This is an automated email from the ASF dual-hosted git repository.
zeroshade pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/arrow-go.git
The following commit(s) were added to refs/heads/main by this push:
new 9e89349a fix(compute/exec): report ListView buffer counts (#1133)
9e89349a is described below
commit 9e89349a8501a7d498c4294f13387f0b8c647625
Author: Minh Vu <[email protected]>
AuthorDate: Wed Aug 12 19:26:02 2026 +0200
fix(compute/exec): report ListView buffer counts (#1133)
### Rationale for this change
`ListView` and `LargeListView` arrays have exactly three buffers:
validity, offsets, and sizes. `ArraySpan.NumBuffers()` currently returns
the default count of two, which leaves the size buffer unset for
zero-length spans and truncates it when any ListView span is converted
back to `array.Data` with `MakeData()`.
### What changes are included in this PR?
Return three buffers for ListView and LargeListView types, document why
that count is exact, and cover both the direct buffer-count contract and
zero-length buffer allocation.
### Are these changes tested?
- `go test ./arrow/compute/exec ./arrow/compute`
### Are there any user-facing changes?
There are no API changes. Compute kernels do not currently accept list
views, so this is preparatory hardening rather than a live compute
corruption fix.
---
arrow/compute/exec/span.go | 4 ++++
arrow/compute/exec/span_test.go | 2 ++
arrow/compute/exec_internals_test.go | 19 +++++++++++++++++++
3 files changed, 25 insertions(+)
diff --git a/arrow/compute/exec/span.go b/arrow/compute/exec/span.go
index d188d39b..c989cfe7 100644
--- a/arrow/compute/exec/span.go
+++ b/arrow/compute/exec/span.go
@@ -590,6 +590,10 @@ func getNumBuffers(dt arrow.DataType) int {
// within a single block (the default 32KB allocation in the
// builder is sufficient for most use cases).
return 3
+ case arrow.LIST_VIEW, arrow.LARGE_LIST_VIEW:
+ // validity + offsets + sizes. Unlike the view types above, this
+ // count is exact rather than a cap on variadic data buffers.
+ return 3
case arrow.EXTENSION:
return getNumBuffers(dt.(arrow.ExtensionType).StorageType())
default:
diff --git a/arrow/compute/exec/span_test.go b/arrow/compute/exec/span_test.go
index 3bdcc286..feff91e3 100644
--- a/arrow/compute/exec/span_test.go
+++ b/arrow/compute/exec/span_test.go
@@ -209,6 +209,8 @@ func TestArraySpan_NumBuffers(t *testing.T) {
{"large binary", fields{Type: arrow.BinaryTypes.LargeBinary},
3},
{"string", fields{Type: arrow.BinaryTypes.String}, 3},
{"large string", fields{Type: arrow.BinaryTypes.LargeString},
3},
+ {"list view", fields{Type:
arrow.ListViewOf(arrow.PrimitiveTypes.Int32)}, 3},
+ {"large list view", fields{Type:
arrow.LargeListViewOf(arrow.PrimitiveTypes.Int32)}, 3},
{"extension", fields{Type: extensions.NewUUIDType()}, 2},
{"int32", fields{Type: arrow.PrimitiveTypes.Int32}, 2},
}
diff --git a/arrow/compute/exec_internals_test.go
b/arrow/compute/exec_internals_test.go
index f813fb1f..9def1be9 100644
--- a/arrow/compute/exec_internals_test.go
+++ b/arrow/compute/exec_internals_test.go
@@ -31,6 +31,7 @@ import (
"github.com/apache/arrow-go/v18/arrow/internal/testing/gen"
"github.com/apache/arrow-go/v18/arrow/memory"
"github.com/apache/arrow-go/v18/arrow/scalar"
+ "github.com/stretchr/testify/assert"
"github.com/stretchr/testify/suite"
)
@@ -580,6 +581,24 @@ func (e *ExecSpanItrSuite) TestZeroLengthInput() {
checkArgs(input)
}
+func TestFillZeroLengthListView(t *testing.T) {
+ for _, dt := range []arrow.DataType{
+ arrow.ListViewOf(arrow.PrimitiveTypes.Int32),
+ arrow.LargeListViewOf(arrow.PrimitiveTypes.Int32),
+ } {
+ t.Run(dt.Name(), func(t *testing.T) {
+ var span exec.ArraySpan
+ exec.FillZeroLength(dt, &span)
+
+ assert.Equal(t, 3, span.NumBuffers())
+ for i := 0; i < span.NumBuffers(); i++ {
+ assert.NotNil(t, span.Buffers[i].Buf)
+ assert.Empty(t, span.Buffers[i].Buf)
+ }
+ })
+ }
+}
+
func TestExecSpanIterator(t *testing.T) {
suite.Run(t, new(ExecSpanItrSuite))
}
