How to do overriding in preprocessing for address?

[tverma27]

Hi

The source data of client is having 3 address line input.
1) Address line 1 usually holds the address ;
2) Addr line 2 is holding a variety of data.
At times, it is having
i) city and state info.
ii) apartment info and city info
iii) only state info
iv) Apartment info
3) Zip column is having 5 digits of zip info.


When in column 2 case i and iii are coming then preprocessing is happening properly but when case ii & iv is happening then is output is not coming as expected.
Preprocessing is moving everything to ADDress domain and then USADDR.set is unable to handle the pattern.
This output is fed to CASS stage; it is able to populate the city and state but it is retaining the ADDress line 2 value as well. As a result the redudnt info is present in address line 2.

We dont want to have the city and state info in addressline 2.

Would like to know how the pattern over-riding is done to move the values in right bucket in preprocessing stage itself or is there some work around to achieve this?

Thanks in advance!

[tverma27]

Example to explain the above scenerios:

Before Standardization
-------------------------------
addr1: 2 CHAMPLAIN RD
addr2: MONMOUTH JCT
zip: 08852

After Standardization
----------------------------
addr1: 2 CHAMPLAIN RD MONMOUTH JCT
City: MONMOUTH JUNCTION
zip: 08852


Before Standardization
-------------------------------
addr1: 2 CHAMPLAIN RD
addr2: APT 2 MONMOUTH JCT
zip: 08852

After Standardization
----------------------------
addr1: 2 CHAMPLAIN RD
ADDR2: APT 2 MONMOUTH JCT
City: MONMOUTH JUNCTION
zip: 08852

[mekrreddy]

try usprep rule set

[tverma27]

Thanks for the reply,
we are using usprep rule then standardization and then cass...
but recently we tried column pattern overriding in USPREP ruleset and it is helping us.

This issue is almost resolved.

Thanks