Include original page url in the file name sent to the editor.

This is similar to what itsalltext does, and allows editors to adjust their
behavior based on the site or page being loaded.

pycl.py

@@ -17,7 +17,7 @@
 import cgi, urlparse
 import subprocess
 import tempfile, time
-import os, sys
+import os, sys, re
 from BaseHTTPServer import BaseHTTPRequestHandler, HTTPServer
 
 temp_has_delete=True
@@ -50,11 +50,15 @@ class Handler(BaseHTTPRequestHandler):
             print l
 
             # write text into file
+            url = self.headers.getheader('x-url')
+            print "url:", url
+            prefix = "chrome_" + re.sub("[^.\w]", "_", re.sub("^.*?//","",url))
             if temp_has_delete==True:
-                f = tempfile.NamedTemporaryFile(delete=False, suffix='.txt')
+                f = tempfile.NamedTemporaryFile(
+                        delete=False, prefix=prefix, suffix='.txt')
                 fname = f.name
             else:
-                tf = tempfile.mkstemp(suffix='.txt')
+                tf = tempfile.mkstemp(prefix=prefix, suffix='.txt')
                 f = os.fdopen(tf[0],"w")
                 fname = tf[1]