You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@arrow.apache.org by gi...@apache.org on 2026/10/15 03:42:21 UTC
(arrow-rs) branch gh-readonly-queue/main/pr-11546-7db9b69d29258d2a5cd545db819647b1cd210fac created (now 2e26ab4b0b)
This is an automated email from the ASF dual-hosted git repository.
github-merge-queue[bot] pushed a change to branch gh-readonly-queue/main/pr-11546-7db9b69d29258d2a5cd545db819647b1cd210fac
in repository https://gitbox.apache.org/repos/asf/arrow-rs.git
at 2e26ab4b0b fix: account for offsets in runarrays when using extend for MutableArrayData (#11546)
This branch includes the following new commits:
new 2e26ab4b0b fix: account for offsets in runarrays when using extend for MutableArrayData (#11546)
The 1 revisions listed above as "new" are entirely new to this
repository and will be described in separate emails. The revisions
listed as "add" were already present in the repository and have only
been added to this reference.
(arrow-rs) 01/01: fix: account for offsets in runarrays when using extend for MutableArrayData (#11546)
Posted by gi...@apache.org.
This is an automated email from the ASF dual-hosted git repository.
github-merge-queue[bot] pushed a commit to branch gh-readonly-queue/main/pr-11546-7db9b69d29258d2a5cd545db819647b1cd210fac
in repository https://gitbox.apache.org/repos/asf/arrow-rs.git
commit 2e26ab4b0baa351ca59945a1988685b974861e63
Author: Jeffrey Vo <je...@gmail.com>
AuthorDate: Thu Oct 15 03:41:57 2026 +0000
fix: account for offsets in runarrays when using extend for MutableArrayData (#11546)
# Which issue does this PR close?
<!--
It is not necessary to raise an issue, but it is encouraged if it is a
non-trivial
enhancement/bug.
If this PR relates to an existing issue, but does not fully resolve it,
please change
the wording below from "Closes #NNN" to "Relates to #NNN" or "Part of
#NNN" to avoid
false closing an issue.
-->
None
# Rationale for this change
<!--
Why are you proposing this change? If this is already explained clearly
in the
issue then this section is not needed. Explaining clearly why changes
are proposed
helps reviewers understand your changes and offer better suggestions for
fixes.
Please try to explain the problem you are trying to solve in terms of
the user-visible
behavior, rather than the implementation.
-->
Correctness edge case when using RunArrays in MutableArrayData
# What changes are included in this PR?
<!--
There is no need to duplicate the description in the issue here but it
is
sometimes worth providing a summary of the individual changes in this
PR.
-->
- Account for `ArrayData` offset of RunArrays when extending
`MutableArrayData`
- Checked conversion of usize length to Run index type to avoid wrapping
# Are these changes tested?
<!--
We typically require tests for all PRs in order to:
1. Prevent the code from being accidentally broken by subsequent changes
2. Serve as another way to document the expected behavior of the code
If tests are not included in your PR, please explain why (for example,
are they
covered by existing tests, if so identify these tests)?
If this PR claims a performance improvement, please include evidence
such as
benchmark results.
DO NOT simply paste a list of commands ran to verify the code; this is
what our
CI is for and it pollutes the PR. ONLY explain if testing was included
with this
PR, or why it wasn't.
-->
Yes
# Are there any significant user-facing changes?
<!--
Significant user-facing changes include:
- Breaking public API change
- Significant change to existing functionality (e.g. function returns
something different)
These need to be explicitly called out to ensure we have proper
documentation if
required and to ensure we follow proper semantic versioning.
DO NOT restate the changes of this PR; if a bug was fixed and already
explained
above, we don't need this to be restated as a user-facing change.
-->
No
---
arrow-data/src/transform/run.rs | 72 +++++++++++++++++++++++++++++++++++------
1 file changed, 63 insertions(+), 9 deletions(-)
diff --git a/arrow-data/src/transform/run.rs b/arrow-data/src/transform/run.rs
index cb239b9a2c..a8e2f4e150 100644
--- a/arrow-data/src/transform/run.rs
+++ b/arrow-data/src/transform/run.rs
@@ -16,7 +16,7 @@
// under the License.
use super::{_MutableArrayData, ArrayData, Extend};
-use arrow_buffer::{ArrowNativeType, Buffer, ToByteSlice};
+use arrow_buffer::{ArrowNativeType, ToByteSlice};
use arrow_schema::{ArrowError, DataType};
use num_traits::CheckedAdd;
@@ -58,8 +58,9 @@ pub fn extend_nulls(mutable: &mut _MutableArrayData, len: usize) -> Result<(), A
macro_rules! extend_nulls_impl {
($run_end_type:ty) => {{
let last_run_end = get_last_run_end::<$run_end_type>(&mutable.child_data[0]);
- let new_value = last_run_end
- .checked_add(<$run_end_type as ArrowNativeType>::usize_as(len))
+ let len = <$run_end_type as ArrowNativeType>::from_usize(len);
+ let new_value = len
+ .and_then(|len| last_run_end.checked_add(len))
.ok_or_else(|| {
ArrowError::InvalidArgumentError(
"run end overflow when extending RunEndEncoded array: \
@@ -92,7 +93,7 @@ type ExtendArrays = (Vec<u8>, Option<(usize, usize)>);
/// Build run ends bytes and values range directly for batch processing
fn build_extend_arrays<T: ArrowNativeType + std::ops::Add<Output = T> + CheckedAdd>(
- buffer: &Buffer,
+ typed_slice: &[T],
length: usize,
start: usize,
len: usize,
@@ -104,9 +105,6 @@ fn build_extend_arrays<T: ArrowNativeType + std::ops::Add<Output = T> + CheckedA
let mut prev_end = 0;
let mut current_run_end = dest_last_run_end;
- // Convert buffer to typed slice once
- let typed_slice: &[T] = buffer.typed_data();
-
for i in 0..length {
if i < typed_slice.len() {
let run_end = typed_slice[i].to_usize().unwrap();
@@ -217,6 +215,8 @@ pub fn build_extend(array: &ArrayData) -> Extend<'_> {
// We need to analyze the source array's run structure
let source_run_ends = &array.child_data()[0];
+ let source_buffer_offset = source_run_ends.offset();
+ let source_buffer_len = source_run_ends.len();
let source_buffer = &source_run_ends.buffers()[0];
// Get the run end type from the mutable array
@@ -230,6 +230,8 @@ pub fn build_extend(array: &ArrayData) -> Extend<'_> {
// Build run ends and values indices directly for batch processing
macro_rules! build_and_process_impl {
($run_end_type:ty) => {{
+ let source_buffer = &source_buffer.typed_data()
+ [source_buffer_offset..source_buffer_offset + source_buffer_len];
let dest_last_run_end =
get_last_run_end::<$run_end_type>(&mutable.child_data[0]);
let (run_ends_bytes, values_range) = build_extend_arrays::<$run_end_type>(
@@ -566,7 +568,7 @@ mod tests {
fn test_build_extend_arrays_int16() {
let buffer = Buffer::from_vec(vec![3i16, 5i16, 8i16]);
let (run_ends_bytes, values_range) =
- build_extend_arrays::<i16>(&buffer, 3, 2, 4, 0i16).unwrap();
+ build_extend_arrays::<i16>(buffer.typed_data(), 3, 2, 4, 0i16).unwrap();
// Logical array: [A, A, A, B, B, C, C, C]
// Requesting indices 2-6 should give us:
@@ -589,7 +591,7 @@ mod tests {
fn test_build_extend_arrays_int64() {
let buffer = Buffer::from_vec(vec![3i64, 5i64, 8i64]);
let (run_ends_bytes, values_range) =
- build_extend_arrays::<i64>(&buffer, 3, 2, 4, 0i64).unwrap();
+ build_extend_arrays::<i64>(buffer.typed_data(), 3, 2, 4, 0i64).unwrap();
// Same logic as above but with i64
assert_eq!(run_ends_bytes.len(), 3 * std::mem::size_of::<i64>());
@@ -712,4 +714,56 @@ mod tests {
"unexpected error: {err}"
);
}
+
+ #[test]
+ fn test_extend_nulls_overflow() {
+ let values = create_int32_array_data(vec![10]);
+ let array = create_run_array_data_int16(vec![100], values);
+
+ let mut mutable = MutableArrayData::new(vec![&array], true, 10);
+ let err = mutable.try_extend_nulls(i16::MAX as usize + 1).unwrap_err();
+ assert_eq!(
+ err.to_string(),
+ "Invalid argument error: run end overflow when extending RunEndEncoded array: use a larger run-end type (e.g. Int64 instead of Int32)"
+ );
+ }
+
+ #[test]
+ fn test_extend_accounts_for_offset() {
+ let dt = DataType::RunEndEncoded(
+ Arc::new(Field::new(
+ Field::REE_RUN_ENDS_FIELD_DEFAULT_NAME,
+ DataType::Int32,
+ false,
+ )),
+ Arc::new(Field::new(
+ Field::REE_VALUES_FIELD_DEFAULT_NAME,
+ DataType::Float32,
+ true,
+ )),
+ );
+
+ let runs = ArrayData::builder(DataType::Int32)
+ .len(2)
+ .offset(2)
+ .add_buffer(vec![-10_i32, -20, 1_i32, 4, 0, 100].into())
+ .build()
+ .unwrap();
+ let values = ArrayData::builder(DataType::Float32)
+ .len(2)
+ .add_buffer(vec![1.5, 2.5].into())
+ .build()
+ .unwrap();
+
+ let source_data = ArrayDataBuilder::new(dt)
+ .len(3)
+ .child_data(vec![runs, values])
+ .build()
+ .unwrap();
+
+ let mut mutable = MutableArrayData::new(vec![&source_data], true, 10);
+ mutable.try_extend(0, 0, 1).unwrap();
+ let data = mutable.freeze();
+ assert_eq!(data, source_data.slice(0, 1));
+ }
}