This is an automated email from the ASF dual-hosted git repository.
Jefffrey pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/arrow-rs.git
The following commit(s) were added to refs/heads/main by this push:
new 2970e0f48f Replace BufferBuilder with Vec in sort_run_downcasted
(#10630)
2970e0f48f is described below
commit 2970e0f48f390d29566e89b02820a7d315079056
Author: cakeni <[email protected]>
AuthorDate: Thu Aug 20 09:23:51 2026 +0800
Replace BufferBuilder with Vec in sort_run_downcasted (#10630)
# Which issue does this PR close?
- Part of #10245.
# Rationale for this change
Using `Vec` instead of `BufferBuilder` can benefit from Rust's optimized
vector implementation. This updates the remaining `BufferBuilder` usage
in `arrow-ord/src/sort.rs`.
# What changes are included in this PR?
- Replace the run-end `BufferBuilder` in `sort_run_downcasted` with a
preallocated `Vec`.
- Push converted run ends directly into the vector and convert it into
the final Arrow buffer.
# Are these changes tested?
Yes. The following checks pass:
- `cargo +stable-x86_64-pc-windows-gnu fmt --all -- --check`
- `cargo +stable-x86_64-pc-windows-gnu clippy -p arrow-ord --all-targets
--all-features --no-deps -- -D warnings`
- `cargo +stable-x86_64-pc-windows-gnu test -p arrow-ord --all-features`
(269 unit tests and 6 doctests passed)
# Are there any user-facing changes?
No.
## AI assistance
OpenAI Codex assisted with drafting this optimization. I reviewed and
verified the final change.
Co-authored-by: Jeffrey Vo <[email protected]>
---
arrow-ord/src/sort.rs | 7 +++----
1 file changed, 3 insertions(+), 4 deletions(-)
diff --git a/arrow-ord/src/sort.rs b/arrow-ord/src/sort.rs
index b814d4079f..c2629245bf 100644
--- a/arrow-ord/src/sort.rs
+++ b/arrow-ord/src/sort.rs
@@ -18,7 +18,6 @@
//! Defines sort kernel for `ArrayRef`
use crate::ord::{DynComparator, make_comparator};
-use arrow_array::builder::BufferBuilder;
use arrow_array::cast::*;
use arrow_array::types::*;
use arrow_array::*;
@@ -722,14 +721,14 @@ fn sort_run_downcasted<R: RunEndIndexType>(
let run_ends = run_array.run_ends();
- let mut new_run_ends_builder =
BufferBuilder::<R::Native>::new(run_ends.len());
+ let mut new_run_ends = Vec::with_capacity(run_ends.len());
let mut new_run_end: usize = 0;
let mut new_physical_len: usize = 0;
let consume_runs = |run_length, _| {
new_run_end += run_length;
new_physical_len += 1;
-
new_run_ends_builder.append(R::Native::from_usize(new_run_end).unwrap());
+ new_run_ends.push(R::Native::from_usize(new_run_end).unwrap());
};
let (values_indices, run_values) = sort_run_inner(run_array, options,
output_len, consume_runs);
@@ -739,7 +738,7 @@ fn sort_run_downcasted<R: RunEndIndexType>(
// The function builds a valid run_ends array and hence need not be
validated.
ArrayDataBuilder::new(R::DATA_TYPE)
.len(new_physical_len)
- .add_buffer(new_run_ends_builder.finish())
+ .add_buffer(new_run_ends.into())
.build_unchecked()
};
