User guide

Operator reference
156 Data Integration with Sybase Avaki Studio
Join types
Avaki Studio supports inner, outer-left, outer-right, and outer-full join types. These
join types work in the same way as their SQL counterparts. They’re summarized here:
Defining join columns
Studio needs to know which columns to match in order to join your input result sets.
You tell it using the table at the bottom of the Join Properties dialog.
• Click the Add button to add a column pair to your join.
• To delete one or more column pairs, select them and click Delete.
• You can edit the specific columns in each pair (When you add a new column pair,
Studio chooses the next available column from the two inputs, which may not nec-
essarily be the columns you want to join on). Note that if you have already defined
columns and you change which input result set is the inner table, Studio swaps your
column choices to match.
 Inner Column: Click in field and use the pull-down list to choose a column from
the inner table.
Join Type Description
Inner Inner joins find the intersection between the two result sets. If no
matches appear in the outer input for a given row in the inner input,
no row is output. If a given inner input matches multiple outer input
rows, then the data from the inner input is repeated for each mach,
combined with each of the outer input’s rows.
Outer Left In the case of the outer left join, if a row in the inner input doesn’t
match anything in the outer input, a row will still be output, but with
the value
NULL for each of the columns which comes from the outer
input.
Outer Right As you might expect, this is the opposite of the outer left join. In the
case of the outer right join, if a row in the outer input doesn’t match
anything in the inner input, a row will still be output, but with the
value
NULL for each of the columns which comes from the inner
input.
Outer Full This is a combination of the outer left join and the outer right join
types. If a row from either input result set doesn’t have a matching
row in the other result set, it is still included, but with
NULL values
for the missing columns.