Skip to content

Optimize Dictionary groupings #21765

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

Open
Rich-T-kid wants to merge 3 commits into
base: main
Choose a base branch
from
+1,018 −2
Open

Optimize Dictionary groupings #21765

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
f7e84cb
revised-version-1
Rich-T-kid Apr 21, 2026
6675b68
updated normalization fn to use a hashmap cache instead of doing a n^…
Rich-T-kid Apr 22, 2026
5d6a9b0
remove extra vector allocations for hash insertions
Rich-T-kid Apr 22, 2026
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
64 changes: 62 additions & 2 deletions datafusion/physical-plan/src/aggregates/group_values/mod.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -23,7 +23,11 @@ use arrow::array::types::{
TimestampMillisecondType, TimestampNanosecondType, TimestampSecondType,
};
use arrow::array::{ArrayRef, downcast_primitive};
use arrow::datatypes::{DataType, SchemaRef, TimeUnit};

use arrow::datatypes::{
DataType, Int8Type, Int16Type, Int32Type, Int64Type, SchemaRef, TimeUnit, UInt8Type,
UInt16Type, UInt32Type, UInt64Type,
};
use datafusion_common::Result;

use datafusion_expr::EmitTo;
Expand All @@ -36,7 +40,9 @@ use datafusion_physical_expr::binary_map::OutputType;
use multi_group_by::GroupValuesColumn;
use row::GroupValuesRows;

pub(crate) use single_group_by::primitive::HashValue;
pub(crate) use single_group_by::{
dictionary::GroupValuesDictionary, primitive::HashValue,
};

use crate::aggregates::{
group_values::single_group_by::{
Expand Down Expand Up @@ -196,6 +202,45 @@ pub fn new_group_values(
DataType::Boolean => {
return Ok(Box::new(GroupValuesBoolean::new()));
}
DataType::Dictionary(key_type, value_type) => {
if supported_single_dictionary_value(value_type) {
return match key_type.as_ref() {
DataType::Int8 => Ok(Box::new(
GroupValuesDictionary::<Int8Type>::new(value_type),
)),
DataType::Int16 => Ok(Box::new(
GroupValuesDictionary::<Int16Type>::new(value_type),
)),
DataType::Int32 => Ok(Box::new(
GroupValuesDictionary::<Int32Type>::new(value_type),
)),
DataType::Int64 => Ok(Box::new(
GroupValuesDictionary::<Int64Type>::new(value_type),
)),
DataType::UInt8 => Ok(Box::new(
GroupValuesDictionary::<UInt8Type>::new(value_type),
)),
DataType::UInt16 => {
Ok(Box::new(GroupValuesDictionary::<UInt16Type>::new(
value_type,
)))
}
DataType::UInt32 => {
Ok(Box::new(GroupValuesDictionary::<UInt32Type>::new(
value_type,
)))
}
DataType::UInt64 => {
Ok(Box::new(GroupValuesDictionary::<UInt64Type>::new(
value_type,
)))
}
_ => Err(datafusion_common::DataFusionError::NotImplemented(
format!("Unsupported dictionary key type: {key_type:?}"),
)),
};
}
}
_ => {}
}
}
Expand All @@ -210,3 +255,18 @@ pub fn new_group_values(
Ok(Box::new(GroupValuesRows::try_new(schema)?))
}
}

fn supported_single_dictionary_value(t: &DataType) -> bool {
match t {
DataType::Utf8 | DataType::LargeUtf8 | DataType::Utf8View => true,
DataType::List(field)
if matches!(
field.data_type(),
DataType::Utf8 | DataType::Utf8View | DataType::LargeUtf8
) =>
{
true
}
_ => false,
}
}
Loading
Loading