diff --git a/planet/spider.py b/planet/spider.py
index 0ad530d..454ef5c 100644
--- a/planet/spider.py
+++ b/planet/spider.py
@@ -14,8 +14,8 @@ except:
     PrettyPrint = None
 
 # Regular expressions to sanitise cache filenames
-re_url_scheme    = re.compile(r'^[^:]*://')
-re_slash         = re.compile(r'[?/]+')
+re_url_scheme    = re.compile(r'^\w+:/*(\w+:|www\.)?')
+re_slash         = re.compile(r'[?/:]+')
 re_initial_cruft = re.compile(r'^[,.]*')
 re_final_cruft   = re.compile(r'[,.]*$')
 
@@ -68,9 +68,13 @@ def spiderFeed(feed):
         xml = reconstitute.reconstitute(data, entry)
         
         file = open(out,'w')
-        if PrettyPrint:
+        try:
             PrettyPrint(reconstitute.reconstitute(data, entry), file)
-        else:
+        except:
+            # known reasons for failure include no pretty printer installed,
+            # and absurdly high levels of markup nesting causing Python to
+            # declare infinite recursion.
+            file.seek(0)
             file.write(reconstitute.reconstitute(data, entry).toxml('utf-8'))
         file.close()
 
diff --git a/tests/test_spider.py b/tests/test_spider.py
index 0954539..3bd9284 100644
--- a/tests/test_spider.py
+++ b/tests/test_spider.py
@@ -24,8 +24,12 @@ class SpiderTest(unittest.TestCase):
     def test_filename(self):
         self.assertEqual('./example.com,index.html',
             filename('.', 'http://example.com/index.html'))
-        self.assertEqual('./www.xn--8ws00zhy3a.com',
+        self.assertEqual('./xn--8ws00zhy3a.com',
             filename('.', u'http://www.\u8a79\u59c6\u65af.com/'))
+        self.assertEqual('./planet.intertwingly.net,2006,testfeed1,1',
+            filename('.', u'tag:planet.intertwingly.net,2006:testfeed1,1'))
+        self.assertEqual('./00000000-0000-0000-0000-000000000000',
+            filename('.', u'urn:uuid:00000000-0000-0000-0000-000000000000'))
 
     def test_spiderFeed(self):
         config.load(configfile)
@@ -37,7 +41,7 @@ class SpiderTest(unittest.TestCase):
 
         # verify that the file names are as expected
         self.assertTrue(workdir + 
-            '/tag:planet.intertwingly.net,2006:testfeed1,1' in files)
+            '/planet.intertwingly.net,2006,testfeed1,1' in files)
 
         # verify that the file timestamps match atom:updated
         for file in files:
@@ -59,7 +63,7 @@ class SpiderTest(unittest.TestCase):
 
         # verify that the file names are as expected
         self.assertTrue(workdir + 
-            '/tag:planet.intertwingly.net,2006:testfeed1,1' in files)
+            '/planet.intertwingly.net,2006,testfeed1,1' in files)
         self.assertTrue(workdir + 
-            '/tag:planet.intertwingly.net,2006:testfeed2,1' in files)
+            '/planet.intertwingly.net,2006,testfeed2,1' in files)