advancedxy commented on code in PR #19627: URL: https://github.com/apache/datafusion/pull/19627#discussion_r2822069099
########## datafusion/spark/src/function/hash/murmur3_hash.rs: ########## @@ -0,0 +1,553 @@ +// 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. + +// This is a Spark-compatible MurmurHash3 implementation. +// The algorithm is based on Austin Appleby's original MurmurHash3: +// https://github.com/aappleby/smhasher/blob/0ff96f7835817a27d0487325b6c16033e2992eb5/src/MurmurHash3.cpp +// Spark's implementation is derived from Guava's Murmur3_32HashFunction: +// https://github.com/google/guava/blob/master/guava/src/com/google/common/hash/Murmur3_32HashFunction.java + +use std::any::Any; +use std::sync::Arc; + +use arrow::array::{ + Array, ArrayRef, ArrowNativeTypeOp, DictionaryArray, Int32Array, + types::ArrowDictionaryKeyType, +}; +use arrow::compute::take; +use arrow::datatypes::{ArrowNativeType, DataType}; +use datafusion_common::{Result, ScalarValue, exec_err, internal_err}; +use datafusion_expr::{ + ColumnarValue, ScalarFunctionArgs, ScalarUDFImpl, Signature, Volatility, +}; + +use super::utils::create_hashes_internal; + +const DEFAULT_SEED: i32 = 42; + +/// Spark-compatible murmur3 hash function. +/// <https://spark.apache.org/docs/latest/api/sql/index.html#hash> +#[derive(Debug, PartialEq, Eq, Hash)] +pub struct SparkMurmur3Hash { + signature: Signature, + aliases: Vec<String>, +} + +impl Default for SparkMurmur3Hash { + fn default() -> Self { + Self::new() + } +} + +impl SparkMurmur3Hash { + pub fn new() -> Self { + Self { + signature: Signature::variadic_any(Volatility::Immutable), + aliases: vec!["hash".to_string()], + } + } +} + +impl ScalarUDFImpl for SparkMurmur3Hash { + fn as_any(&self) -> &dyn Any { + self + } + + fn name(&self) -> &str { + "murmur3_hash" + } + + fn aliases(&self) -> &[String] { + &self.aliases + } + + fn signature(&self) -> &Signature { + &self.signature + } + + fn return_type(&self, _arg_types: &[DataType]) -> Result<DataType> { + Ok(DataType::Int32) + } + + fn invoke_with_args(&self, args: ScalarFunctionArgs) -> Result<ColumnarValue> { + if args.args.is_empty() { + return exec_err!("murmur3_hash requires at least one argument"); + } + + let num_rows = args.number_rows; + + // Initialize hashes with seed + let mut hashes: Vec<u32> = vec![DEFAULT_SEED as u32; num_rows]; + + let arrays = ColumnarValue::values_to_arrays(&args.args)?; + + // Hash each column + for (i, col) in arrays.iter().enumerate() { + hash_column_murmur3(col, &mut hashes, i == 0)?; + } + + // Convert to Int32 + let result: Vec<i32> = hashes.into_iter().map(|h| h as i32).collect(); + let result_array = Int32Array::from(result); + + if num_rows == 1 { + Ok(ColumnarValue::Scalar(ScalarValue::Int32(Some( + result_array.value(0), + )))) + } else { + Ok(ColumnarValue::Array(Arc::new(result_array))) + } Review Comment: nit: I think we could move the result array initialization to the else branch to avoid the Int32Array allocation for one element row calls. -- 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]
