Merge pull request #5328 from yan12125/fix_5226

[Yahoo/NBCSports] Fix 5226 and add support for NBC sports

5 files changed, 85 insertions, 1 deletions

diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py
index 43bac0252..9fddb8e32 100644
--- a/youtube_dl/extractor/__init__.py
+++ b/youtube_dl/extractor/__init__.py
@@ -311,6 +311,8 @@ from .nba import NBAIE
 from .nbc import (
     NBCIE,
     NBCNewsIE,
+    NBCSportsIE,
+    NBCSportsVPlayerIE,
 )
 from .ndr import NDRIE
 from .ndtv import NDTVIE
diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py
index 042d23a13..9ddf36f6b 100644
--- a/youtube_dl/extractor/generic.py
+++ b/youtube_dl/extractor/generic.py
@@ -29,6 +29,7 @@ from ..utils import (
     xpath_text,
 )
 from .brightcove import BrightcoveIE
+from .nbc import NBCSportsVPlayerIE
 from .ooyala import OoyalaIE
 from .rutv import RUTVIE
 from .smotri import SmotriIE
@@ -639,6 +640,15 @@ class GenericIE(InfoExtractor):
                 'upload_date': '20150228',
                 'title': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
             }
+        },
+        # NBC Sports vplayer embeds
+        {
+            'url': 'http://bbs.clutchfans.net/showthread.php?t=244180',
+            'info_dict': {
+                'id': '_hqLjQ95yx8Z',
+                'ext': 'flv'
+            },
+            'skip': 'This content expired on 9/17/14 12:23 PM',
         }
     ]
 
@@ -1252,6 +1262,11 @@ class GenericIE(InfoExtractor):
         if mobj is not None:
             return self.url_result('5min:%s' % mobj.group('id'), 'FiveMin')
 
+        # Look for NBC Sports VPlayer embeds
+        nbc_sports_url = NBCSportsVPlayerIE._extract_url(webpage)
+        if nbc_sports_url:
+            return self.url_result(nbc_sports_url, 'NBCSportsVPlayer')
+
         def check_video(vurl):
             if YoutubeIE.suitable(vurl):
                 return True
diff --git a/youtube_dl/extractor/nbc.py b/youtube_dl/extractor/nbc.py
index 80a01c778..b7f6a5366 100644
--- a/youtube_dl/extractor/nbc.py
+++ b/youtube_dl/extractor/nbc.py
@@ -50,6 +50,58 @@ class NBCIE(InfoExtractor):
         return self.url_result(theplatform_url)
 
 
+class NBCSportsVPlayerIE(InfoExtractor):
+    _VALID_URL = r'https?://vplayer\.nbcsports\.com/(?:[^/]+/)+(?P<id>[0-9a-zA-Z_]+)'
+
+    _TESTS = [{
+        'url': 'https://vplayer.nbcsports.com/p/BxmELC/nbcsports_share/select/9CsDKds0kvHI',
+        'info_dict': {
+            'id': '9CsDKds0kvHI',
+            'ext': 'flv',
+            'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d',
+            'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson',
+        }
+    }, {
+        'note': 'This video is already expired. It\'s for testing _VALID_URL',
+        'url': 'http://vplayer.nbcsports.com/p/BxmELC/nbc_embedshare/select/_hqLjQ95yx8Z',
+        'only_matching': True,
+    }]
+
+    @staticmethod
+    def _extract_url(webpage):
+        iframe_m = re.search(
+            r'<iframe[^>]+src="(?P<url>https?://vplayer\.nbcsports\.com/[^"]+)"', webpage)
+        if iframe_m:
+            return iframe_m.group('url')
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+        theplatform_url = self._og_search_video_url(webpage)
+        return self.url_result(theplatform_url, 'ThePlatform')
+
+
+class NBCSportsIE(InfoExtractor):
+    # Does not include https becuase its certificate is invalid
+    _VALID_URL = r'http://www\.nbcsports\.com//?(?:[^/]+/)+(?P<id>[0-9a-z-]+)'
+
+    _TEST = {
+        'url': 'http://www.nbcsports.com//college-basketball/ncaab/tom-izzo-michigan-st-has-so-much-respect-duke',
+        'info_dict': {
+            'id': 'PHJSaFWbrTY9',
+            'ext': 'flv',
+            'title': 'Tom Izzo, Michigan St. has \'so much respect\' for Duke',
+            'description': 'md5:ecb459c9d59e0766ac9c7d5d0eda8113',
+        }
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+        return self.url_result(
+            NBCSportsVPlayerIE._extract_url(webpage), 'NBCSportsVPlayer')
+
+
 class NBCNewsIE(InfoExtractor):
     _VALID_URL = r'''(?x)https?://(?:www\.)?nbcnews\.com/
         (?:video/.+?/(?P<id>\d+)|
diff --git a/youtube_dl/extractor/theplatform.py b/youtube_dl/extractor/theplatform.py
index feac666f7..0e3e627f4 100644
--- a/youtube_dl/extractor/theplatform.py
+++ b/youtube_dl/extractor/theplatform.py
@@ -92,7 +92,7 @@ class ThePlatformIE(InfoExtractor):
             error_msg = next(
                 n.attrib['abstract']
                 for n in meta.findall(_x('.//smil:ref'))
-                if n.attrib.get('title') == 'Geographic Restriction')
+                if n.attrib.get('title') == 'Geographic Restriction' or n.attrib.get('title') == 'Expired')
         except StopIteration:
             pass
         else:
diff --git a/youtube_dl/extractor/yahoo.py b/youtube_dl/extractor/yahoo.py
index 97dbac4cc..b777159c5 100644
--- a/youtube_dl/extractor/yahoo.py
+++ b/youtube_dl/extractor/yahoo.py
@@ -17,6 +17,8 @@ from ..utils import (
     int_or_none,
 )
 
+from .nbc import NBCSportsVPlayerIE
+
 
 class YahooIE(InfoExtractor):
     IE_DESC = 'Yahoo screen and movies'
@@ -129,6 +131,15 @@ class YahooIE(InfoExtractor):
         }, {
             'url': 'https://gma.yahoo.com/pizza-delivery-man-surprised-huge-tip-college-kids-195200785.html',
             'only_matching': True,
+        }, {
+            'note': 'NBC Sports embeds',
+            'url': 'http://sports.yahoo.com/blogs/ncaab-the-dagger/tyler-kalinoski-s-buzzer-beater-caps-davidson-s-comeback-win-185609842.html?guid=nbc_cbk_davidsonbuzzerbeater_150313',
+            'info_dict': {
+                'id': '9CsDKds0kvHI',
+                'ext': 'flv',
+                'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d',
+                'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson',
+            }
         }
     ]
 
@@ -151,6 +162,10 @@ class YahooIE(InfoExtractor):
                 items = json.loads(items_json)
                 video_id = items[0]['id']
                 return self._get_info(video_id, display_id, webpage)
+        # Look for NBCSports iframes
+        nbc_sports_url = NBCSportsVPlayerIE._extract_url(webpage)
+        if nbc_sports_url:
+            return self.url_result(nbc_sports_url, 'NBCSportsVPlayer')
 
         items_json = self._search_regex(
             r'mediaItems: ({.*?})$', webpage, 'items', flags=re.MULTILINE,