The join condition defines when incoming rows are joined. It includes fields from both sources that must match to join source rows.
You define one or more conditions based on equality between the master and detail data. For example, if two sets of employee data contain employee ID numbers, the following condition matches rows with the same employee IDs in both sets of data:
EMP_ID1 = EMP_ID2
Use one or more join conditions. Additional join conditions increase the time necessary to join the data. When you use multiple join conditions, the mapping task evaluates the conditions in the order that you specify.
Both fields in a condition must have the same data type. If you need to use two fields with non-matching data types, convert the data types so they match.
For example, when you try to join Char and Varchar data, any spaces that pad Char values are included as part of the string. Both fields might include the value "Shoes," but because the Char(40) field includes 35 trailing spaces, the values do not match. To ensure that the values match, change the data type of one field to match the other.
In advanced mode, the join condition can’t contain a binary data type or evaluate to a binary data type.
Note: The Joiner transformation does not match null values. To join rows with null values, you can replace null values with default values, and then join on the default values.