rdblue commented on code in PR #6371: URL: https://github.com/apache/iceberg/pull/6371#discussion_r1055951820
########## spark/v3.3/spark/src/main/java/org/apache/iceberg/spark/source/SparkPartitioningAwareScan.java: ########## @@ -0,0 +1,267 @@ +/* + * 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.io.IOException; +import java.io.UncheckedIOException; +import java.util.Collections; +import java.util.List; +import java.util.Set; +import org.apache.iceberg.BaseScanTaskGroup; +import org.apache.iceberg.PartitionField; +import org.apache.iceberg.PartitionScanTask; +import org.apache.iceberg.PartitionSpec; +import org.apache.iceberg.Scan; +import org.apache.iceberg.ScanTask; +import org.apache.iceberg.ScanTaskGroup; +import org.apache.iceberg.Schema; +import org.apache.iceberg.StructLike; +import org.apache.iceberg.Table; +import org.apache.iceberg.exceptions.ValidationException; +import org.apache.iceberg.expressions.Expression; +import org.apache.iceberg.io.CloseableIterable; +import org.apache.iceberg.relocated.com.google.common.collect.Lists; +import org.apache.iceberg.relocated.com.google.common.collect.Sets; +import org.apache.iceberg.spark.Spark3Util; +import org.apache.iceberg.spark.SparkReadConf; +import org.apache.iceberg.types.Types.StructType; +import org.apache.iceberg.util.StructLikeSet; +import org.apache.iceberg.util.TableScanUtil; +import org.apache.spark.sql.SparkSession; +import org.apache.spark.sql.connector.expressions.Transform; +import org.apache.spark.sql.connector.read.SupportsReportPartitioning; +import org.apache.spark.sql.connector.read.partitioning.KeyGroupedPartitioning; +import org.apache.spark.sql.connector.read.partitioning.Partitioning; +import org.apache.spark.sql.connector.read.partitioning.UnknownPartitioning; +import org.slf4j.Logger; +import org.slf4j.LoggerFactory; + +abstract class SparkPartitioningAwareScan<T extends PartitionScanTask> extends SparkScan + implements SupportsReportPartitioning { + + private static final Logger LOG = LoggerFactory.getLogger(SparkPartitioningAwareScan.class); + + private final Scan<?, ? extends ScanTask, ? extends ScanTaskGroup<?>> scan; + private final boolean preserveDataGrouping; + + private Set<PartitionSpec> specs = null; // lazy cache of scanned specs + private List<T> tasks = null; // lazy cache of uncombined tasks + private List<ScanTaskGroup<T>> taskGroups = null; // lazy cache of task groups + private StructType groupingKeyType = null; // lazy cache of the grouping key type + private Transform[] groupingKeyTransforms = null; // lazy cache of grouping key transforms + private StructLikeSet groupingKeys = null; // lazy cache of grouping keys + + SparkPartitioningAwareScan( + SparkSession spark, + Table table, + Scan<?, ? extends ScanTask, ? extends ScanTaskGroup<?>> scan, + SparkReadConf readConf, + Schema expectedSchema, + List<Expression> filters) { + + super(spark, table, readConf, expectedSchema, filters); + + this.scan = scan; + this.preserveDataGrouping = readConf.preserveDataGrouping(); + + if (scan == null) { + this.specs = Collections.emptySet(); + this.tasks = Collections.emptyList(); + this.taskGroups = Collections.emptyList(); + } + } + + protected abstract Class<T> taskJavaClass(); + + protected Scan<?, ? extends ScanTask, ? extends ScanTaskGroup<?>> scan() { + return scan; + } + + @Override + public Partitioning outputPartitioning() { + if (groupingKeyType().fields().isEmpty()) { + LOG.info("Reporting UnknownPartitioning with {} partition(s)", taskGroups().size()); + return new UnknownPartitioning(taskGroups().size()); + } else { + LOG.info( + "Reporting KeyGroupedPartitioning by {} with {} partition(s)", + groupingKeyTransforms(), + taskGroups().size()); + return new KeyGroupedPartitioning(groupingKeyTransforms(), taskGroups().size()); + } + } + + @Override + protected StructType groupingKeyType() { + if (groupingKeyType == null) { + if (preserveDataGrouping) { + this.groupingKeyType = computeGroupingKeyType(); + } else { + this.groupingKeyType = StructType.of(); + } + } + + return groupingKeyType; + } + + private StructType computeGroupingKeyType() { + return org.apache.iceberg.Partitioning.groupingKeyType(expectedSchema(), specs()); + } + + private Transform[] groupingKeyTransforms() { + if (groupingKeyTransforms == null) { + List<PartitionField> groupingKeyFields = Lists.newArrayList(); + + Set<Integer> seenFieldIds = Sets.newHashSet(); + + for (PartitionSpec spec : specs()) { Review Comment: @aokolnychyi, I don't quite understand why this doesn't just get the field IDs from `groupingKeyType()`. In order to add an ID to `groupingKeyFields`, the ID must be in `groupingKeyType()`. Using that type would also ensure that each ID is used only once, so there would be no need for `seenFieldIds`. The main reason for this that I can think of is that this depends on the field order from from looping through specs like this... but shouldn't these transforms actually match the order of `groupingKeyType`? Rather than depending on the order to match between the key type and the transforms here, I think it would make more sense to just build a map from partition field ID to partition field, then loop over the grouping key fields in order. -- 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
