Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
195 changes: 195 additions & 0 deletions datafusion/functions/benches/pad.rs
Original file line number Diff line number Diff line change
Expand Up @@ -24,6 +24,7 @@ use arrow::util::bench_util::{
create_string_array_with_len, create_string_view_array_with_len,
};
use criterion::{Criterion, SamplingMode, criterion_group, criterion_main};
use datafusion_common::ScalarValue;
use datafusion_common::config::ConfigOptions;
use datafusion_expr::{ColumnarValue, ScalarFunctionArgs};
use datafusion_functions::unicode;
Expand Down Expand Up @@ -172,6 +173,32 @@ fn create_pad_args<O: OffsetSizeTrait>(
}
}

/// Create args for pad benchmark with scalar length and fill (common pattern:
/// `lpad(column, 20, '0')`).
fn create_scalar_pad_args<O: OffsetSizeTrait>(
size: usize,
str_len: usize,
target_len: i64,
fill: &str,
use_string_view: bool,
) -> Vec<ColumnarValue> {
if use_string_view {
let string_array = create_string_view_array_with_len(size, 0.1, str_len, false);
vec![
ColumnarValue::Array(Arc::new(string_array)),
ColumnarValue::Scalar(ScalarValue::Int64(Some(target_len))),
ColumnarValue::Scalar(ScalarValue::Utf8(Some(fill.to_string()))),
]
} else {
let string_array = create_string_array_with_len::<O>(size, 0.1, str_len);
vec![
ColumnarValue::Array(Arc::new(string_array)),
ColumnarValue::Scalar(ScalarValue::Int64(Some(target_len))),
ColumnarValue::Scalar(ScalarValue::Utf8(Some(fill.to_string()))),
]
}
}

fn criterion_benchmark(c: &mut Criterion) {
for size in [1024, 4096] {
let mut group = c.benchmark_group(format!("lpad size={size}"));
Expand Down Expand Up @@ -336,6 +363,90 @@ fn criterion_benchmark(c: &mut Criterion) {
},
);

// --- Scalar length + fill benchmarks ---

// Utf8 with scalar length and fill (3-arg)
let args = create_scalar_pad_args::<i32>(size, 5, 20, "x", false);
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 utf8 scalar [size={size}, str_len=5, target=20, fill='x']"),
|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 with scalar length and fill (3-arg)
let args = create_scalar_pad_args::<i32>(size, 5, 20, "x", 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 scalar [size={size}, str_len=5, target=20, fill='x']"
),
|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),
}))
})
},
);

// Utf8 with scalar length and unicode fill
let args = create_scalar_pad_args::<i32>(size, 5, 20, "é", false);
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 utf8 scalar unicode [size={size}, str_len=5, target=20, fill='é']"
),
|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),
}))
})
},
);

group.finish();
}

Expand Down Expand Up @@ -502,6 +613,90 @@ fn criterion_benchmark(c: &mut Criterion) {
},
);

// --- Scalar length + fill benchmarks ---

// Utf8 with scalar length and fill (3-arg)
let args = create_scalar_pad_args::<i32>(size, 5, 20, "x", false);
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 utf8 scalar [size={size}, str_len=5, target=20, fill='x']"),
|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 with scalar length and fill (3-arg)
let args = create_scalar_pad_args::<i32>(size, 5, 20, "x", 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 scalar [size={size}, str_len=5, target=20, fill='x']"
),
|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),
}))
})
},
);

// Utf8 with scalar length and unicode fill
let args = create_scalar_pad_args::<i32>(size, 5, 20, "é", false);
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 utf8 scalar unicode [size={size}, str_len=5, target=20, fill='é']"
),
|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),
}))
})
},
);

group.finish();
}
}
Expand Down
18 changes: 18 additions & 0 deletions datafusion/functions/src/unicode/common.rs
Original file line number Diff line number Diff line change
Expand Up @@ -23,14 +23,32 @@ use arrow::array::{
};
use arrow::datatypes::DataType;
use arrow_buffer::{NullBuffer, ScalarBuffer};
use datafusion_common::ScalarValue;
use datafusion_common::cast::{
as_generic_string_array, as_int64_array, as_string_view_array,
};
use datafusion_common::exec_err;
use datafusion_expr::ColumnarValue;
use std::cmp::Ordering;
use std::ops::Range;
use std::sync::Arc;

/// If `cv` is a non-null scalar string, return its value.
pub(crate) fn try_as_scalar_str(cv: &ColumnarValue) -> Option<&str> {
match cv {
ColumnarValue::Scalar(s) => s.try_as_str().flatten(),
_ => None,
}
}

/// If `cv` is a non-null scalar Int64, return its value.
pub(crate) fn try_as_scalar_i64(cv: &ColumnarValue) -> Option<i64> {
match cv {
ColumnarValue::Scalar(ScalarValue::Int64(v)) => *v,
_ => None,
}
}

/// A trait for `left` and `right` byte slicing operations
pub(crate) trait LeftRightSlicer {
fn slice(string: &str, n: i64) -> Range<usize>;
Expand Down
Loading