 'clickhole',
             }
         },
+        # SnagFilms embed
+        {
+            'url': 'http://whilewewatch.blogspot.ru/2012/06/whilewewatch-whilewewatch-gripping.html',
+            'info_dict': {
+                'id': '74849a00-85a9-11e1-9660-123139220831',
+                'ext': 'mp4',
+                'title': '#whilewewatch',
+            }
+        },
         # AdobeTVVideo embed
         {
             'url': 'https://helpx.adobe.com/acrobat/how-to/new-experience-acrobat-dc.html?set=acrobat--get-started--essential-beginners',
         # AdobeTVVideo embed
         {
             'url': 'https://helpx.adobe.com/acrobat/how-to/new-experience-acrobat-dc.html?set=acrobat--get-started--essential-beginners',
@@ -860,6 +1024,27 @@ class GenericIE(InfoExtractor):
                 'description': 'New experience with Acrobat DC',
                 'duration': 248.667,
             },
                 'description': 'New experience with Acrobat DC',
                 'duration': 248.667,
             },
+        },
+        # ScreenwaveMedia embed
+        {
+            'url': 'http://www.thecinemasnob.com/the-cinema-snob/a-nightmare-on-elm-street-2-freddys-revenge1',
+            'md5': '24ace5baba0d35d55c6810b51f34e9e0',
+            'info_dict': {
+                'id': 'cinemasnob-55d26273809dd',
+                'ext': 'mp4',
+                'title': 'cinemasnob',
+            },
+        },
+        # BrightcoveInPageEmbed embed
+        {
+            'url': 'http://www.geekandsundry.com/tabletop-bonus-wils-final-thoughts-on-dread/',
+            'info_dict': {
+                'id': '4238694884001',
+                'ext': 'flv',
+                'title': 'Tabletop: Dread, Last Thoughts',
+                'description': 'Tabletop: Dread, Last Thoughts',
+                'duration': 51690,
+            },
         }
     ]
 
         }
     ]
 
@@ -1065,11 +1250,15 @@ class GenericIE(InfoExtractor):
 
         self.report_extraction(video_id)
 
 
         self.report_extraction(video_id)
 
-        # Is it an RSS feed?
+        # Is it an RSS feed, a SMIL file or a XSPF playlist?
         try:
         try:
-            doc = parse_xml(webpage)
+            doc = compat_etree_fromstring(webpage.encode('utf-8'))
             if doc.tag == 'rss':
                 return self._extract_rss(url, video_id, doc)
             if doc.tag == 'rss':
                 return self._extract_rss(url, video_id, doc)
+            elif re.match(r'^(?:{[^}]+})?smil$', doc.tag):
+                return self._parse_smil(doc, url, video_id)
+            elif doc.tag == '{http://xspf.org/ns/0/}playlist':
+                return self.playlist_result(self._parse_xspf(doc, video_id), video_id)
         except compat_xml_parse_error:
             pass
 
         except compat_xml_parse_error:
             pass
 
@@ -1081,7 +1270,7 @@ class GenericIE(InfoExtractor):
         # Sometimes embedded video player is hidden behind percent encoding
         # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
         # Unescaping the whole page allows to handle those cases in a generic way
         # Sometimes embedded video player is hidden behind percent encoding
         # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
         # Unescaping the whole page allows to handle those cases in a generic way
-        webpage = compat_urllib_parse.unquote(webpage)
+        webpage = compat_urllib_parse_unquote(webpage)
 
         # it's tempting to parse this further, but you would
         # have to take into account all the variations like
 
         # it's tempting to parse this further, but you would
         # have to take into account all the variations like
@@ -1132,6 +1321,11 @@ class GenericIE(InfoExtractor):
                 'entries': entries,
             }
 
                 'entries': entries,
             }
 
+        # Look for Brightcove In Page Embed:
+        brightcove_in_page_embed_url = BrightcoveInPageEmbedIE._extract_url(webpage)
+        if brightcove_in_page_embed_url:
+            return self.url_result(brightcove_in_page_embed_url, 'BrightcoveInPageEmbed')
+
         # Look for embedded rtl.nl player
         matches = re.findall(
             r'<iframe[^>]+?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+(?:video_)?embed[^"]+)"',
         # Look for embedded rtl.nl player
         matches = re.findall(
             r'<iframe[^>]+?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+(?:video_)?embed[^"]+)"',
@@ -1143,6 +1337,12 @@ class GenericIE(InfoExtractor):
         if vimeo_url is not None:
             return self.url_result(vimeo_url)
 
         if vimeo_url is not None:
             return self.url_result(vimeo_url)
 
+        vid_me_embed_url = self._search_regex(
+            r'src=[\'"](https?://vid\.me/[^\'"]+)[\'"]',
+            webpage, 'vid.me embed', default=None)
+        if vid_me_embed_url is not None:
+            return self.url_result(vid_me_embed_url, 'Vidme')
+
         # Look for embedded YouTube player
         matches = re.findall(r'''(?x)
             (?:
         # Look for embedded YouTube player
         matches = re.findall(r'''(?x)
             (?:
@@ -1269,7 +1469,7 @@ class GenericIE(InfoExtractor):
             return self.url_result(mobj.group('url'))
 
         # Look for Ooyala videos
             return self.url_result(mobj.group('url'))
 
         # Look for Ooyala videos
-        mobj = (re.search(r'player\.ooyala\.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
+        mobj = (re.search(r'player\.ooyala\.com/[^"?]+[?#][^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
                 re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage) or
                 re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P<ec>.{32})[\'"]\)', webpage) or
                 re.search(r'data-ooyala-video-id\s*=\s*[\'"](?P<ec>.{32})[\'"]', webpage))
                 re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage) or
                 re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P<ec>.{32})[\'"]\)', webpage) or
                 re.search(r'data-ooyala-video-id\s*=\s*[\'"](?P<ec>.{32})[\'"]', webpage))
@@ -1335,7 +1535,7 @@ class GenericIE(InfoExtractor):
             return self.url_result(mobj.group('url'))
         mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
         if mobj is not None:
             return self.url_result(mobj.group('url'))
         mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
         if mobj is not None:
-            return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
+            return self.url_result(compat_urllib_parse_unquote(mobj.group('url')))
 
         # Look for funnyordie embed
         matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
 
         # Look for funnyordie embed
         matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
@@ -1398,11 +1598,23 @@ class GenericIE(InfoExtractor):
         if mobj is not None:
             return self.url_result(mobj.group('url'), 'ArteTVEmbed')
 
         if mobj is not None:
             return self.url_result(mobj.group('url'), 'ArteTVEmbed')
 
+        # Look for embedded francetv player
+        mobj = re.search(
+            r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?://)?embed\.francetv\.fr/\?ue=.+?)\1',
+            webpage)
+        if mobj is not None:
+            return self.url_result(mobj.group('url'))
+
         # Look for embedded smotri.com player
         smotri_url = SmotriIE._extract_url(webpage)
         if smotri_url:
             return self.url_result(smotri_url, 'Smotri')
 
         # Look for embedded smotri.com player
         smotri_url = SmotriIE._extract_url(webpage)
         if smotri_url:
             return self.url_result(smotri_url, 'Smotri')
 
+        # Look for embedded Myvi.ru player
+        myvi_url = MyviIE._extract_url(webpage)
+        if myvi_url:
+            return self.url_result(myvi_url)
+
         # Look for embeded soundcloud player
         mobj = re.search(
             r'<iframe\s+(?:[a-zA-Z0-9_-]+="[^"]+"\s+)*src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
         # Look for embeded soundcloud player
         mobj = re.search(
             r'<iframe\s+(?:[a-zA-Z0-9_-]+="[^"]+"\s+)*src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
@@ -1420,12 +1632,9 @@ class GenericIE(InfoExtractor):
             return self.url_result(url, ie='Vulture')
 
         # Look for embedded mtvservices player
             return self.url_result(url, ie='Vulture')
 
         # Look for embedded mtvservices player
-        mobj = re.search(
-            r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"',
-            webpage)
-        if mobj is not None:
-            url = unescapeHTML(mobj.group('url'))
-            return self.url_result(url, ie='MTVServicesEmbedded')
+        mtvservices_url = MTVServicesEmbeddedIE._extract_url(webpage)
+        if mtvservices_url:
+            return self.url_result(mtvservices_url, ie='MTVServicesEmbedded')
 
         # Look for embedded yahoo player
         mobj = re.search(
 
         # Look for embedded yahoo player
         mobj = re.search(
@@ -1464,7 +1673,7 @@ class GenericIE(InfoExtractor):
             return self.url_result(mobj.group('url'), 'MLB')
 
         mobj = re.search(
             return self.url_result(mobj.group('url'), 'MLB')
 
         mobj = re.search(
-            r'<iframe[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
+            r'<(?:iframe|script)[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
             webpage)
         if mobj is not None:
             return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
             webpage)
         if mobj is not None:
             return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
@@ -1482,8 +1691,8 @@ class GenericIE(InfoExtractor):
             return self.url_result(mobj.group('url'), 'Zapiks')
 
         # Look for Kaltura embeds
             return self.url_result(mobj.group('url'), 'Zapiks')
 
         # Look for Kaltura embeds
-        mobj = re.search(
-            r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_id'\s*:\s*'(?P<id>[^']+)',", webpage)
+        mobj = (re.search(r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_?[Ii]d'\s*:\s*'(?P<id>[^']+)',", webpage) or
+                re.search(r'(?s)(?P<q1>["\'])(?:https?:)?//cdnapi(?:sec)?\.kaltura\.com/.*?(?:p|partner_id)/(?P<partner_id>\d+).*?(?P=q1).*?entry_?[Ii]d\s*:\s*(?P<q2>["\'])(?P<id>.+?)(?P=q2)', webpage))
         if mobj is not None:
             return self.url_result('kaltura:%(partner_id)s:%(id)s' % mobj.groupdict(), 'Kaltura')
 
         if mobj is not None:
             return self.url_result('kaltura:%(partner_id)s:%(id)s' % mobj.groupdict(), 'Kaltura')
 
@@ -1550,6 +1759,16 @@ class GenericIE(InfoExtractor):
         if onionstudios_url:
             return self.url_result(onionstudios_url)
 
         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 ScreenwaveMedia embeds
+        mobj = re.search(ScreenwaveMediaIE.EMBED_PATTERN, webpage)
+        if mobj is not None:
+            return self.url_result(unescapeHTML(mobj.group('url')), 'ScreenwaveMedia')
+
         # Look for AdobeTVVideo embeds
         mobj = re.search(
             r'<iframe[^>]+src=[\'"]((?:https?:)?//video\.tv\.adobe\.com/v/\d+[^"]+)[\'"]',
         # Look for AdobeTVVideo embeds
         mobj = re.search(
             r'<iframe[^>]+src=[\'"]((?:https?:)?//video\.tv\.adobe\.com/v/\d+[^"]+)[\'"]',
@@ -1587,7 +1806,7 @@ class GenericIE(InfoExtractor):
         if not found:
             # Broaden the findall a little bit: JWPlayer JS loader
             found = filter_video(re.findall(
         if not found:
             # Broaden the findall a little bit: JWPlayer JS loader
             found = filter_video(re.findall(
-                r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
+                r'[^A-Za-z0-9]?(?:file|video_url)["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
         if not found:
             # Flow player
             found = filter_video(re.findall(r'''(?xs)
         if not found:
             # Flow player
             found = filter_video(re.findall(r'''(?xs)
@@ -1613,7 +1832,7 @@ class GenericIE(InfoExtractor):
                 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
         if not found:
             # HTML5 video
                 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
         if not found:
             # HTML5 video
-            found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage)
+            found = re.findall(r'(?s)<(?:video|audio)[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage)
         if not found:
             REDIRECT_REGEX = r'[0-9]{,2};\s*(?:URL|url)=\'?([^\'"]+)'
             found = re.search(
         if not found:
             REDIRECT_REGEX = r'[0-9]{,2};\s*(?:URL|url)=\'?([^\'"]+)'
             found = re.search(
@@ -1624,9 +1843,12 @@ class GenericIE(InfoExtractor):
                 # Look also in Refresh HTTP header
                 refresh_header = head_response.headers.get('Refresh')
                 if refresh_header:
                 # Look also in Refresh HTTP header
                 refresh_header = head_response.headers.get('Refresh')
                 if refresh_header:
+                    # In python 2 response HTTP headers are bytestrings
+                    if sys.version_info < (3, 0) and isinstance(refresh_header, str):
+                        refresh_header = refresh_header.decode('iso-8859-1')
                     found = re.search(REDIRECT_REGEX, refresh_header)
             if found:
                     found = re.search(REDIRECT_REGEX, refresh_header)
             if found:
-                new_url = compat_urlparse.urljoin(url, found.group(1))
+                new_url = compat_urlparse.urljoin(url, unescapeHTML(found.group(1)))
                 self.report_following_redirect(new_url)
                 return {
                     '_type': 'url',
                 self.report_following_redirect(new_url)
                 return {
                     '_type': 'url',
@@ -1638,7 +1860,7 @@ class GenericIE(InfoExtractor):
         entries = []
         for video_url in found:
             video_url = compat_urlparse.urljoin(url, video_url)
         entries = []
         for video_url in found:
             video_url = compat_urlparse.urljoin(url, video_url)
-            video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
+            video_id = compat_urllib_parse_unquote(os.path.basename(video_url))
 
             # Sometimes, jwplayer extraction will result in a YouTube URL
             if YoutubeIE.suitable(video_url):
 
             # Sometimes, jwplayer extraction will result in a YouTube URL
             if YoutubeIE.suitable(video_url):
@@ -1648,7 +1870,8 @@ class GenericIE(InfoExtractor):
             # here's a fun little line of code for you:
             video_id = os.path.splitext(video_id)[0]
 
             # here's a fun little line of code for you:
             video_id = os.path.splitext(video_id)[0]
 
-            if determine_ext(video_url) == 'smil':
+            ext = determine_ext(video_url)
+            if ext == 'smil':
                 entries.append({
                     'id': video_id,
                     'formats': self._extract_smil_formats(video_url, video_id),
                 entries.append({
                     'id': video_id,
                     'formats': self._extract_smil_formats(video_url, video_id),
@@ -1656,6 +1879,8 @@ class GenericIE(InfoExtractor):
                     'title': video_title,
                     'age_limit': age_limit,
                 })
                     'title': video_title,
                     'age_limit': age_limit,
                 })
+            elif ext == 'xspf':
+                return self.playlist_result(self._extract_xspf_playlist(video_url, video_id), video_id)
             else:
                 entries.append({
                     'id': video_id,
             else:
                 entries.append({
                     'id': video_id,
Totally unverified clone of youtube-dl