Skip to content

migrate string functions to inovke_with_args #14722

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
alamb merged 3 commits into from
Feb 18, 2025
Merged

migrate string functions to inovke_with_args #14722

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
b19ae17
migrate string functions to inovke_with_args
zjregee Feb 18, 2025
955ddae
move clone of args in bench out of black_box
zjregee Feb 18, 2025
90fe439
modify obsolete calls in to_hex bench
zjregee Feb 18, 2025
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
15 changes: 12 additions & 3 deletions datafusion/functions/benches/concat.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,10 +16,11 @@
// under the License.

use arrow::array::ArrayRef;
use arrow::datatypes::DataType;
use arrow::util::bench_util::create_string_array_with_len;
use criterion::{criterion_group, criterion_main, BenchmarkId, Criterion};
use datafusion_common::ScalarValue;
use datafusion_expr::ColumnarValue;
use datafusion_expr::{ColumnarValue, ScalarFunctionArgs};
use datafusion_functions::string::concat;
use std::sync::Arc;

Expand All @@ -39,8 +40,16 @@ fn criterion_benchmark(c: &mut Criterion) {
let mut group = c.benchmark_group("concat function");
group.bench_function(BenchmarkId::new("concat", size), |b| {
b.iter(|| {
// TODO use invoke_with_args
criterion::black_box(concat().invoke_batch(&args, size).unwrap())
let args_cloned = args.clone();
criterion::black_box(
concat()
.invoke_with_args(ScalarFunctionArgs {
args: args_cloned,
number_rows: size,
return_type: &DataType::Utf8,
})
.unwrap(),
)
})
});
group.finish();
Expand Down
51 changes: 38 additions & 13 deletions datafusion/functions/benches/lower.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,11 +18,12 @@
extern crate criterion;

use arrow::array::{ArrayRef, StringArray, StringViewBuilder};
use arrow::datatypes::DataType;
use arrow::util::bench_util::{
create_string_array_with_len, create_string_view_array_with_len,
};
use criterion::{black_box, criterion_group, criterion_main, Criterion};
use datafusion_expr::ColumnarValue;
use datafusion_expr::{ColumnarValue, ScalarFunctionArgs};
use datafusion_functions::string;
use std::sync::Arc;

Expand Down Expand Up @@ -125,8 +126,12 @@ fn criterion_benchmark(c: &mut Criterion) {
let args = create_args1(size, 32);
c.bench_function(&format!("lower_all_values_are_ascii: {}", size), |b| {
b.iter(|| {
// TODO use invoke_with_args
black_box(lower.invoke_batch(&args, size))
let args_cloned = args.clone();
black_box(lower.invoke_with_args(ScalarFunctionArgs {
args: args_cloned,
number_rows: size,
return_type: &DataType::Utf8,
}))
})
});

Expand All @@ -135,8 +140,12 @@ fn criterion_benchmark(c: &mut Criterion) {
&format!("lower_the_first_value_is_nonascii: {}", size),
|b| {
b.iter(|| {
// TODO use invoke_with_args
black_box(lower.invoke_batch(&args, size))
let args_cloned = args.clone();
black_box(lower.invoke_with_args(ScalarFunctionArgs {
args: args_cloned,
number_rows: size,
return_type: &DataType::Utf8,
}))
})
},
);
Expand All @@ -146,8 +155,12 @@ fn criterion_benchmark(c: &mut Criterion) {
&format!("lower_the_middle_value_is_nonascii: {}", size),
|b| {
b.iter(|| {
// TODO use invoke_with_args
black_box(lower.invoke_batch(&args, size))
let args_cloned = args.clone();
black_box(lower.invoke_with_args(ScalarFunctionArgs {
args: args_cloned,
number_rows: size,
return_type: &DataType::Utf8,
}))
})
},
);
Expand All @@ -167,8 +180,12 @@ fn criterion_benchmark(c: &mut Criterion) {
&format!("lower_all_values_are_ascii_string_views: size: {}, str_len: {}, null_density: {}, mixed: {}",
size, str_len, null_density, mixed),
|b| b.iter(|| {
// TODO use invoke_with_args
black_box(lower.invoke_batch(&args, size))
let args_cloned = args.clone();
black_box(lower.invoke_with_args(ScalarFunctionArgs{
args: args_cloned,
number_rows: size,
return_type: &DataType::Utf8,
}))
}),
);

Expand All @@ -177,8 +194,12 @@ fn criterion_benchmark(c: &mut Criterion) {
&format!("lower_all_values_are_ascii_string_views: size: {}, str_len: {}, null_density: {}, mixed: {}",
size, str_len, null_density, mixed),
|b| b.iter(|| {
// TODO use invoke_with_args
black_box(lower.invoke_batch(&args, size))
let args_cloned = args.clone();
black_box(lower.invoke_with_args(ScalarFunctionArgs{
args: args_cloned,
number_rows: size,
return_type: &DataType::Utf8,
}))
}),
);

Expand All @@ -187,8 +208,12 @@ fn criterion_benchmark(c: &mut Criterion) {
&format!("lower_some_values_are_nonascii_string_views: size: {}, str_len: {}, non_ascii_density: {}, null_density: {}, mixed: {}",
size, str_len, 0.1, null_density, mixed),
|b| b.iter(|| {
// TODO use invoke_with_args
black_box(lower.invoke_batch(&args, size))
let args_cloned = args.clone();
black_box(lower.invoke_with_args(ScalarFunctionArgs{
args: args_cloned,
number_rows: size,
return_type: &DataType::Utf8,
}))
}),
);
}
Expand Down
11 changes: 8 additions & 3 deletions datafusion/functions/benches/ltrim.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,12 +18,13 @@
extern crate criterion;

use arrow::array::{ArrayRef, LargeStringArray, StringArray, StringViewArray};
use arrow::datatypes::DataType;
use criterion::{
black_box, criterion_group, criterion_main, measurement::Measurement, BenchmarkGroup,
Criterion, SamplingMode,
};
use datafusion_common::ScalarValue;
use datafusion_expr::{ColumnarValue, ScalarUDF};
use datafusion_expr::{ColumnarValue, ScalarFunctionArgs, ScalarUDF};
use datafusion_functions::string;
use rand::{distributions::Alphanumeric, rngs::StdRng, Rng, SeedableRng};
use std::{fmt, sync::Arc};
Expand Down Expand Up @@ -141,8 +142,12 @@ fn run_with_string_type<M: Measurement>(
),
|b| {
b.iter(|| {
// TODO use invoke_with_args
black_box(ltrim.invoke_batch(&args, size))
let args_cloned = args.clone();
black_box(ltrim.invoke_with_args(ScalarFunctionArgs {
args: args_cloned,
number_rows: size,
return_type: &DataType::Utf8,
}))
})
},
);
Expand Down
59 changes: 44 additions & 15 deletions datafusion/functions/benches/repeat.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,11 +18,12 @@
extern crate criterion;

use arrow::array::{ArrayRef, Int64Array, OffsetSizeTrait};
use arrow::datatypes::DataType;
use arrow::util::bench_util::{
create_string_array_with_len, create_string_view_array_with_len,
};
use criterion::{black_box, criterion_group, criterion_main, Criterion, SamplingMode};
use datafusion_expr::ColumnarValue;
use datafusion_expr::{ColumnarValue, ScalarFunctionArgs};
use datafusion_functions::string;
use std::sync::Arc;
use std::time::Duration;
Expand Down Expand Up @@ -73,8 +74,12 @@ fn criterion_benchmark(c: &mut Criterion) {
),
|b| {
b.iter(|| {
// TODO use invoke_with_args
black_box(repeat.invoke_batch(&args, repeat_times as usize))
let args_cloned = args.clone();
black_box(repeat.invoke_with_args(ScalarFunctionArgs {
args: args_cloned,
number_rows: repeat_times as usize,
return_type: &DataType::Utf8,
}))
})
},
);
Expand All @@ -87,8 +92,12 @@ fn criterion_benchmark(c: &mut Criterion) {
),
|b| {
b.iter(|| {
// TODO use invoke_with_args
black_box(repeat.invoke_batch(&args, repeat_times as usize))
let args_cloned = args.clone();
black_box(repeat.invoke_with_args(ScalarFunctionArgs {
args: args_cloned,
number_rows: repeat_times as usize,
return_type: &DataType::Utf8,
}))
})
},
);
Expand All @@ -101,8 +110,12 @@ fn criterion_benchmark(c: &mut Criterion) {
),
|b| {
b.iter(|| {
// TODO use invoke_with_args
black_box(repeat.invoke_batch(&args, repeat_times as usize))
let args_cloned = args.clone();
black_box(repeat.invoke_with_args(ScalarFunctionArgs {
args: args_cloned,
number_rows: repeat_times as usize,
return_type: &DataType::Utf8,
}))
})
},
);
Expand All @@ -124,8 +137,12 @@ fn criterion_benchmark(c: &mut Criterion) {
),
|b| {
b.iter(|| {
// TODO use invoke_with_args
black_box(repeat.invoke_batch(&args, repeat_times as usize))
let args_cloned = args.clone();
black_box(repeat.invoke_with_args(ScalarFunctionArgs {
args: args_cloned,
number_rows: repeat_times as usize,
return_type: &DataType::Utf8,
}))
})
},
);
Expand All @@ -138,8 +155,12 @@ fn criterion_benchmark(c: &mut Criterion) {
),
|b| {
b.iter(|| {
// TODO use invoke_with_args
black_box(repeat.invoke_batch(&args, size))
let args_cloned = args.clone();
black_box(repeat.invoke_with_args(ScalarFunctionArgs {
args: args_cloned,
number_rows: repeat_times as usize,
return_type: &DataType::Utf8,
}))
})
},
);
Expand All @@ -152,8 +173,12 @@ fn criterion_benchmark(c: &mut Criterion) {
),
|b| {
b.iter(|| {
// TODO use invoke_with_args
black_box(repeat.invoke_batch(&args, repeat_times as usize))
let args_cloned = args.clone();
black_box(repeat.invoke_with_args(ScalarFunctionArgs {
args: args_cloned,
number_rows: repeat_times as usize,
return_type: &DataType::Utf8,
}))
})
},
);
Expand All @@ -175,8 +200,12 @@ fn criterion_benchmark(c: &mut Criterion) {
),
|b| {
b.iter(|| {
// TODO use invoke_with_args
black_box(repeat.invoke_batch(&args, size))
let args_cloned = args.clone();
black_box(repeat.invoke_with_args(ScalarFunctionArgs {
args: args_cloned,
number_rows: repeat_times as usize,
return_type: &DataType::Utf8,
}))
})
},
);
Expand Down
32 changes: 25 additions & 7 deletions datafusion/functions/benches/to_hex.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,12 +17,10 @@

extern crate criterion;

use arrow::{
datatypes::{Int32Type, Int64Type},
util::bench_util::create_primitive_array,
};
use arrow::datatypes::{DataType, Int32Type, Int64Type};
use arrow::util::bench_util::create_primitive_array;
use criterion::{black_box, criterion_group, criterion_main, Criterion};
use datafusion_expr::ColumnarValue;
use datafusion_expr::{ColumnarValue, ScalarFunctionArgs};
use datafusion_functions::string;
use std::sync::Arc;

Expand All @@ -33,13 +31,33 @@ fn criterion_benchmark(c: &mut Criterion) {
let batch_len = i32_array.len();
let i32_args = vec![ColumnarValue::Array(i32_array)];
c.bench_function(&format!("to_hex i32 array: {}", size), |b| {
b.iter(|| black_box(hex.invoke_batch(&i32_args, batch_len).unwrap()))
b.iter(|| {
let args_cloned = i32_args.clone();
black_box(
hex.invoke_with_args(ScalarFunctionArgs {
args: args_cloned,
number_rows: batch_len,
return_type: &DataType::Utf8,
})
.unwrap(),
)
})
});
let i64_array = Arc::new(create_primitive_array::<Int64Type>(size, 0.2));
let batch_len = i64_array.len();
let i64_args = vec![ColumnarValue::Array(i64_array)];
c.bench_function(&format!("to_hex i64 array: {}", size), |b| {
b.iter(|| black_box(hex.invoke_batch(&i64_args, batch_len).unwrap()))
b.iter(|| {
let args_cloned = i64_args.clone();
black_box(
hex.invoke_with_args(ScalarFunctionArgs {
args: args_cloned,
number_rows: batch_len,
return_type: &DataType::Utf8,
})
.unwrap(),
)
})
});
}

Expand Down
11 changes: 8 additions & 3 deletions datafusion/functions/benches/upper.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,9 +17,10 @@

extern crate criterion;

use arrow::datatypes::DataType;
use arrow::util::bench_util::create_string_array_with_len;
use criterion::{black_box, criterion_group, criterion_main, Criterion};
use datafusion_expr::ColumnarValue;
use datafusion_expr::{ColumnarValue, ScalarFunctionArgs};
use datafusion_functions::string;
use std::sync::Arc;

Expand All @@ -38,8 +39,12 @@ fn criterion_benchmark(c: &mut Criterion) {
let args = create_args(size, 32);
c.bench_function("upper_all_values_are_ascii", |b| {
b.iter(|| {
// TODO use invoke_with_args
black_box(upper.invoke_batch(&args, size))
let args_cloned = args.clone();
black_box(upper.invoke_with_args(ScalarFunctionArgs {
args: args_cloned,
number_rows: size,
return_type: &DataType::Utf8,
}))
})
});
}
Expand Down
10 changes: 9 additions & 1 deletion datafusion/functions/benches/uuid.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,13 +17,21 @@

extern crate criterion;

use arrow::datatypes::DataType;
use criterion::{black_box, criterion_group, criterion_main, Criterion};
use datafusion_expr::ScalarFunctionArgs;
use datafusion_functions::string;

fn criterion_benchmark(c: &mut Criterion) {
let uuid = string::uuid();
c.bench_function("uuid", |b| {
b.iter(|| black_box(uuid.invoke_batch(&[], 1024)))
b.iter(|| {
black_box(uuid.invoke_with_args(ScalarFunctionArgs {
args: vec![],
number_rows: 1024,
return_type: &DataType::Utf8,
}))
})
});
}

Expand Down
Loading
Loading