ajantha-bhat commented on code in PR #13480: URL: https://github.com/apache/iceberg/pull/13480#discussion_r2197749058
########## spark/v3.5/spark/src/main/java/org/apache/iceberg/spark/procedures/ComputePartitionStatsProcedure.java: ########## @@ -0,0 +1,130 @@ +/* + * 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.procedures; + +import org.apache.iceberg.PartitionStatisticsFile; +import org.apache.iceberg.Table; +import org.apache.iceberg.UpdatePartitionStatistics; +import org.apache.iceberg.actions.ComputePartitionStats; +import org.apache.iceberg.actions.ComputePartitionStats.Result; +import org.apache.iceberg.spark.actions.SparkActions; +import org.apache.iceberg.spark.procedures.SparkProcedures.ProcedureBuilder; +import org.apache.spark.sql.catalyst.InternalRow; +import org.apache.spark.sql.connector.catalog.Identifier; +import org.apache.spark.sql.connector.catalog.TableCatalog; +import org.apache.spark.sql.connector.iceberg.catalog.ProcedureParameter; +import org.apache.spark.sql.types.DataTypes; +import org.apache.spark.sql.types.Metadata; +import org.apache.spark.sql.types.StructField; +import org.apache.spark.sql.types.StructType; +import org.apache.spark.unsafe.types.UTF8String; + +/** + * A procedure that computes the stats incrementally after the snapshot that has partition stats + * file till the given snapshot (uses current snapshot if not specified) and writes the combined + * result into a {@link PartitionStatisticsFile} after merging the stats for a given snapshot. Does + * a full compute if previous statistics file does not exist. Also registers the {@link + * PartitionStatisticsFile} to table metadata. + * + * @see SparkActions#computePartitionStats(Table) + */ +public class ComputePartitionStatsProcedure extends BaseProcedure { + + private static final ProcedureParameter TABLE_PARAM = + ProcedureParameter.required("table", DataTypes.StringType); + private static final ProcedureParameter SNAPSHOT_ID_PARAM = + ProcedureParameter.optional("snapshot_id", DataTypes.LongType); + private static final ProcedureParameter FULL_REFRESH_PARAM = + ProcedureParameter.optional("full_refresh", DataTypes.BooleanType); + + private static final ProcedureParameter[] PARAMETERS = + new ProcedureParameter[] {TABLE_PARAM, SNAPSHOT_ID_PARAM, FULL_REFRESH_PARAM}; + + private static final StructType OUTPUT_TYPE = + new StructType( + new StructField[] { + new StructField( + "partition_statistics_file", DataTypes.StringType, true, Metadata.empty()) + }); + + public static ProcedureBuilder builder() { + return new Builder<ComputePartitionStatsProcedure>() { + @Override + protected ComputePartitionStatsProcedure doBuild() { + return new ComputePartitionStatsProcedure(tableCatalog()); + } + }; + } + + private ComputePartitionStatsProcedure(TableCatalog tableCatalog) { + super(tableCatalog); + } + + @Override + public ProcedureParameter[] parameters() { + return PARAMETERS; + } + + @Override + public StructType outputType() { + return OUTPUT_TYPE; + } + + @Override + public InternalRow[] call(InternalRow args) { + ProcedureInput input = new ProcedureInput(spark(), tableCatalog(), PARAMETERS, args); + Identifier tableIdent = input.ident(TABLE_PARAM); + Long snapshotId = input.asLong(SNAPSHOT_ID_PARAM, null); + Boolean fullRefresh = input.asBoolean(FULL_REFRESH_PARAM, false); + + return modifyIcebergTable( + tableIdent, + table -> { + if (fullRefresh) { + UpdatePartitionStatistics updateStats = table.updatePartitionStatistics(); + table + .partitionStatisticsFiles() + .forEach(file -> updateStats.removePartitionStatistics(file.snapshotId())); + updateStats.commit(); Review Comment: Agree. I just removed the full refresh option now. Since the underlaying action will fallback to full compute if stats are corrupted automatically, we don't have use case for full refresh yet. If there are new use case in the future, we can support it from underlaying API to make it atomic instead of using remove stats and compute method. PR is ready now. -- 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
