pyarrow.compute.case_when#

pyarrow.compute.case_when(cond, /, *cases, memory_pool=None)#

Choose values based on multiple conditions.

cond must be a struct of Boolean values. cases can be a mix of scalar and array arguments (of any type, but all must be the same type or castable to a common type), with either exactly one datum per child of cond, or one more cases than children of cond (in which case we have an โ€œelseโ€ value).

Each row of the output will be the corresponding value of the first datum in cases for which the corresponding child of cond is true, or otherwise the โ€œelseโ€ value (if given), or null.

Essentially, this implements a switch-case or if-else, if-elseโ€ฆ statement.

Parameters:
condArray-like or scalar-like

Argument to compute function.

*casesArray-like or scalar-like

Argument to compute function.

memory_poolpyarrow.MemoryPool, optional

If not passed, will allocate memory from the default memory pool.