Message-ID: <4C9AB69F.7080009@postnewspapers.com.au>
Date: Thu, 23 Sep 2010 10:08:31 +0800
From: Craig Ringer <craig@postnewspapers.com.au>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.1; en-GB;
	rv:1.9.1.12) Gecko/20100914 Thunderbird/3.0.8
MIME-Version: 1.0
To: Andreas <maps.on@gmx.net>
CC: pgsql-sql@postgresql.org
Subject: Re: HowTo divide  streetname from house-nr ?
References: <4C9AAC85.1000001@gmx.net>
In-Reply-To: <4C9AAC85.1000001@gmx.net>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit

On 23/09/2010 9:25 AM, Andreas wrote:
> Hi,
>
> how could I divide streetnames from housenumbers ?

You can't, reliably. There are papers written on the subject of 
addressing, address analysis, addressing in databases, etc.

How would you handle the address:

   Person's Name
   RD3 Clemo Rd
   Whangarei

? "RD3" is "Rural Delivery Area 3". The posties deliver within the area 
by named recipient. This is a real address scheme.

The world is full of weird and wacky addressing. IMO, unless you're 
willing to confine your schema to only handling addresses of a 
particular area you know about, don't try to normalize address. Even 
then, I wouldn't try to normalize addresses with text processing, I'd 
ask the user to do it during form entry or not do it at all.

If you're trying to find duplicate addresses, matching addreses, etc, 
then IMO you're better off using existing tools that do this with 
free-form addresses using national phone databases, postcode databases, etc.

-- 
Craig Ringer

Tech-related writing at http://soapyfrogs.blogspot.com/