mxm commented on code in PR #15996: URL: https://github.com/apache/iceberg/pull/15996#discussion_r3148656490
########## flink/v2.1/flink/src/main/java/org/apache/iceberg/flink/maintenance/operator/ReadCommand.java: ########## @@ -0,0 +1,98 @@ +/* + * 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.flink.maintenance.operator; + +import java.io.Serializable; +import java.util.Collections; +import java.util.List; +import org.apache.flink.annotation.Internal; +import org.apache.iceberg.FileFormat; +import org.apache.iceberg.relocated.com.google.common.collect.Lists; + +/** + * Command from the {@link EqualityConvertPlanner} to the {@link EqualityConvertReader}. Describes a + * single file to read and index. + * + * <p>{@code equalityFieldIds} is in the order the planner read it from the delete file metadata + * (not necessarily sorted). The worker's keyBy uses the canonical sorted order separately when + * deduplicating field sets. + */ +@Internal +public class ReadCommand implements Serializable { Review Comment: At the moment, `newIncrementalChangelogScan().planFiles()` throws here: https://github.com/apache/iceberg/blob/324d7cd99c5d9f0e53697506c6cbc6d46052ee42/core/src/main/java/org/apache/iceberg/BaseIncrementalChangelogScan.java#L109 #14264 removes this restriction, but `DeletedRowsScanTask` has a different delete-file model. It ships a task for all the affected data files of a delete file. For an unpartitioned table with equality deletes, one equality delete potentially affects every data file on main, so a single delete file would show up in N tasks (one per affected file). The current worker doesn't care which data files are affected, we first build an index of the table, then resolve the deletes against it. I'm not sure the model works for us. -- 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]
