perf: improve performance of lpad/rpad by reusing buffers (#19558)

Optimized lpad and rpad functions to eliminate per-row allocations by
reusing buffers for graphemes and fill characters.

The previous implementation allocated new Vec<&str> for graphemes and
Vec<char> for fill characters on every row, which was inefficient. This
optimization introduces reusable buffers that are allocated once and
cleared/refilled for each row.

Changes:
- lpad: Added graphemes_buf and fill_chars_buf outside loops, clear and
refill per row instead of allocating new Vec each time
- rpad: Added graphemes_buf outside loops to reuse across iterations
- Both functions now allocate buffers once and reuse them for all rows
- Buffers are cleared and reused for each row via .clear() and .extend()

Optimization impact:
- For lpad with fill parameter: Eliminates 2 Vec allocations per row
(graphemes + fill_chars)
- For lpad without fill: Eliminates 1 Vec allocation per row (graphemes)
- For rpad: Eliminates 1 Vec allocation per row (graphemes)

This optimization is particularly effective for:
- Large arrays with many rows
- Strings with multiple graphemes (unicode characters)
- Workloads with custom fill patterns

Benchmark results comparing main vs optimized branch:

lpad benchmarks:
- size=1024, str_len=5, target=20: 116.53 µs -> 63.226 µs (45.7% faster)
- size=1024, str_len=20, target=50: 314.07 µs -> 190.30 µs (39.4%
faster)
- size=4096, str_len=5, target=20: 467.35 µs -> 261.29 µs (44.1% faster)
- size=4096, str_len=20, target=50: 1.2286 ms -> 754.24 µs (38.6%
faster)

rpad benchmarks:
- size=1024, str_len=5, target=20: 113.89 µs -> 72.645 µs (36.2% faster)
- size=1024, str_len=20, target=50: 313.68 µs -> 202.98 µs (35.3%
faster)
- size=4096, str_len=5, target=20: 456.08 µs -> 295.57 µs (35.2% faster)
- size=4096, str_len=20, target=50: 1.2523 ms -> 818.47 µs (34.6%
faster)

Overall improvements: 35-46% faster across all workloads

## 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. For example
`Closes #123` indicates that this PR will close issue #123.
-->

- Closes #.

## 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.
-->

## 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 add the `api
change` label.
-->

Author: viirya

datafusion/functions/benches/pad.rs

@@ -15,20 +15,22 @@
 // specific language governing permissions and limitations
 // under the License.
 
-use arrow::array::{ArrayRef, ArrowPrimitiveType, OffsetSizeTrait, PrimitiveArray};
+extern crate criterion;
+
+use arrow::array::{ArrowPrimitiveType, OffsetSizeTrait, PrimitiveArray};
 use arrow::datatypes::{DataType, Field, Int64Type};
 use arrow::util::bench_util::{
     create_string_array_with_len, create_string_view_array_with_len,
 };
-use criterion::{BenchmarkId, Criterion, criterion_group, criterion_main};
-use datafusion_common::DataFusionError;
+use criterion::{Criterion, SamplingMode, criterion_group, criterion_main};
 use datafusion_common::config::ConfigOptions;
 use datafusion_expr::{ColumnarValue, ScalarFunctionArgs};
-use datafusion_functions::unicode::{lpad, rpad};
+use datafusion_functions::unicode;
 use rand::Rng;
 use rand::distr::{Distribution, Uniform};
 use std::hint::black_box;
 use std::sync::Arc;
+use std::time::Duration;
 
 struct Filter<Dist> {
     dist: Dist,
@@ -67,104 +69,260 @@ where
         .collect()
 }
 
-fn create_args<O: OffsetSizeTrait>(
+/// Create args for pad benchmark
+fn create_pad_args<O: OffsetSizeTrait>(
     size: usize,
     str_len: usize,
-    force_view_types: bool,
+    target_len: usize,
+    use_string_view: bool,
 ) -> Vec<ColumnarValue> {
-    let length_array = Arc::new(create_primitive_array::<Int64Type>(size, 0.0, str_len));
-
-    if !force_view_types {
-        let string_array =
-            Arc::new(create_string_array_with_len::<O>(size, 0.1, str_len));
-        let fill_array = Arc::new(create_string_array_with_len::<O>(size, 0.1, str_len));
+    let length_array =
+        Arc::new(create_primitive_array::<Int64Type>(size, 0.0, target_len));
 
+    if use_string_view {
+        let string_array = create_string_view_array_with_len(size, 0.1, str_len, false);
+        let fill_array = create_string_view_array_with_len(size, 0.1, str_len, false);
         vec![
-            ColumnarValue::Array(string_array),
-            ColumnarValue::Array(Arc::clone(&length_array) as ArrayRef),
-            ColumnarValue::Array(fill_array),
+            ColumnarValue::Array(Arc::new(string_array)),
+            ColumnarValue::Array(length_array),
+            ColumnarValue::Array(Arc::new(fill_array)),
         ]
     } else {
-        let string_array =
-            Arc::new(create_string_view_array_with_len(size, 0.1, str_len, false));
-        let fill_array =
-            Arc::new(create_string_view_array_with_len(size, 0.1, str_len, false));
-
+        let string_array = create_string_array_with_len::<O>(size, 0.1, str_len);
+        let fill_array = create_string_array_with_len::<O>(size, 0.1, str_len);
         vec![
-            ColumnarValue::Array(string_array),
-            ColumnarValue::Array(Arc::clone(&length_array) as ArrayRef),
-            ColumnarValue::Array(fill_array),
+            ColumnarValue::Array(Arc::new(string_array)),
+            ColumnarValue::Array(length_array),
+            ColumnarValue::Array(Arc::new(fill_array)),
         ]
     }
 }
 
-#[expect(clippy::needless_pass_by_value)]
-fn invoke_pad_with_args(
-    args: Vec<ColumnarValue>,
-    number_rows: usize,
-    left_pad: bool,
-) -> Result<ColumnarValue, DataFusionError> {
-    let arg_fields = args
-        .iter()
-        .enumerate()
-        .map(|(idx, arg)| Field::new(format!("arg_{idx}"), arg.data_type(), true).into())
-        .collect::<Vec<_>>();
-    let config_options = Arc::new(ConfigOptions::default());
-
-    let scalar_args = ScalarFunctionArgs {
-        args: args.clone(),
-        arg_fields,
-        number_rows,
-        return_field: Field::new("f", DataType::Utf8, true).into(),
-        config_options: Arc::clone(&config_options),
-    };
+fn criterion_benchmark(c: &mut Criterion) {
+    for size in [1024, 4096] {
+        let mut group = c.benchmark_group(format!("lpad size={size}"));
+        group.sampling_mode(SamplingMode::Flat);
+        group.sample_size(10);
+        group.measurement_time(Duration::from_secs(10));
 
-    if left_pad {
-        lpad().invoke_with_args(scalar_args)
-    } else {
-        rpad().invoke_with_args(scalar_args)
-    }
-}
+        // Utf8 type
+        let args = create_pad_args::<i32>(size, 5, 20, false);
+        let arg_fields = args
+            .iter()
+            .enumerate()
+            .map(|(idx, arg)| {
+                Field::new(format!("arg_{idx}"), arg.data_type(), true).into()
+            })
+            .collect::<Vec<_>>();
+        let config_options = Arc::new(ConfigOptions::default());
 
-fn criterion_benchmark(c: &mut Criterion) {
-    for size in [1024, 2048] {
-        let mut group = c.benchmark_group("lpad function");
+        group.bench_function(
+            format!("lpad utf8 [size={size}, str_len=5, target=20]"),
+            |b| {
+                b.iter(|| {
+                    let args_cloned = args.clone();
+                    black_box(unicode::lpad().invoke_with_args(ScalarFunctionArgs {
+                        args: args_cloned,
+                        arg_fields: arg_fields.clone(),
+                        number_rows: size,
+                        return_field: Field::new("f", DataType::Utf8, true).into(),
+                        config_options: Arc::clone(&config_options),
+                    }))
+                })
+            },
+        );
 
-        let args = create_args::<i32>(size, 32, false);
+        // StringView type
+        let args = create_pad_args::<i32>(size, 5, 20, true);
+        let arg_fields = args
+            .iter()
+            .enumerate()
+            .map(|(idx, arg)| {
+                Field::new(format!("arg_{idx}"), arg.data_type(), true).into()
+            })
+            .collect::<Vec<_>>();
 
-        group.bench_function(BenchmarkId::new("utf8 type", size), |b| {
-            b.iter(|| black_box(invoke_pad_with_args(args.clone(), size, true).unwrap()))
-        });
+        group.bench_function(
+            format!("lpad stringview [size={size}, str_len=5, target=20]"),
+            |b| {
+                b.iter(|| {
+                    let args_cloned = args.clone();
+                    black_box(unicode::lpad().invoke_with_args(ScalarFunctionArgs {
+                        args: args_cloned,
+                        arg_fields: arg_fields.clone(),
+                        number_rows: size,
+                        return_field: Field::new("f", DataType::Utf8View, true).into(),
+                        config_options: Arc::clone(&config_options),
+                    }))
+                })
+            },
+        );
 
-        let args = create_args::<i64>(size, 32, false);
-        group.bench_function(BenchmarkId::new("largeutf8 type", size), |b| {
-            b.iter(|| black_box(invoke_pad_with_args(args.clone(), size, true).unwrap()))
-        });
+        // Utf8 type with longer strings
+        let args = create_pad_args::<i32>(size, 20, 50, false);
+        let arg_fields = args
+            .iter()
+            .enumerate()
+            .map(|(idx, arg)| {
+                Field::new(format!("arg_{idx}"), arg.data_type(), true).into()
+            })
+            .collect::<Vec<_>>();
 
-        let args = create_args::<i32>(size, 32, true);
-        group.bench_function(BenchmarkId::new("stringview type", size), |b| {
-            b.iter(|| black_box(invoke_pad_with_args(args.clone(), size, true).unwrap()))
-        });
+        group.bench_function(
+            format!("lpad utf8 [size={size}, str_len=20, target=50]"),
+            |b| {
+                b.iter(|| {
+                    let args_cloned = args.clone();
+                    black_box(unicode::lpad().invoke_with_args(ScalarFunctionArgs {
+                        args: args_cloned,
+                        arg_fields: arg_fields.clone(),
+                        number_rows: size,
+                        return_field: Field::new("f", DataType::Utf8, true).into(),
+                        config_options: Arc::clone(&config_options),
+                    }))
+                })
+            },
+        );
+
+        // StringView type with longer strings
+        let args = create_pad_args::<i32>(size, 20, 50, true);
+        let arg_fields = args
+            .iter()
+            .enumerate()
+            .map(|(idx, arg)| {
+                Field::new(format!("arg_{idx}"), arg.data_type(), true).into()
+            })
+            .collect::<Vec<_>>();
+
+        group.bench_function(
+            format!("lpad stringview [size={size}, str_len=20, target=50]"),
+            |b| {
+                b.iter(|| {
+                    let args_cloned = args.clone();
+                    black_box(unicode::lpad().invoke_with_args(ScalarFunctionArgs {
+                        args: args_cloned,
+                        arg_fields: arg_fields.clone(),
+                        number_rows: size,
+                        return_field: Field::new("f", DataType::Utf8View, true).into(),
+                        config_options: Arc::clone(&config_options),
+                    }))
+                })
+            },
+        );
 
         group.finish();
+    }
+
+    for size in [1024, 4096] {
+        let mut group = c.benchmark_group(format!("rpad size={size}"));
+        group.sampling_mode(SamplingMode::Flat);
+        group.sample_size(10);
+        group.measurement_time(Duration::from_secs(10));
+
+        // Utf8 type
+        let args = create_pad_args::<i32>(size, 5, 20, false);
+        let arg_fields = args
+            .iter()
+            .enumerate()
+            .map(|(idx, arg)| {
+                Field::new(format!("arg_{idx}"), arg.data_type(), true).into()
+            })
+            .collect::<Vec<_>>();
+        let config_options = Arc::new(ConfigOptions::default());
+
+        group.bench_function(
+            format!("rpad utf8 [size={size}, str_len=5, target=20]"),
+            |b| {
+                b.iter(|| {
+                    let args_cloned = args.clone();
+                    black_box(unicode::rpad().invoke_with_args(ScalarFunctionArgs {
+                        args: args_cloned,
+                        arg_fields: arg_fields.clone(),
+                        number_rows: size,
+                        return_field: Field::new("f", DataType::Utf8, true).into(),
+                        config_options: Arc::clone(&config_options),
+                    }))
+                })
+            },
+        );
+
+        // StringView type
+        let args = create_pad_args::<i32>(size, 5, 20, true);
+        let arg_fields = args
+            .iter()
+            .enumerate()
+            .map(|(idx, arg)| {
+                Field::new(format!("arg_{idx}"), arg.data_type(), true).into()
+            })
+            .collect::<Vec<_>>();
+
+        group.bench_function(
+            format!("rpad stringview [size={size}, str_len=5, target=20]"),
+            |b| {
+                b.iter(|| {
+                    let args_cloned = args.clone();
+                    black_box(unicode::rpad().invoke_with_args(ScalarFunctionArgs {
+                        args: args_cloned,
+                        arg_fields: arg_fields.clone(),
+                        number_rows: size,
+                        return_field: Field::new("f", DataType::Utf8View, true).into(),
+                        config_options: Arc::clone(&config_options),
+                    }))
+                })
+            },
+        );
 
-        let mut group = c.benchmark_group("rpad function");
+        // Utf8 type with longer strings
+        let args = create_pad_args::<i32>(size, 20, 50, false);
+        let arg_fields = args
+            .iter()
+            .enumerate()
+            .map(|(idx, arg)| {
+                Field::new(format!("arg_{idx}"), arg.data_type(), true).into()
+            })
+            .collect::<Vec<_>>();
 
-        let args = create_args::<i32>(size, 32, false);
-        group.bench_function(BenchmarkId::new("utf8 type", size), |b| {
-            b.iter(|| black_box(invoke_pad_with_args(args.clone(), size, false).unwrap()))
-        });
+        group.bench_function(
+            format!("rpad utf8 [size={size}, str_len=20, target=50]"),
+            |b| {
+                b.iter(|| {
+                    let args_cloned = args.clone();
+                    black_box(unicode::rpad().invoke_with_args(ScalarFunctionArgs {
+                        args: args_cloned,
+                        arg_fields: arg_fields.clone(),
+                        number_rows: size,
+                        return_field: Field::new("f", DataType::Utf8, true).into(),
+                        config_options: Arc::clone(&config_options),
+                    }))
+                })
+            },
+        );
 
-        let args = create_args::<i64>(size, 32, false);
-        group.bench_function(BenchmarkId::new("largeutf8 type", size), |b| {
-            b.iter(|| black_box(invoke_pad_with_args(args.clone(), size, false).unwrap()))
-        });
+        // StringView type with longer strings
+        let args = create_pad_args::<i32>(size, 20, 50, true);
+        let arg_fields = args
+            .iter()
+            .enumerate()
+            .map(|(idx, arg)| {
+                Field::new(format!("arg_{idx}"), arg.data_type(), true).into()
+            })
+            .collect::<Vec<_>>();
 
-        // rpad for stringview type
-        let args = create_args::<i32>(size, 32, true);
-        group.bench_function(BenchmarkId::new("stringview type", size), |b| {
-            b.iter(|| black_box(invoke_pad_with_args(args.clone(), size, false).unwrap()))
-        });
+        group.bench_function(
+            format!("rpad stringview [size={size}, str_len=20, target=50]"),
+            |b| {
+                b.iter(|| {
+                    let args_cloned = args.clone();
+                    black_box(unicode::rpad().invoke_with_args(ScalarFunctionArgs {
+                        args: args_cloned,
+                        arg_fields: arg_fields.clone(),
+                        number_rows: size,
+                        return_field: Field::new("f", DataType::Utf8View, true).into(),
+                        config_options: Arc::clone(&config_options),
+                    }))
+                })
+            },
+        );
 
         group.finish();
     }