debian/NEWS: Update with notes about split audio/video downloads.

[youtubedl] / youtube_dl / extractor / generic.py

diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py
index 764070635160d96487baa53b4370b28e8a957316..69e0a7bd271dd0965f5c1f6f9c3a7cdce7a3da0b 100644 (file)
--- a/youtube_dl/extractor/generic.py
+++ b/youtube_dl/extractor/generic.py
@@ -11,6 +11,8 @@ from ..utils import (
     compat_urlparse,
 
     ExtractorError,
     compat_urlparse,
 
     ExtractorError,

+    smuggle_url,
+    unescapeHTML,

 )
 from .brightcove import BrightcoveIE
 
 )
 from .brightcove import BrightcoveIE
 


@@ -29,6 +31,17 @@ class GenericIE(InfoExtractor):
                 u"title": u"R\u00e9gis plante sa Jeep"
             }
         },
                 u"title": u"R\u00e9gis plante sa Jeep"
             }
         },

+        # embedded vimeo video
+        {
+            u'url': u'http://skillsmatter.com/podcast/home/move-semanticsperfect-forwarding-and-rvalue-references',
+            u'file': u'22444065.mp4',
+            u'md5': u'2903896e23df39722c33f015af0666e2',
+            u'info_dict': {
+                u'title': u'ACCU 2011: Move Semantics,Perfect Forwarding, and Rvalue references- Scott Meyers- 13/04/2011',
+                u"uploader_id": u"skillsmatter",
+                u"uploader": u"Skills Matter",
+            }
+        }

     ]
 
     def report_download_webpage(self, video_id):
     ]
 
     def report_download_webpage(self, video_id):


@@ -117,16 +130,31 @@ class GenericIE(InfoExtractor):
         except ValueError:
             # since this is the last-resort InfoExtractor, if
             # this error is thrown, it'll be thrown here
         except ValueError:
             # since this is the last-resort InfoExtractor, if
             # this error is thrown, it'll be thrown here

-            raise ExtractorError(u'Invalid URL: %s' % url)
+            raise ExtractorError(u'Failed to download URL: %s' % url)

 
         self.report_extraction(video_id)
         # Look for BrightCove:
 
         self.report_extraction(video_id)
         # Look for BrightCove:

-        m_brightcove = re.search(r'<object.+?class=([\'"]).*?BrightcoveExperience.*?\1.+?</object>', webpage, re.DOTALL)
+        m_brightcove = re.search(r'<object[^>]+?class=([\'"])[^>]*?BrightcoveExperience.*?\1.+?</object>', webpage, re.DOTALL)

         if m_brightcove is not None:
             self.to_screen(u'Brightcove video detected.')
             bc_url = BrightcoveIE._build_brighcove_url(m_brightcove.group())
             return self.url_result(bc_url, 'Brightcove')
 
         if m_brightcove is not None:
             self.to_screen(u'Brightcove video detected.')
             bc_url = BrightcoveIE._build_brighcove_url(m_brightcove.group())
             return self.url_result(bc_url, 'Brightcove')
 

+        # Look for embedded Vimeo player
+        mobj = re.search(
+            r'<iframe[^>]+?src="(https?://player.vimeo.com/video/.+?)"', webpage)
+        if mobj:
+            player_url = unescapeHTML(mobj.group(1))
+            surl = smuggle_url(player_url, {'Referer': url})
+            return self.url_result(surl, 'Vimeo')
+
+        # Look for embedded YouTube player
+        mobj = re.search(
+            r'<iframe[^>]+?src="(https?://(?:www\.)?youtube.com/embed/.+?)"', webpage)
+        if mobj:
+            surl = unescapeHTML(mobj.group(1))
+            return self.url_result(surl, 'Youtube')
+

         # Start with something easy: JW Player in SWFObject
         mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
         if mobj is None:
         # Start with something easy: JW Player in SWFObject
         mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
         if mobj is None:


@@ -149,12 +177,12 @@ class GenericIE(InfoExtractor):
             # HTML5 video
             mobj = re.search(r'<video[^<]*(?:>.*?<source.*?)? src="([^"]+)"', webpage, flags=re.DOTALL)
         if mobj is None:
             # HTML5 video
             mobj = re.search(r'<video[^<]*(?:>.*?<source.*?)? src="([^"]+)"', webpage, flags=re.DOTALL)
         if mobj is None:

-            raise ExtractorError(u'Invalid URL: %s' % url)
+            raise ExtractorError(u'Unsupported URL: %s' % url)

 
         # It's possible that one of the regexes
         # matched, but returned an empty group:
         if mobj.group(1) is None:
 
         # It's possible that one of the regexes
         # matched, but returned an empty group:
         if mobj.group(1) is None:

-            raise ExtractorError(u'Invalid URL: %s' % url)
+            raise ExtractorError(u'Did not find a valid video URL at %s' % url)

 
         video_url = mobj.group(1)
         video_url = compat_urlparse.urljoin(url, video_url)
 
         video_url = mobj.group(1)
         video_url = compat_urlparse.urljoin(url, video_url)