[generic] Support embedded vimeo videos (#1602)

test/test_utils.py

@@ -1,4 +1,5 @@
 #!/usr/bin/env python
+# coding: utf-8
 
 # Allow direct execution
 import os
@@ -21,6 +22,8 @@
     find_xpath_attr,
     get_meta_content,
     xpath_with_ns,
+    smuggle_url,
+    unsmuggle_url,
 )
 
 if sys.version_info < (3, 0):
@@ -155,5 +158,18 @@ def test_xpath_with_ns(self):
         self.assertEqual(find('media:song/media:author').text, u'The Author')
         self.assertEqual(find('media:song/url').text, u'http://server.com/download.mp3')
 
+    def test_smuggle_url(self):
+        data = {u"ö": u"ö", u"abc": [3]}
+        url = 'https://foo.bar/baz?x=y#a'
+        smug_url = smuggle_url(url, data)
+        unsmug_url, unsmug_data = unsmuggle_url(smug_url)
+        self.assertEqual(url, unsmug_url)
+        self.assertEqual(data, unsmug_data)
+
+        res_url, res_data = unsmuggle_url(url)
+        self.assertEqual(res_url, url)
+        self.assertEqual(res_data, None)
+
+
 if __name__ == '__main__':
     unittest.main()

youtube_dl/extractor/generic.py

@@ -11,6 +11,8 @@
     compat_urlparse,
 
     ExtractorError,
+    smuggle_url,
+    unescapeHTML,
 )
 from .brightcove import BrightcoveIE
 
@@ -29,6 +31,17 @@ class GenericIE(InfoExtractor):
                 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):
@@ -127,6 +140,14 @@ def _real_extract(self, url):
             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\s+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')
+
         # Start with something easy: JW Player in SWFObject
         mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
         if mobj is None:

youtube_dl/extractor/vimeo.py

@@ -11,6 +11,7 @@
     get_element_by_attribute,
     ExtractorError,
     std_headers,
+    unsmuggle_url,
 )
 
 class VimeoIE(InfoExtractor):
@@ -53,7 +54,7 @@ class VimeoIE(InfoExtractor):
                 u'title': u'Kathy Sierra: Building the minimum Badass User, Business of Software',
                 u'uploader': u'The BLN & Business of Software',
             },
-        },
+        }
     ]
 
     def _login(self):
@@ -98,6 +99,12 @@ def _real_initialize(self):
         self._login()
 
     def _real_extract(self, url, new_video=True):
+        url, data = unsmuggle_url(url)
+        headers = std_headers
+        if data is not None:
+            headers = headers.copy()
+            headers.update(data)
+
         # Extract ID from URL
         mobj = re.match(self._VALID_URL, url)
         if mobj is None:
@@ -112,7 +119,7 @@ def _real_extract(self, url, new_video=True):
             url = 'https://vimeo.com/' + video_id
 
         # Retrieve video webpage to extract further information
-        request = compat_urllib_request.Request(url, None, std_headers)
+        request = compat_urllib_request.Request(url, None, headers)
         webpage = self._download_webpage(request, video_id)
 
         # Now we begin extracting as much information as we can from what we

youtube_dl/utils.py

@@ -945,3 +945,20 @@ def read(self, *args):
 
 def shell_quote(args):
     return ' '.join(map(pipes.quote, args))
+
+
+def smuggle_url(url, data):
+    """ Pass additional data in a URL for internal use. """
+
+    sdata = compat_urllib_parse.urlencode(
+        {u'__youtubedl_smuggle': json.dumps(data)})
+    return url + u'#' + sdata
+
+
+def unsmuggle_url(smug_url):
+    if not '#__youtubedl_smuggle' in smug_url:
+        return smug_url, None
+    url, _, sdata = smug_url.rpartition(u'#')
+    jsond = compat_parse_qs(sdata)[u'__youtubedl_smuggle'][0]
+    data = json.loads(jsond)
+    return url, data