deniskuzZ commented on code in PR #6878: URL: https://github.com/apache/hive/pull/6878#discussion_r4293928292
########## ql/src/java/org/apache/hadoop/hive/ql/udf/generic/TruncUtils.java: ########## @@ -0,0 +1,108 @@ +/* + * 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.hadoop.hive.ql.udf.generic; + +import java.math.BigDecimal; +import java.math.RoundingMode; + +import org.apache.hadoop.hive.common.type.HiveDecimal; +import org.apache.hadoop.hive.serde2.io.HiveDecimalWritable; + +/** + * Utility class for generic trunc UDF and its vectorized implementations. + * <p> + * Truncation drops the digits after the given scale, rounding towards zero. A negative scale + * zeroes digits left of the decimal point. The input is never scaled into a long, so the result + * is exact over the whole range of the input type. + */ +public final class TruncUtils { + + // HiveDecimal rounding takes the legacy BigDecimal rounding mode constants. + @SuppressWarnings("deprecation") + private static final int DECIMAL_ROUND_DOWN = BigDecimal.ROUND_DOWN; + + // Every decimal is smaller than 10^38 in magnitude, so lower scales also truncate it to zero. + private static final int MIN_DECIMAL_SCALE = -HiveDecimal.MAX_PRECISION; + + // 10^0 .. 10^18. Every long is smaller than 10^19 in magnitude. + private static final long[] POWERS_OF_TEN = new long[19]; Review Comment: FastHiveDecimalImpl, CastLongToDecimal64 and VectorizationContext hold the identical 19-entry table, and TruncUtils would make it a fourth copy maybe add ```` /** * Returns 10^exponent, for 0 <= exponent <= {@link #MAX_LONG_POWER_OF_TEN}. */ public static long powerOfTen(int exponent) { return powerOfTenTable[exponent]; } ```` in FastHiveDecimalImpl and reuse -- 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]
