pvary commented on code in PR #3131: URL: https://github.com/apache/hive/pull/3131#discussion_r846717142
########## iceberg/iceberg-handler/src/main/java/org/apache/iceberg/mr/hive/IcebergAcidUtil.java: ########## @@ -0,0 +1,139 @@ +/* + * 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.mr.hive; + +import java.util.List; +import java.util.Map; +import java.util.Objects; +import org.apache.iceberg.MetadataColumns; +import org.apache.iceberg.Schema; +import org.apache.iceberg.Table; +import org.apache.iceberg.data.GenericRecord; +import org.apache.iceberg.data.Record; +import org.apache.iceberg.deletes.PositionDelete; +import org.apache.iceberg.relocated.com.google.common.collect.Lists; +import org.apache.iceberg.relocated.com.google.common.collect.Maps; +import org.apache.iceberg.types.Types; +import org.apache.iceberg.util.StructProjection; + +public class IcebergAcidUtil { + + private IcebergAcidUtil() { + } + + private static final Types.NestedField PARTITION_STRUCT_META_COL = null; // placeholder value in the map + private static final Map<Types.NestedField, Integer> DELETE_FILE_READ_META_COLS = Maps.newLinkedHashMap(); + + static { + DELETE_FILE_READ_META_COLS.put(MetadataColumns.SPEC_ID, 0); + DELETE_FILE_READ_META_COLS.put(PARTITION_STRUCT_META_COL, 1); + DELETE_FILE_READ_META_COLS.put(MetadataColumns.FILE_PATH, 2); + DELETE_FILE_READ_META_COLS.put(MetadataColumns.ROW_POSITION, 3); + } + + private static final Types.NestedField PARTITION_HASH_META_COL = Types.NestedField.required( + MetadataColumns.PARTITION_COLUMN_ID, MetadataColumns.PARTITION_COLUMN_NAME, Types.LongType.get()); + private static final Map<Types.NestedField, Integer> DELETE_SERDE_META_COLS = Maps.newLinkedHashMap(); + + static { + DELETE_SERDE_META_COLS.put(MetadataColumns.SPEC_ID, 0); + DELETE_SERDE_META_COLS.put(PARTITION_HASH_META_COL, 1); + DELETE_SERDE_META_COLS.put(MetadataColumns.FILE_PATH, 2); + DELETE_SERDE_META_COLS.put(MetadataColumns.ROW_POSITION, 3); + } + + /** + * @param dataCols The columns of the original file read schema + * @param table The table object - it is used for populating the partition struct meta column + * @return The schema for reading files, extended with metadata columns needed for deletes + */ + public static Schema createFileReadSchemaForDelete(List<Types.NestedField> dataCols, Table table) { + List<Types.NestedField> cols = Lists.newArrayListWithCapacity(dataCols.size() + DELETE_FILE_READ_META_COLS.size()); + DELETE_FILE_READ_META_COLS.forEach((metaCol, index) -> { + if (metaCol == PARTITION_STRUCT_META_COL) { + cols.add(MetadataColumns.metadataColumn(table, MetadataColumns.PARTITION_COLUMN_NAME)); + } else { + cols.add(metaCol); + } + }); + cols.addAll(dataCols); + return new Schema(cols); + } + + /** + * @param dataCols The columns of the serde projection schema + * @return The schema for SerDe operations, extended with metadata columns needed for deletes + */ + public static Schema createSerdeSchemaForDelete(List<Types.NestedField> dataCols) { + List<Types.NestedField> cols = Lists.newArrayListWithCapacity(dataCols.size() + DELETE_SERDE_META_COLS.size()); + DELETE_SERDE_META_COLS.forEach((metaCol, index) -> cols.add(metaCol)); + cols.addAll(dataCols); + return new Schema(cols); + } + + /** + * @param rec The record read by the file scan task, which contains both the metadata fields and the row data fields + * @param rowData The record object to populate with the rowData fields only + * @return The position delete object + */ + public static PositionDelete<Record> getPositionDelete(Record rec, GenericRecord rowData) { + PositionDelete<Record> positionDelete = PositionDelete.create(); + String filePath = rec.get(DELETE_SERDE_META_COLS.get(MetadataColumns.FILE_PATH), String.class); + long filePosition = rec.get(DELETE_SERDE_META_COLS.get(MetadataColumns.ROW_POSITION), Long.class); + + int dataOffset = DELETE_SERDE_META_COLS.size(); // position in the rec where the actual row data begins + for (int i = dataOffset; i < rec.size(); ++i) { Review Comment: Do we want to have a config for writing deletes without the original data? It would result in faster writes (less data to read, and less data to write), but we might lose some read performance because of missing statistics. ACID tables write only the metadata. Maybe a table level, plus a global config. Maybe in a follow-up PR -- 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]
