[mlb] Fix extraction of articles

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

diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py
index e645d1bb32d26f1e93ba57c93bd4b211d362d2fa..3d756e8481e0aba09d2290cfd5a4a8b21369aa7d 100644 (file)
--- a/youtube_dl/extractor/generic.py
+++ b/youtube_dl/extractor/generic.py
@@ -35,6 +35,9 @@ from .rutv import RUTVIE
 from .smotri import SmotriIE
 from .condenast import CondeNastIE
 from .udn import UDNEmbedIE
+from .senateisvp import SenateISVPIE
+from .bliptv import BlipTVIE
+from .svt import SVTIE
 
 
 class GenericIE(InfoExtractor):
@@ -643,6 +646,17 @@ class GenericIE(InfoExtractor):
                 'title': 'Facebook Creates "On This Day" | Crunch Report',
             },
         },
+        # SVT embed
+        {
+            'url': 'http://www.svt.se/sport/ishockey/jagr-tacklar-giroux-under-intervjun',
+            'info_dict': {
+                'id': '2900353',
+                'ext': 'flv',
+                'title': 'Här trycker Jagr till Giroux (under SVT-intervjun)',
+                'duration': 27,
+                'age_limit': 0,
+            },
+        },
         # RSS feed with enclosure
         {
             'url': 'http://podcastfeeds.nbcnews.com/audio/podcast/MSNBC-MADDOW-NETCAST-M4V.xml',
@@ -1072,12 +1086,14 @@ class GenericIE(InfoExtractor):
             }
 
         # Look for embedded blip.tv player
-        mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)
-        if mobj:
-            return self.url_result('http://blip.tv/a/a-' + mobj.group(1), 'BlipTV')
-        mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9_]+)', webpage)
-        if mobj:
-            return self.url_result(mobj.group(1), 'BlipTV')
+        bliptv_url = BlipTVIE._extract_url(webpage)
+        if bliptv_url:
+            return self.url_result(bliptv_url, 'BlipTV')
+
+        # Look for SVT player
+        svt_url = SVTIE._extract_url(webpage)
+        if svt_url:
+            return self.url_result(svt_url, 'SVT')
 
         # Look for embedded condenast player
         matches = re.findall(
@@ -1290,6 +1306,10 @@ class GenericIE(InfoExtractor):
         mobj = re.search(
             r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
             webpage)
+        if not mobj:
+            mobj = re.search(
+                r'data-video-link=["\'](?P<url>http://m.mlb.com/video/[^"\']+)',
+                webpage)
         if mobj is not None:
             return self.url_result(mobj.group('url'), 'MLB')
 
@@ -1365,6 +1385,11 @@ class GenericIE(InfoExtractor):
             return self.url_result(
                 compat_urlparse.urljoin(url, mobj.group('url')), 'UDNEmbed')
 
+        # Look for Senate ISVP iframe
+        senate_isvp_url = SenateISVPIE._search_iframe_url(webpage)
+        if senate_isvp_url:
+            return self.url_result(surl, 'SenateISVP')
+
         def check_video(vurl):
             if YoutubeIE.suitable(vurl):
                 return True
@@ -1432,7 +1457,7 @@ class GenericIE(InfoExtractor):
                 if refresh_header:
                     found = re.search(REDIRECT_REGEX, refresh_header)
             if found:
-                new_url = found.group(1)
+                new_url = compat_urlparse.urljoin(url, found.group(1))
                 self.report_following_redirect(new_url)
                 return {
                     '_type': 'url',