Parsing a text string Address field into multiple fields, HELP needed

Legacy problem here.I have a 7000 documents which contain a single text field for address’s. I now need to split these out into country, city, address 1 to 5 and postcode.

Problem is that as they are stored as a single text string, I have no comminality in the address’s.

eg1: “Arrow Stevenage”

eg 2: “TDG, Howley Lane, Warrington, WA1 2EB”

eg 3: “Gold Group House Croydon CR30GG”

eg 4: “12003 Cent Avenue, Century City, LA, 90211”

Now I know that these are a mess, but does anyone out there know of or have any code, or any ideas, that can help me parse these strings out? Even if it will only give a 50% success, I need help here.

Subject: Parsing a text string Address field into multiple fields, HELP needed.

You have a couple of options…none of which will get you 100% but since you are only shooting for 50%, you could look at some of these options:

  1. Try converting your string to a list using @Explode (Split in LS) with the separator being a comma (,). Then you could try to examine each element in the list and try to determine what field it should map to. You could use a combination of @length and @matches or other string comparison function to help figure out what piece of the address you are looking at.

  2. Similar to above except using the @Word function to your string into separate parts.

Good Luck!