Added 9 digits zip support + cities db updates

README

@@ -82,7 +82,7 @@ The state of the address, preferably following the city and a comma. Always two
 
 `zip`
 
-The 5 digit zip code of the address, preferably following the state. 9 digit zips not yet supported. E.g. 123 W. Mifflin St., Madison, WI __53703__
+The 5 or 9 digit zip code of the address, preferably following the state. Supported 9 digit zips format is (xxxxx-xxxx). E.g. 123 W. Mifflin St., Madison, WI __53703__
 
 `full_address()`
 

README.rst

@@ -101,8 +101,8 @@ Always two capitalized letters. E.g. 123 W. Mifflin St., Madison, **WI**
 
 ``zip``
 
-The 5 digit zip code of the address, preferably following the state. 9
-digit zips not yet supported. E.g. 123 W. Mifflin St., Madison, WI
+The 5 or 9 digit zip code of the address, preferably following the state. 9
+digit zips supported in the format (xxxxx-xxx). E.g. 123 W. Mifflin St., Madison, WI
 **53703**
 
 ``full_address()``

address/address.py

@@ -280,7 +280,7 @@ def preprocess_address(self, address):
 
     def check_zip(self, token):
         """
-        Returns true if token is matches a zip code (5 numbers). Zip code must be the last token in an address (minus anything
+        Returns true if token is matches a zip code (5 numbers or 9 numbers). Zip code must be the last token in an address (minus anything
         removed during preprocessing such as --2 units.
         """
         if self.zip is None:
@@ -290,6 +290,8 @@ def check_zip(self, token):
                 # print "zip check", len(token) == 5, re.match(r"\d{5}", token)
             if len(token) == 5 and re.match(r"\d{5}", token):
                 self.zip = self._clean(token)
+            if re.match(r"\d{5}-?\d{4}", token):
+                self.zip = self._clean(token)
 
                 return True
         return False