Select Stage Row Distinct
Determines if the select-to-stage tables apply a row number function based on the defined key, or use a distinct based on the full row including the HASH.
When disabled, the generated select-to-stage query applies a ROW_NUMBER() function partitioned by the primary key (and the effective date when history is persisted), ordered by the effective date, and keeps the first row of each partition. When enabled, the query uses SELECT DISTINCT over the full row (including the row hash) instead of the row-number selection.
When to Change This Setting
Keep the default N for the standard key-based row selection. Enable it when the source can deliver full-row duplicates that should be reduced by distinctness rather than by key-based row numbering.
Related Settings
| Setting | Interaction |
|---|---|
| Select Blob Row Order By | The order direction of the row-number selection used when this setting is disabled |
| Delta Collapse Rows | Row collapsing only applies with the row-number selection; enabling distinct selection bypasses it |
| Delta Is Late Arriving | With late-arriving deltas, enabling distinct selection adds a row-hash comparison to the delta join |
Notes
- This setting is part of the Staging settings category.
- The default value for this setting is
N.