coderfender commented on code in PR #2600: URL: https://github.com/apache/datafusion-comet/pull/2600#discussion_r2501555396
########## native/spark-expr/src/agg_funcs/sum_int.rs: ########## @@ -0,0 +1,570 @@ +// 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. + +use crate::{arithmetic_overflow_error, EvalMode}; +use arrow::array::{ + cast::AsArray, Array, ArrayRef, ArrowNativeTypeOp, ArrowPrimitiveType, BooleanArray, + Int64Array, PrimitiveArray, +}; +use arrow::datatypes::{ + ArrowNativeType, DataType, Field, FieldRef, Int16Type, Int32Type, Int64Type, Int8Type, +}; +use datafusion::common::{DataFusionError, Result as DFResult, ScalarValue}; +use datafusion::logical_expr::function::{AccumulatorArgs, StateFieldsArgs}; +use datafusion::logical_expr::Volatility::Immutable; +use datafusion::logical_expr::{ + Accumulator, AggregateUDFImpl, EmitTo, GroupsAccumulator, Signature, +}; +use std::{any::Any, sync::Arc}; + +#[derive(Debug, PartialEq, Eq, Hash)] +pub struct SumInteger { + signature: Signature, + eval_mode: EvalMode, +} + +impl SumInteger { + pub fn try_new(data_type: DataType, eval_mode: EvalMode) -> DFResult<Self> { + match data_type { + DataType::Int8 | DataType::Int16 | DataType::Int32 | DataType::Int64 => Ok(Self { + signature: Signature::user_defined(Immutable), + eval_mode, + }), + _ => Err(DataFusionError::Internal( + "Invalid data type for SumInteger".into(), + )), + } + } +} + +impl AggregateUDFImpl for SumInteger { + fn as_any(&self) -> &dyn Any { + self + } + + fn name(&self) -> &str { + "sum" + } + + fn signature(&self) -> &Signature { + &self.signature + } + + fn return_type(&self, _arg_types: &[DataType]) -> DFResult<DataType> { + Ok(DataType::Int64) + } + + fn accumulator(&self, _acc_args: AccumulatorArgs) -> DFResult<Box<dyn Accumulator>> { + Ok(Box::new(SumIntegerAccumulator::new(self.eval_mode))) + } + + fn state_fields(&self, _args: StateFieldsArgs) -> DFResult<Vec<FieldRef>> { + if self.eval_mode == EvalMode::Try { + Ok(vec![ + Arc::new(Field::new("sum", DataType::Int64, true)), + Arc::new(Field::new("has_all_nulls", DataType::Boolean, false)), + ]) + } else { + Ok(vec![Arc::new(Field::new("sum", DataType::Int64, true))]) + } + } + + fn groups_accumulator_supported(&self, _args: AccumulatorArgs) -> bool { + true + } + + fn create_groups_accumulator( + &self, + _args: AccumulatorArgs, + ) -> DFResult<Box<dyn GroupsAccumulator>> { + Ok(Box::new(SumIntGroupsAccumulator::new(self.eval_mode))) + } +} + +#[derive(Debug)] +struct SumIntegerAccumulator { + sum: Option<i64>, + eval_mode: EvalMode, + has_all_nulls: bool, +} + +impl SumIntegerAccumulator { + fn new(eval_mode: EvalMode) -> Self { + if eval_mode == EvalMode::Try { + Self { + // Try mode starts with 0 (because if this is init to None we cant say if it is none due to all nulls or due to an overflow + sum: Some(0), + has_all_nulls: true, + eval_mode, + } + } else { + Self { + sum: None, + has_all_nulls: false, + eval_mode, + } + } + } +} + +impl Accumulator for SumIntegerAccumulator { + fn update_batch(&mut self, values: &[ArrayRef]) -> DFResult<()> { + // accumulator internal to add sum and return null sum (and has_nulls false) if there is an overflow in Try Eval mode + fn update_sum_internal<T>( + int_array: &PrimitiveArray<T>, + eval_mode: EvalMode, + mut sum: i64, + ) -> Result<Option<i64>, DataFusionError> + where + T: ArrowPrimitiveType, + { + for i in 0..int_array.len() { + if !int_array.is_null(i) { + let v = int_array.value(i).to_i64().ok_or_else(|| { + DataFusionError::Internal("Failed to convert value to i64".to_string()) Review Comment: Thank you for the review . I believe we wouldnt be needing these checks at all given how we check the data types multiple times (from planning phase , internal functions) -- 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]
