You can create a data structure that parses name data from a single column into multiple columns. For example, you can create columns for salutations, first names, middle names, and surnames.

You can create a data structure that parses a single column of address data into multiple columns that describe a deliverable address.

Configure the transformation with reference tables that contain recognizable address elements, such as ZIP Codes, state names, and city names. Create a token parsing strategy that writes each address element to a new port.

You cannot use a reference table to parse street address data from an input string, because street name and number data is too general to be captured in a reference table. However, you can use the Overflow port to capture this data. When you have parsed all city, state, and ZIP data from an address, the remaining data contains street information.

For example, use a token parsing strategy to split the following address into address elements:

123 MAIN ST NW STE 12 ANYTOWN NY 12345

The parsing strategy can write the address elements to the following columns:

Column Name	Data
Overflow	123 MAIN ST NW STE 12
City	ANYTOWN
State	NY
ZIP	12345

You can create a data structure that parses a single column of product data into multiple columns that describe the product inventory details.

Configure the transformation with token sets that contain inventory elements, such as dimension, color, and weight. Create a token parsing strategy that writes each inventory element to a new port

For example, use a token parsing strategy to split the following paint description into separate inventory elements:

500ML Red Matt Exterior

The parsing strategy can write the values to the following columns:

Column Name	Data
Size	500ML
Color	Red
Style	Matt
Exterior	Y