asolimando commented on code in PR #20768: URL: https://github.com/apache/datafusion/pull/20768#discussion_r2905735318
########## datafusion/common/src/utils/aggregate.rs: ########## @@ -0,0 +1,117 @@ +// Licensed to the Apache Software Foundation (ASF) under one +// or more contributor license agreements. See the NOTICE file +// distributed with this work for additional information +// regarding copyright ownership. The ASF licenses this file +// to you under the Apache License, Version 2.0 (the +// "License"); you may not use this file except in compliance +// with the License. You may obtain a copy of the License at +// +// http://www.apache.org/licenses/LICENSE-2.0 +// +// Unless required by applicable law or agreed to in writing, +// software distributed under the License is distributed on an +// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +// KIND, either express or implied. See the License for the +// specific language governing permissions and limitations +// under the License. + +//! Scalar-level aggregation utilities for statistics merging. +//! +//! Provides a cheap pairwise [`ScalarValue`] addition that directly +//! extracts inner primitive values, avoiding the expensive +//! `ScalarValue::add` path (which round-trips through Arrow arrays). + +use crate::stats::Precision; +use crate::{Result, ScalarValue}; + +/// Add two [`ScalarValue`]s by directly extracting and adding their +/// inner primitive values. +/// +/// This avoids `ScalarValue::add` which converts both operands to +/// single-element Arrow arrays, runs the `add_wrapping` kernel, and +/// converts the result back — 3 heap allocations per call. +/// +/// For non-primitive types, falls back to `ScalarValue::add`. +pub(crate) fn scalar_add(lhs: &ScalarValue, rhs: &ScalarValue) -> Result<ScalarValue> { + macro_rules! add_wrapping { + ($lhs:expr, $rhs:expr, $VARIANT:ident) => { + match ($lhs, $rhs) { + (ScalarValue::$VARIANT(Some(a)), ScalarValue::$VARIANT(Some(b))) => { + Ok(ScalarValue::$VARIANT(Some(a.wrapping_add(*b)))) + } + (ScalarValue::$VARIANT(None), other) + | (other, ScalarValue::$VARIANT(None)) => Ok(other.clone()), + _ => unreachable!(), + } + }; + } + + macro_rules! add_decimal { + ($lhs:expr, $rhs:expr, $VARIANT:ident) => { + match ($lhs, $rhs) { + ( + ScalarValue::$VARIANT(Some(a), p, s), + ScalarValue::$VARIANT(Some(b), _, _), + ) => Ok(ScalarValue::$VARIANT(Some(a.wrapping_add(*b)), *p, *s)), + (ScalarValue::$VARIANT(None, _, _), other) + | (other, ScalarValue::$VARIANT(None, _, _)) => Ok(other.clone()), + _ => unreachable!(), + } + }; + } + + macro_rules! add_float { + ($lhs:expr, $rhs:expr, $VARIANT:ident) => { + match ($lhs, $rhs) { + (ScalarValue::$VARIANT(Some(a)), ScalarValue::$VARIANT(Some(b))) => { + Ok(ScalarValue::$VARIANT(Some(*a + *b))) + } + (ScalarValue::$VARIANT(None), other) + | (other, ScalarValue::$VARIANT(None)) => Ok(other.clone()), + _ => unreachable!(), + } + }; + } + + match lhs { + ScalarValue::Int8(_) => add_wrapping!(lhs, rhs, Int8), + ScalarValue::Int16(_) => add_wrapping!(lhs, rhs, Int16), + ScalarValue::Int32(_) => add_wrapping!(lhs, rhs, Int32), + ScalarValue::Int64(_) => add_wrapping!(lhs, rhs, Int64), + ScalarValue::UInt8(_) => add_wrapping!(lhs, rhs, UInt8), + ScalarValue::UInt16(_) => add_wrapping!(lhs, rhs, UInt16), + ScalarValue::UInt32(_) => add_wrapping!(lhs, rhs, UInt32), + ScalarValue::UInt64(_) => add_wrapping!(lhs, rhs, UInt64), + ScalarValue::Float16(_) => add_float!(lhs, rhs, Float16), + ScalarValue::Float32(_) => add_float!(lhs, rhs, Float32), + ScalarValue::Float64(_) => add_float!(lhs, rhs, Float64), + ScalarValue::Decimal32(_, _, _) => add_decimal!(lhs, rhs, Decimal32), + ScalarValue::Decimal64(_, _, _) => add_decimal!(lhs, rhs, Decimal64), + ScalarValue::Decimal128(_, _, _) => add_decimal!(lhs, rhs, Decimal128), + ScalarValue::Decimal256(_, _, _) => add_decimal!(lhs, rhs, Decimal256), + // Fallback: use the existing ScalarValue::add + _ => lhs.add(rhs), + } +} + +/// [`Precision`]-aware sum of two [`ScalarValue`] precisions using +/// cheap direct addition via [`scalar_add`]. +/// +/// Mirrors the semantics of `Precision<ScalarValue>::add` but avoids Review Comment: Q: is there any way we can make sure that `Precision<ScalarValue>::add` and `precision_add` don't drift apart over time? At the very least we could add a comment in both places to signal they should be kept in sync manually. ########## datafusion/common/src/utils/aggregate.rs: ########## @@ -0,0 +1,117 @@ +// Licensed to the Apache Software Foundation (ASF) under one +// or more contributor license agreements. See the NOTICE file +// distributed with this work for additional information +// regarding copyright ownership. The ASF licenses this file +// to you under the Apache License, Version 2.0 (the +// "License"); you may not use this file except in compliance +// with the License. You may obtain a copy of the License at +// +// http://www.apache.org/licenses/LICENSE-2.0 +// +// Unless required by applicable law or agreed to in writing, +// software distributed under the License is distributed on an +// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +// KIND, either express or implied. See the License for the +// specific language governing permissions and limitations +// under the License. + +//! Scalar-level aggregation utilities for statistics merging. +//! +//! Provides a cheap pairwise [`ScalarValue`] addition that directly +//! extracts inner primitive values, avoiding the expensive +//! `ScalarValue::add` path (which round-trips through Arrow arrays). + +use crate::stats::Precision; +use crate::{Result, ScalarValue}; + +/// Add two [`ScalarValue`]s by directly extracting and adding their +/// inner primitive values. +/// +/// This avoids `ScalarValue::add` which converts both operands to +/// single-element Arrow arrays, runs the `add_wrapping` kernel, and +/// converts the result back — 3 heap allocations per call. +/// +/// For non-primitive types, falls back to `ScalarValue::add`. +pub(crate) fn scalar_add(lhs: &ScalarValue, rhs: &ScalarValue) -> Result<ScalarValue> { + macro_rules! add_wrapping { + ($lhs:expr, $rhs:expr, $VARIANT:ident) => { + match ($lhs, $rhs) { + (ScalarValue::$VARIANT(Some(a)), ScalarValue::$VARIANT(Some(b))) => { + Ok(ScalarValue::$VARIANT(Some(a.wrapping_add(*b)))) Review Comment: Correct me if I am wrong, but it seems that `wrapping_add` would wrap the sum upon overflow, producing a negative number, I am not sure this is what we want for statistics, I propose we either mark it `Absent` or at the latest mark it as `Inexact` (which doesn't seem to be happening) as done in [Precision<ScalarValue>::add](https://github.com/apache/datafusion/blob/main/datafusion/common/src/stats.rs#L145-L147). Not in scope for this PR, but on the subject I also think [`pub sum_value: Precision<ScalarValue>`](https://github.com/apache/datafusion/blob/main/datafusion/common/src/stats.rs#L757), unlike `min` and `max` should not have matched the data type of the column, but a wide type like `Int64` to minimize the chances of overflow. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
