ajantha-bhat commented on code in PR #10176: URL: https://github.com/apache/iceberg/pull/10176#discussion_r1698436823
########## core/src/main/java/org/apache/iceberg/PartitionStatsUtil.java: ########## @@ -0,0 +1,227 @@ +/* + * 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; + +import java.util.Map; +import org.apache.iceberg.data.GenericRecord; +import org.apache.iceberg.data.Record; +import org.apache.iceberg.io.CloseableIterable; +import org.apache.iceberg.relocated.com.google.common.base.Preconditions; +import org.apache.iceberg.types.Types; +import org.apache.iceberg.util.PartitionUtil; + +public class PartitionStatsUtil { + + private PartitionStatsUtil() {} + + public enum Column { + PARTITION, + SPEC_ID, + DATA_RECORD_COUNT, + DATA_FILE_COUNT, + TOTAL_DATA_FILE_SIZE_IN_BYTES, + POSITION_DELETE_RECORD_COUNT, + POSITION_DELETE_FILE_COUNT, + EQUALITY_DELETE_RECORD_COUNT, + EQUALITY_DELETE_FILE_COUNT, + TOTAL_RECORD_COUNT, + LAST_UPDATED_AT, + LAST_UPDATED_SNAPSHOT_ID + } + + /** + * Generates a Schema object as per partition statistics spec based on the given partition type. + * + * @param partitionType the struct type that defines the structure of the partition. + * @return a Schema object that corresponds to the provided partition type. + */ + public static Schema schema(Types.StructType partitionType) { + if (partitionType.fields().isEmpty()) { + throw new IllegalArgumentException("getting schema for an unpartitioned table"); + } + + return new Schema( + Types.NestedField.required(1, Column.PARTITION.name(), partitionType), + Types.NestedField.required(2, Column.SPEC_ID.name(), Types.IntegerType.get()), + Types.NestedField.required(3, Column.DATA_RECORD_COUNT.name(), Types.LongType.get()), + Types.NestedField.required(4, Column.DATA_FILE_COUNT.name(), Types.IntegerType.get()), + Types.NestedField.required( + 5, Column.TOTAL_DATA_FILE_SIZE_IN_BYTES.name(), Types.LongType.get()), + Types.NestedField.optional( + 6, Column.POSITION_DELETE_RECORD_COUNT.name(), Types.LongType.get()), + Types.NestedField.optional( + 7, Column.POSITION_DELETE_FILE_COUNT.name(), Types.IntegerType.get()), + Types.NestedField.optional( + 8, Column.EQUALITY_DELETE_RECORD_COUNT.name(), Types.LongType.get()), + Types.NestedField.optional( + 9, Column.EQUALITY_DELETE_FILE_COUNT.name(), Types.IntegerType.get()), + Types.NestedField.optional(10, Column.TOTAL_RECORD_COUNT.name(), Types.LongType.get()), + Types.NestedField.optional(11, Column.LAST_UPDATED_AT.name(), Types.LongType.get()), + Types.NestedField.optional( + 12, Column.LAST_UPDATED_SNAPSHOT_ID.name(), Types.LongType.get())); + } + + /** + * Creates an iterable of partition stats records from a given manifest file, using the specified + * table and record schema. + * + * @param table the table from which the manifest file is derived. + * @param manifest the manifest file containing metadata about the records. + * @param recordSchema the schema defining the structure of the records. + * @return a CloseableIterable of partition stats records as defined by the manifest file and + * record schema. + */ + public static CloseableIterable<Record> fromManifest( + Table table, ManifestFile manifest, Schema recordSchema) { + return CloseableIterable.transform( + ManifestFiles.open(manifest, table.io(), table.specs()) + .select(BaseScan.scanColumns(manifest.content())) + .liveEntries(), + entry -> fromManifestEntry(entry, table, recordSchema)); + } + + /** + * Appends statistics from one Record to another. + * + * @param toRecord the Record to which statistics will be appended. + * @param fromRecord the Record from which statistics will be sourced. + */ + public static void appendStatsFromRecord(Record toRecord, Record fromRecord) { + Preconditions.checkState(toRecord != null, "Record to update cannot be null"); + Preconditions.checkState(fromRecord != null, "Record to update from cannot be null"); + + toRecord.set( + Column.SPEC_ID.ordinal(), + Math.max( + (int) toRecord.get(Column.SPEC_ID.ordinal()), + (int) fromRecord.get(Column.SPEC_ID.ordinal()))); + checkAndIncrementLong(toRecord, fromRecord, Column.DATA_RECORD_COUNT); + checkAndIncrementInt(toRecord, fromRecord, Column.DATA_FILE_COUNT); + checkAndIncrementLong(toRecord, fromRecord, Column.TOTAL_DATA_FILE_SIZE_IN_BYTES); + checkAndIncrementLong(toRecord, fromRecord, Column.POSITION_DELETE_RECORD_COUNT); + checkAndIncrementInt(toRecord, fromRecord, Column.POSITION_DELETE_FILE_COUNT); + checkAndIncrementLong(toRecord, fromRecord, Column.EQUALITY_DELETE_RECORD_COUNT); + checkAndIncrementInt(toRecord, fromRecord, Column.EQUALITY_DELETE_FILE_COUNT); + checkAndIncrementLong(toRecord, fromRecord, Column.TOTAL_RECORD_COUNT); + if (fromRecord.get(Column.LAST_UPDATED_AT.ordinal()) != null) { + if (toRecord.get(Column.LAST_UPDATED_AT.ordinal()) == null Review Comment: > Hm, can we actually compute last_updated_at if at least one side has null? `fromRecord` side if it is null, then that snapshot might have expired and older than `toRecord` side. So, it won't be latest. Hence, no need to update. If `toRecord` side is null then `fromRecord` might be latest hence updating. So, when we go through all records for that partition, it will have the latest snapshot info. -- 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...@iceberg.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@iceberg.apache.org For additional commands, e-mail: issues-h...@iceberg.apache.org
