lidavidm commented on a change in pull request #10987:
URL: https://github.com/apache/arrow/pull/10987#discussion_r695973782
##########
File path: cpp/src/arrow/compute/kernels/scalar_if_else.cc
##########
@@ -1827,9 +1827,131 @@ Status ExecArrayCoalesce(KernelContext* ctx, const
ExecBatch& batch, Datum* out)
return Status::OK();
}
+// Special case: implement 'coalesce' for an array and a scalar for any
+// fixed-width type (a 'fill_null' operation)
+template <typename Type>
+Status ExecArrayScalarCoalesce(KernelContext* ctx, Datum left, Datum right,
+ int64_t length, Datum* out) {
+ ArrayData* output = out->mutable_array();
+ const int64_t out_offset = output->offset;
+ uint8_t* out_valid = output->buffers[0]->mutable_data();
+ uint8_t* out_values = output->buffers[1]->mutable_data();
+
+ const ArrayData& left_arr = *left.array();
+ const uint8_t* left_valid =
+ left_arr.MayHaveNulls() ? left_arr.buffers[0]->data() : nullptr;
+ arrow::internal::OptionalBitBlockCounter bit_counter(left_valid,
left_arr.offset,
Review comment:
In this case it does better, though I could imagine it doing badly at
~50% nulls.
```
--------------------------------------------------------------------------------------
Benchmark Time CPU Iterations
UserCounters...
--------------------------------------------------------------------------------------
Before:
CoalesceScalarBench64/0 1321249 ns 1321175 ns 508
bytes_per_second=5.9133G/s items_per_second=793.67M/s length=1048.58k null%=1
num_args=2
CoalesceScalarBench64/2 5049228 ns 5049073 ns 133
bytes_per_second=1.54731G/s items_per_second=207.677M/s length=1048.58k
null%=25 num_args=2
CoalesceScalarStringBench/0 87407512 ns 87404678 ns 7
bytes_per_second=5.70364G/s items_per_second=11.9968M/s length=1048.58k null%=1
num_args=2
CoalesceScalarStringBench/2 77686321 ns 77685936 ns 8
bytes_per_second=4.87622G/s items_per_second=13.4976M/s length=1048.58k
null%=25 num_args=2
After:
CoalesceScalarBench64/0 988642 ns 988606 ns 664
bytes_per_second=7.90255G/s items_per_second=1060.66M/s length=1048.58k null%=1
num_args=2
CoalesceScalarBench64/2 4537487 ns 4537432 ns 154
bytes_per_second=1.72179G/s items_per_second=231.095M/s length=1048.58k
null%=25 num_args=2
CoalesceScalarStringBench/0 103582667 ns 103575544 ns 6
bytes_per_second=4.81315G/s items_per_second=10.1238M/s length=1048.58k null%=1
num_args=2
CoalesceScalarStringBench/2 79813292 ns 79807705 ns 8
bytes_per_second=4.74658G/s items_per_second=13.1388M/s length=1048.58k
null%=25 num_args=2
```
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
