amogh-jahagirdar commented on code in PR #5398: URL: https://github.com/apache/iceberg/pull/5398#discussion_r934128197
########## spark/v3.3/spark/src/main/java/org/apache/iceberg/spark/actions/GenerateSymlinkManifestAction.java: ########## @@ -0,0 +1,145 @@ +/* + * 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.actions; + +import java.util.Arrays; +import java.util.List; +import org.apache.iceberg.MetadataTableType; +import org.apache.iceberg.Partitioning; +import org.apache.iceberg.Table; +import org.apache.iceberg.actions.GenerateSymlinkManifest; +import org.apache.iceberg.actions.GenerateSymlinkManifestActionResult; +import org.apache.iceberg.exceptions.ValidationException; +import org.apache.iceberg.relocated.com.google.common.collect.Lists; +import org.apache.iceberg.spark.JobGroupInfo; +import org.apache.iceberg.spark.SparkTableUtil; +import org.apache.iceberg.types.Types; +import org.apache.spark.sql.DataFrameWriter; +import org.apache.spark.sql.Dataset; +import org.apache.spark.sql.Row; +import org.apache.spark.sql.SparkSession; + +public class GenerateSymlinkManifestAction extends BaseSparkAction<GenerateSymlinkManifestAction> + implements GenerateSymlinkManifest { + + private final Table table; + private boolean ignoreDeleteFiles = false; + private String rootLocation; + + private static final String SYMLINK_MANIFEST_DEFAULT_DIRECTORY = "_symlink_format_manifest"; + + protected GenerateSymlinkManifestAction(SparkSession spark, Table table) { + super(spark); + ValidationException.check( + table.currentSnapshot() != null, "Cannot generate symlink manifest for empty table"); + this.table = table; + } + + @Override + public GenerateSymlinkManifestAction.Result execute() { + JobGroupInfo info = newJobGroupInfo("GENERATE-SYMLINK-MANIFEST", jobDesc()); + return withJobGroupInfo(info, this::doExecute); + } + + @Override + protected GenerateSymlinkManifestAction self() { + return this; + } + + @Override + public GenerateSymlinkManifest ignoreDeleteFiles() { + this.ignoreDeleteFiles = true; + return this; + } + + @Override + public GenerateSymlinkManifest symlinkManifestRootLocation(String rootLocation) { + this.rootLocation = rootLocation; + return this; + } + + private GenerateSymlinkManifest.Result doExecute() { + long snapshot = table.currentSnapshot().snapshotId(); + Dataset<Row> entries = + SparkTableUtil.loadMetadataTable(spark(), table, MetadataTableType.ENTRIES); + // If we're not ignoring delete files, check if there are any delete manifests for the snapshot. + if (!ignoreDeleteFiles) { + boolean deletesExist = entries.filter("status < 2 and data_file.content > 0").isEmpty(); + if (deletesExist) { + throw new UnsupportedOperationException( + "Cannot generate symlink manifest when there are delete files. Set ignore delete files for generating a " + + "symlink manifest of only data files"); Review Comment: @szehon-ho @aokolnychyi @jackye1995 Appreciate any inputs on if there is a more efficient way to just validate if there are delete files beyond filtering on the whole entries table. basically, I'm still checking is there a way at higher layers of the metadata tree to filter for the existence of delete files for a given table snapshot rather than having to do a filter on the list of data/delete files in the current table state? My original thinking was to leverage the AllManifests metadata table and filter based on reference_snapshot_id and the manifest type. Since a manifest file can only contain deletes or only contain data files, I filtered if there are any delete manifests for the given reference_snapshot_Id but reference_snapshot_Id is the snapshot when that manifest is produced which isn't what we really want. -- 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]
