This is an automated email from the ASF dual-hosted git repository.
jeffreyvo 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 294961b2d9 feat: add new `try_append_value_n()` function to
`GenericByteViewBuilder` (#9040)
294961b2d9 is described below
commit 294961b2d93b5c0691e545b2316a58afeefe799d
Author: Lanqing Yang <[email protected]>
AuthorDate: Thu Dec 25 20:50:14 2025 +0800
feat: add new `try_append_value_n()` function to `GenericByteViewBuilder`
(#9040)
# Which issue does this PR close?
<!--
We generally require a GitHub issue to be filed for all bug fixes and
enhancements and this helps us generate change logs for our releases.
You can link an issue to this PR using the GitHub syntax.
-->
- Closes #9034.
# Rationale for this change
remove boiler plate code upstream, ensure only hash once instead n
times.
# 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.
-->
# 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)?
-->
# Are there any user-facing changes?
<!--
If there are user-facing changes then we may require documentation to be
updated before approving the PR.
If there are any breaking changes to public APIs, please call them out.
-->
Signed-off-by: lyang24 <[email protected]>
---
.../src/builder/generic_bytes_view_builder.rs | 119 +++++++++++++++++++++
1 file changed, 119 insertions(+)
diff --git a/arrow-array/src/builder/generic_bytes_view_builder.rs
b/arrow-array/src/builder/generic_bytes_view_builder.rs
index d7ff7ce051..2d60187344 100644
--- a/arrow-array/src/builder/generic_bytes_view_builder.rs
+++ b/arrow-array/src/builder/generic_bytes_view_builder.rs
@@ -430,6 +430,53 @@ impl<T: ByteViewType + ?Sized> GenericByteViewBuilder<T> {
};
}
+ /// Append the same value `n` times into the builder
+ ///
+ /// This is more efficient than calling [`Self::try_append_value`] `n`
times,
+ /// especially when deduplication is enabled, as it only hashes the value
once.
+ ///
+ /// # Errors
+ ///
+ /// Returns an error if
+ /// - String buffer count exceeds `u32::MAX`
+ /// - String length exceeds `u32::MAX`
+ ///
+ /// # Example
+ /// ```
+ /// # use arrow_array::builder::StringViewBuilder;
+ /// # use arrow_array::Array;
+ /// let mut builder = StringViewBuilder::new().with_deduplicate_strings();
+ ///
+ /// // Append "hello" 1000 times efficiently
+ /// builder.try_append_value_n("hello", 1000)?;
+ ///
+ /// let array = builder.finish();
+ /// assert_eq!(array.len(), 1000);
+ ///
+ /// // All values are "hello"
+ /// for value in array.iter() {
+ /// assert_eq!(value, Some("hello"));
+ /// }
+ /// # Ok::<(), arrow_schema::ArrowError>(())
+ /// ```
+ #[inline]
+ pub fn try_append_value_n(
+ &mut self,
+ value: impl AsRef<T::Native>,
+ n: usize,
+ ) -> Result<(), ArrowError> {
+ if n == 0 {
+ return Ok(());
+ }
+ // Process value once (handles deduplication, buffer management, view
creation)
+ self.try_append_value(value)?;
+ // Reuse the view (n-1) times
+ let view = *self.views_buffer.last().unwrap();
+ self.views_buffer.extend(std::iter::repeat_n(view, n - 1));
+ self.null_buffer_builder.append_n_non_nulls(n - 1);
+ Ok(())
+ }
+
/// Append a null value into the builder
#[inline]
pub fn append_null(&mut self) {
@@ -884,4 +931,76 @@ mod tests {
MAX_BLOCK_SIZE as usize
);
}
+
+ #[test]
+ fn test_append_value_n() {
+ // Test with inline strings (<=12 bytes)
+ let mut builder = StringViewBuilder::new();
+
+ builder.try_append_value_n("hello", 100).unwrap();
+ builder.append_value("world");
+ builder.try_append_value_n("foo", 50).unwrap();
+
+ let array = builder.finish();
+ assert_eq!(array.len(), 151);
+ assert_eq!(array.null_count(), 0);
+
+ // Verify the values
+ for i in 0..100 {
+ assert_eq!(array.value(i), "hello");
+ }
+ assert_eq!(array.value(100), "world");
+ for i in 101..151 {
+ assert_eq!(array.value(i), "foo");
+ }
+
+ // All inline strings should have no data buffers
+ assert_eq!(array.data_buffers().len(), 0);
+ }
+
+ #[test]
+ fn test_append_value_n_with_deduplication() {
+ let long_string = "This is a very long string that exceeds the inline
length";
+
+ // Test with deduplication enabled
+ let mut builder = StringViewBuilder::new().with_deduplicate_strings();
+
+ // First append the string once to add it to the hash map
+ builder.append_value(long_string);
+
+ // Then append_n the same string - should deduplicate and reuse the
existing value
+ builder.try_append_value_n(long_string, 999).unwrap();
+
+ let array = builder.finish();
+ assert_eq!(array.len(), 1000);
+ assert_eq!(array.null_count(), 0);
+
+ // Verify all values are the same
+ for i in 0..1000 {
+ assert_eq!(array.value(i), long_string);
+ }
+
+ // With deduplication, should only have 1 data buffer containing the
string once
+ assert_eq!(array.data_buffers().len(), 1);
+
+ // All views should be identical
+ let first_view = array.views()[0];
+ for view in array.views().iter() {
+ assert_eq!(*view, first_view);
+ }
+ }
+
+ #[test]
+ fn test_append_value_n_zero() {
+ let mut builder = StringViewBuilder::new();
+
+ builder.append_value("first");
+ builder.try_append_value_n("should not appear", 0).unwrap();
+ builder.append_value("second");
+
+ let array = builder.finish();
+ assert_eq!(array.len(), 2);
+ assert_eq!(array.value(0), "first");
+ assert_eq!(array.value(1), "second");
+ }
}
