amogh-jahagirdar commented on code in PR #13310: URL: https://github.com/apache/iceberg/pull/13310#discussion_r2205497055
########## spark/v4.0/spark/src/main/java/org/apache/iceberg/spark/source/ExtractRowLineage.java: ########## @@ -0,0 +1,92 @@ +/* + * 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.source; + +import java.util.List; +import java.util.function.Function; +import org.apache.iceberg.MetadataColumns; +import org.apache.iceberg.Schema; +import org.apache.iceberg.relocated.com.google.common.base.Preconditions; +import org.apache.iceberg.relocated.com.google.common.collect.ImmutableList; +import org.apache.spark.sql.catalyst.InternalRow; +import org.apache.spark.sql.catalyst.ProjectingInternalRow; +import org.apache.spark.sql.catalyst.expressions.GenericInternalRow; +import org.apache.spark.sql.types.LongType$; +import org.apache.spark.sql.types.StructType; +import scala.collection.JavaConverters; + +class ExtractRowLineage implements Function<InternalRow, InternalRow> { + private static final StructType ROW_LINEAGE_SCHEMA = + new StructType() + .add(MetadataColumns.ROW_ID.name(), LongType$.MODULE$, true) + .add(MetadataColumns.LAST_UPDATED_SEQUENCE_NUMBER.name(), LongType$.MODULE$, true); + private final Schema writeSchema; + + private ProjectingInternalRow cachedRowLineageProjection; + + ExtractRowLineage(Schema writeSchema) { + this.writeSchema = writeSchema; + } + + @Override + public InternalRow apply(InternalRow meta) { + // If output schema is null, i.e. deletes in MoR, or row lineage is not required on write, + // return a null row + if (writeSchema == null || writeSchema.findField(MetadataColumns.ROW_ID.name()) == null) { + return null; + } + + // If metadata row is null, return a row where both fields are null + if (meta == null) { + return new GenericInternalRow(2); + } + + ProjectingInternalRow metaProj = (ProjectingInternalRow) meta; + // Use cached ordinals if they exist + if (cachedRowLineageProjection != null) { + cachedRowLineageProjection.project(metaProj); + return cachedRowLineageProjection; + } + + // Otherwise, discover ordinals and set values + Integer rowIdOrdinal = null; + Integer lastUpdatedOrdinal = null; + for (int i = 0; i < metaProj.numFields(); i++) { + String fieldName = metaProj.schema().fields()[i].name(); + if (fieldName.equals(MetadataColumns.ROW_ID.name())) { + rowIdOrdinal = i; + } else if (fieldName.equals(MetadataColumns.LAST_UPDATED_SEQUENCE_NUMBER.name())) { + lastUpdatedOrdinal = i; + } + } + + Preconditions.checkArgument(rowIdOrdinal != null, "Expected to find row ID in metadata row"); + Preconditions.checkArgument( + lastUpdatedOrdinal != null, + "Expected to find last updated sequence number in metadata row"); + + List<Object> rowLineageProjectionOrdinals = ImmutableList.of(rowIdOrdinal, lastUpdatedOrdinal); Review Comment: Unfortunately, this is made challenging by the java scala conversion. projectingInternalRow expects a IndexedSeq[Int] where Int is primitive in Scala, wheras Java List<Integer> is a boxed integer. It's not possible to convert directly in a type safe manner which is why this is initialized as a List<Object> and scala will correctly take it in after the conversion. The alternative is we convert to buffer[integer] and buffer[integer] into another indexed seq and that may work. But rather than doing that just having a list<object> to convert seemed more simpler even though it's a bit awkward. -- 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]
