In Dataflow Gen2, you plan to merge the Sales table with ExceptionRecords so that records present in ExceptionRecords are excluded from Sales, while maintaining query folding. Which applied steps should you configure?

In Dataflow Gen2, you plan to merge the Sales table with ExceptionRecords so that records present in ExceptionRecords are excluded from Sales, while maintaining query folding. Which applied steps should you configure?

• A. Merge (left anti join) applied step, and then the expand columns applied step
• B. Merge (inner join) applied step, and then the remove columns applied step
• C. Join (full outer) applied step, and then the expand columns applied step
• D. Merge (left outer) applied step, and then the add columns applied step

Answer: (A)

The idea being tested is filtering a table by removing any rows that appear in a secondary table using an anti join, while keeping the query folded so the work is pushed to the data source. Using a left anti join in the merge step checks each row in Sales against ExceptionRecords and keeps only those rows that have no match in ExceptionRecords. That precisely excludes the records that are listed as exceptions, so Sales is filtered correctly. Following that with an expand step shapes the final columns without pulling in unnecessary data, and both steps are standard, foldable operations in Dataflow Gen2, so the query stays folded and efficiently executed by the source. Choosing any of the other approaches would not achieve the same result: an inner join would keep only rows present in both tables (the opposite of what you want), a full outer join would bring in all rows from both tables and require extra steps to filter, and a left outer join would retain all Sales rows (including those in exceptions) and an add columns step wouldn’t perform the needed exclusion.

The idea being tested is filtering a table by removing any rows that appear in a secondary table using an anti join, while keeping the query folded so the work is pushed to the data source.

Using a left anti join in the merge step checks each row in Sales against ExceptionRecords and keeps only those rows that have no match in ExceptionRecords. That precisely excludes the records that are listed as exceptions, so Sales is filtered correctly. Following that with an expand step shapes the final columns without pulling in unnecessary data, and both steps are standard, foldable operations in Dataflow Gen2, so the query stays folded and efficiently executed by the source.

Choosing any of the other approaches would not achieve the same result: an inner join would keep only rows present in both tables (the opposite of what you want), a full outer join would bring in all rows from both tables and require extra steps to filter, and a left outer join would retain all Sales rows (including those in exceptions) and an add columns step wouldn’t perform the needed exclusion.