hanyuzheng7 commented on code in PR #23173: URL: https://github.com/apache/flink/pull/23173#discussion_r1510162242
########## flink-table/flink-table-runtime/src/main/java/org/apache/flink/table/runtime/functions/scalar/ArrayExceptFunction.java: ########## @@ -0,0 +1,134 @@ +/* + * 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. + */ + +package org.apache.flink.table.runtime.functions.scalar; + +import org.apache.flink.annotation.Internal; +import org.apache.flink.table.data.ArrayData; +import org.apache.flink.table.data.GenericArrayData; +import org.apache.flink.table.functions.BuiltInFunctionDefinitions; +import org.apache.flink.table.functions.FunctionContext; +import org.apache.flink.table.functions.SpecializedFunction; +import org.apache.flink.table.runtime.util.EqualityAndHashcodeProvider; +import org.apache.flink.table.runtime.util.ObjectContainer; +import org.apache.flink.table.types.CollectionDataType; +import org.apache.flink.table.types.DataType; +import org.apache.flink.util.FlinkRuntimeException; + +import javax.annotation.Nullable; + +import java.util.ArrayList; +import java.util.HashMap; +import java.util.List; +import java.util.Map; + +/** Implementation of {@link BuiltInFunctionDefinitions#ARRAY_EXCEPT}. */ +@Internal +public class ArrayExceptFunction extends BuiltInScalarFunction { + private final ArrayData.ElementGetter elementGetter; + private final EqualityAndHashcodeProvider equalityAndHashcodeProvider; + + public ArrayExceptFunction(SpecializedFunction.SpecializedContext context) { + super(BuiltInFunctionDefinitions.ARRAY_EXCEPT, context); + final DataType dataType = + ((CollectionDataType) context.getCallContext().getArgumentDataTypes().get(0)) + .getElementDataType() + .toInternal(); + elementGetter = ArrayData.createElementGetter(dataType.toInternal().getLogicalType()); + this.equalityAndHashcodeProvider = new EqualityAndHashcodeProvider(context, dataType); + } + + @Override + public void open(FunctionContext context) throws Exception { + equalityAndHashcodeProvider.open(context); + } + + public @Nullable ArrayData eval(ArrayData arrayOne, ArrayData arrayTwo) { + try { + if (arrayOne == null || arrayTwo == null) { + return null; + } + + List<Object> list = new ArrayList<>(); + Map<ObjectContainer, Integer> map = new HashMap<>(); + for (int pos = 0; pos < arrayOne.size(); pos++) { + final Object element = elementGetter.getElementOrNull(arrayOne, pos); + if (element == null) { + map.put(null, map.getOrDefault(null, 0) + 1); + } else { + ObjectContainer objectContainer = new ObjectContainer(element); + map.put(objectContainer, map.getOrDefault(objectContainer, 0) + 1); + } + } + for (int pos = 0; pos < arrayTwo.size(); pos++) { + final Object element = elementGetter.getElementOrNull(arrayTwo, pos); + if (element == null) { + if (map.containsKey(null)) { + int count = map.get(null); + count--; + if (count == 0) { + map.remove(null); + } else { + map.put(null, count); + } + } + } else { + ObjectContainer objectContainer = new ObjectContainer(element); + if (map.containsKey(objectContainer)) { + int count = map.get(objectContainer); + count--; + if (count == 0) { + map.remove(objectContainer); + } else { + map.put(objectContainer, count); + } + } + } + } + for (int pos = 0; pos < arrayOne.size(); pos++) { Review Comment: > Maybe I am tired while reviewing the PR, but why do you go twice over `arrayOne`? > > Shouldn't you build cardinality map from `arrayTwo` and when going throug `arrayOne` check if the element is present in the cardinality map. If the element is present don't add it to the list and decrease the count in the map, if it is no longer in the cardinality map add it to the list? If we build a cardinality map from arrayTwo, we still need to iterate over arrayOne twice. The first iteration is to decrease the count in the map, and the second is to guarantee the order. I think I can use LinkedHashMap, which can guarantee the order of the keys. -- 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: issues-unsubscr...@flink.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
