- PartitionColumns
- Specify the same columns as the PARTITION BY clause (in any order).
- NumberOfRows
- [Required if you omit PivotColumn.] Use NumberOfRows when no column contains pivot keys, but you can specify a maximum number of rows in any partition. The function pivots the input rows into this number of columns in the output table.
If a partition has fewer than number_of_rows rows, the function adds NULL values; if a partition has more than number_of_rows rows, the function omits the extra rows.
If you use NumberOfRows, you must use the ORDER BY clause to order the input data; otherwise, the contents of the output table columns may vary from run to run.
- PivotColumn
- [Required if you omit NumberOfRows.] Specify the name of the input column that contains the pivot keys.
If pivot_column contains numeric values, the function casts them to VARCHAR; function performance improves slightly if you specify NumericPivotKey ('true').
- PivotKeys
- [Required if you specify PivotColumn.] Specify the values in pivot_column to use as pivot keys. The function ignores rows that contain other values in pivot_column.
- DefaultPivotValues
- [Optional with PivotKeys, disallowed otherwise.] Specify exactly one default value for each pivot_key. The nth default_pivot_value applies to the nth pivot_key. The data type of each default_pivot_value must be compatible with the data type of target_column.
- NumericPivotKey
- [Optional] Use this syntax element only with the PivotColumn syntax element. If pivot_column is numeric, NumericPivotKey ('true') improves function performance slightly.
- TargetColumns
- [Optional] Specify the names of the target columns (input columns that contain the values to pivot).