[eagleplatform] Checking direct HTTP links

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

diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py
index f3de738f765819da7cbda1332e9b6b6dcada5052..a95501d86b6874e2775046e142fb81c4469ed2a0 100644 (file)
--- a/youtube_dl/extractor/generic.py
+++ b/youtube_dl/extractor/generic.py
@@ -60,6 +60,7 @@ from .googledrive import GoogleDriveIE
 from .jwplatform import JWPlatformIE
 from .digiteka import DigitekaIE
 from .instagram import InstagramIE
+from .liveleak import LiveLeakIE
 
 
 class GenericIE(InfoExtractor):
@@ -104,7 +105,8 @@ class GenericIE(InfoExtractor):
                 'skip_download': True,  # infinite live stream
             },
             'expected_warnings': [
-                r'501.*Not Implemented'
+                r'501.*Not Implemented',
+                r'400.*Bad Request',
             ],
         },
         # Direct link with incorrect MIME type
@@ -235,6 +237,7 @@ class GenericIE(InfoExtractor):
                 'ext': 'mp4',
                 'title': 'car-20120827-manifest',
                 'formats': 'mincount:9',
+                'upload_date': '20130904',
             },
             'params': {
                 'format': 'bestvideo',
@@ -406,19 +409,6 @@ class GenericIE(InfoExtractor):
                 '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',
-            'info_dict': {
-                'id': 'national-signing-day-rationalizations-itll-be-ok-itll-be-ok',
-                'title': '25 lies you will tell yourself on National Signing Day - SBNation.com',
-            },
-            'playlist_mincount': 3,
-            'params': {
-                'skip_download': True,
-            },
-            'add_ie': ['Ooyala'],
-        },
         # embed.ly video
         {
             'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
@@ -607,7 +597,11 @@ 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'],
         },
@@ -740,8 +734,11 @@ 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
@@ -890,6 +887,7 @@ 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',
@@ -904,6 +902,7 @@ 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',
@@ -992,6 +991,9 @@ 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
@@ -1044,6 +1046,9 @@ 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
@@ -1135,12 +1140,39 @@ 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
                 'skip_download': True,
             },
         },
+        # Another form of arte.tv embed
+        {
+            'url': 'http://www.tv-replay.fr/redirection/09-04-16/arte-reportage-arte-11508975.html',
+            'md5': '850bfe45417ddf221288c88a0cffe2e2',
+            'info_dict': {
+                'id': '030273-562_PLUS7-F',
+                'ext': 'mp4',
+                'title': 'ARTE Reportage - Nulle part, en France',
+                'description': 'md5:e3a0e8868ed7303ed509b9e3af2b870d',
+                'upload_date': '20160409',
+            },
+        },
+        # LiveLeak embed
+        {
+            'url': 'http://www.wykop.pl/link/3088787/',
+            'md5': 'ace83b9ed19b21f68e1b50e844fdf95d',
+            'info_dict': {
+                'id': '874_1459135191',
+                'ext': 'mp4',
+                'title': 'Man shows poor quality of new apartment building',
+                'description': 'The wall is like a sand pile.',
+                'uploader': 'Lake8737',
+            }
+        },
     ]
 
     def report_following_redirect(self, new_url):
@@ -1715,7 +1747,7 @@ class GenericIE(InfoExtractor):
 
         # Look for embedded arte.tv player
         mobj = re.search(
-            r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
+            r'<(?:script|iframe) [^>]*?src="(?P<url>http://www\.arte\.tv/(?:playerv2/embed|arte_vp/index)[^"]+)"',
             webpage)
         if mobj is not None:
             return self.url_result(mobj.group('url'), 'ArteTVEmbed')
@@ -1943,7 +1975,13 @@ class GenericIE(InfoExtractor):
         # Look for Instagram embeds
         instagram_embed_url = InstagramIE._extract_embed_url(webpage)
         if instagram_embed_url is not None:
-            return self.url_result(instagram_embed_url, InstagramIE.ie_key())
+            return self.url_result(
+                self._proto_relative_url(instagram_embed_url), InstagramIE.ie_key())
+
+        # Look for LiveLeak embeds
+        liveleak_url = LiveLeakIE._extract_url(webpage)
+        if liveleak_url:
+            return self.url_result(liveleak_url, 'LiveLeak')
 
         def check_video(vurl):
             if YoutubeIE.suitable(vurl):
@@ -2026,6 +2064,7 @@ 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))