Gathers the specified rows (including null values) of a set of columns.
Gathers the rows of the source columns according to gather_map
such that row "i" in the resulting table's columns will contain row "gather_map[i]" from the source columns. The number of rows in the result table will be equal to the number of elements in gather_map
.
A negative value i
in the gather_map
is interpreted as i+n
, where n
is the number of rows in the source_table
.
For dictionary columns, the keys column component is copied and not trimmed if the gather results in abandoned key elements.
- Exceptions
-
- Parameters
-
source_table | The input columns whose rows will be gathered |
gather_map | View into a non-nullable column of integral indices that maps the rows in the source columns to rows in the destination columns. |
bounds_policy | Policy to apply to account for possible out-of-bounds indices DONT_CHECK skips all bounds checking for gather map values. NULLIFY coerces rows that corresponds to out-of-bounds indices in the gather map to be null elements. Callers should use DONT_CHECK when they are certain that the gather_map contains only valid indices for better performance. If policy is set to DONT_CHECK and there are out-of-bounds indices in the gather map, the behavior is undefined. Defaults to DONT_CHECK . |
stream | CUDA stream used for device memory operations and kernel launches |
mr | Device memory resource used to allocate the returned table's device memory |
- Returns
- Result of the gather