rdblue commented on code in PR #5513: URL: https://github.com/apache/iceberg/pull/5513#discussion_r945205832
########## spark/v3.3/spark/src/main/java/org/apache/iceberg/spark/functions/BucketFunction.java: ########## @@ -0,0 +1,303 @@ +/* + * 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.iceberg.spark.functions; + +import java.nio.ByteBuffer; +import java.util.Set; +import org.apache.iceberg.relocated.com.google.common.collect.ImmutableSet; +import org.apache.iceberg.util.BucketUtil; +import org.apache.spark.sql.catalyst.InternalRow; +import org.apache.spark.sql.connector.catalog.functions.BoundFunction; +import org.apache.spark.sql.connector.catalog.functions.ScalarFunction; +import org.apache.spark.sql.connector.catalog.functions.UnboundFunction; +import org.apache.spark.sql.types.BinaryType; +import org.apache.spark.sql.types.ByteType; +import org.apache.spark.sql.types.DataType; +import org.apache.spark.sql.types.DataTypes; +import org.apache.spark.sql.types.DateType; +import org.apache.spark.sql.types.Decimal; +import org.apache.spark.sql.types.DecimalType; +import org.apache.spark.sql.types.IntegerType; +import org.apache.spark.sql.types.LongType; +import org.apache.spark.sql.types.ShortType; +import org.apache.spark.sql.types.StringType; +import org.apache.spark.sql.types.StructField; +import org.apache.spark.sql.types.StructType; +import org.apache.spark.sql.types.TimestampType; +import org.apache.spark.unsafe.types.UTF8String; + +/** + * A Spark function implementation for the Iceberg bucket transform. + * + * <p>Example usage: {@code SELECT system.bucket(128, 'abc')}, which returns the bucket 122. + * + * <p>Note that for performance reasons, the given input number of buckets is not validated in the + * implementations used in code-gen. The number of buckets must be positive to give meaningful + * results. + */ +public class BucketFunction implements UnboundFunction { + private static final int NUM_BUCKETS_ORDINAL = 0; + private static final int VALUE_ORDINAL = 1; + private static final Set<DataType> SUPPORTED_NUM_BUCKETS_TYPES = + ImmutableSet.of(DataTypes.ByteType, DataTypes.ShortType, DataTypes.IntegerType); + + @Override + public BoundFunction bind(StructType inputType) { + if (inputType.size() != 2) { + throw new UnsupportedOperationException( + "Wrong number of inputs (expected numBuckets and value)"); + } + + StructField numBucketsField = inputType.fields()[NUM_BUCKETS_ORDINAL]; + StructField valueField = inputType.fields()[VALUE_ORDINAL]; + + if (!SUPPORTED_NUM_BUCKETS_TYPES.contains(numBucketsField.dataType())) { + throw new UnsupportedOperationException( + "Expected number of buckets to be tinyint, shortint or int"); + } + + // TODO - Support UUID type. + DataType type = valueField.dataType(); + if (type instanceof DateType) { + return new BucketInt(DataTypes.DateType); + } else if (type instanceof ByteType + || type instanceof ShortType + || type instanceof IntegerType) { + return new BucketInt(DataTypes.IntegerType); + } else if (type instanceof LongType) { + return new BucketLong(DataTypes.LongType); + } else if (type instanceof TimestampType) { + return new BucketLong(DataTypes.TimestampType); + } else if (type instanceof DecimalType) { + return new BucketDecimal(((DecimalType) type).precision(), ((DecimalType) type).scale()); + } else if (type instanceof StringType) { + return new BucketString(); + } else if (type instanceof BinaryType) { + return new BucketBinary(); + } else { + throw new UnsupportedOperationException( + "Expected bucketed column to be date, tinyint, smallint, int, bigint, decimal, timestamp, string, or binary"); + } + } + + @Override + public String description() { + return name() + + "(numBuckets, col) - Call Iceberg's bucket transform\n" + + " numBuckets :: number of buckets to divide the rows into, e.g. bucket(100, 34) -> 79 (must be a tinyint, smallint, or int)\n" + + " col :: column to bucket (must be a date, integer, long, timestamp, decimal, string, or binary)"; + } + + @Override + public String name() { + return "bucket"; + } + + public abstract static class BucketBase implements ScalarFunction<Integer> { + @Override + public String name() { + return "bucket"; + } + + @Override + public DataType resultType() { + return DataTypes.IntegerType; + } + } + + // Used for both int and date - tinyint and smallint are upcasted to int by Spark. + public static class BucketInt extends BucketBase { + private final DataType sqlType; + + // magic method used in codegen + public static int invoke(int numBuckets, int value) { + return (BucketUtil.hashInteger(value) & Integer.MAX_VALUE) % numBuckets; + } + + public BucketInt(DataType sqlType) { + this.sqlType = sqlType; + } + + @Override + public DataType[] inputTypes() { + return new DataType[] {DataTypes.IntegerType, sqlType}; Review Comment: Oh, I see. This is for `DateType`. Looks fine to me. -- 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]
