Package org.apache.drill.exec.planner

Class AbstractPartitionDescriptor

java.lang.Object

org.apache.drill.exec.planner.AbstractPartitionDescriptor

All Implemented Interfaces:

Iterable<List<PartitionLocation>>, PartitionDescriptor

Direct Known Subclasses:

FileSystemPartitionDescriptor, HivePartitionDescriptor, ParquetPartitionDescriptor

public abstract class AbstractPartitionDescriptor
extends Object
implements PartitionDescriptor, Iterable<List<PartitionLocation>>

Abstract base class for file system based partition descriptors and Hive partition descriptors.

Field Summary

Fields

Modifier and Type	Field	Description
protected List<List<PartitionLocation>>	locationSuperList	A sequence of sublists of partition locations combined into a single super list.
protected boolean	sublistsCreated	Indicates if the sublists of the partition locations has been created

Fields inherited from interface org.apache.drill.exec.planner.PartitionDescriptor

PARTITION_BATCH_SIZE

Constructor Summary

Constructors

Constructor	Description
AbstractPartitionDescriptor()

Method Summary

Modifier and Type	Method	Description
protected abstract void	createPartitionSublists()	Create sublists of the partition locations, each sublist of size at most PartitionDescriptor.PARTITION_BATCH_SIZE
org.apache.calcite.rel.core.TableScan	createTableScan(List<PartitionLocation> newPartitions, org.apache.hadoop.fs.Path cacheFileRoot, boolean isAllPruned, MetadataContext metaContext)	Create a new TableScan rel node, given the lists of new partitions or new files to scan and a path to a metadata cache file
Iterator<List<PartitionLocation>>	iterator()	Iterator that traverses over the super list of partition locations and each time returns a single sublist of partition locations.
boolean	supportsMetadataCachePruning()

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Methods inherited from interface java.lang.Iterable

forEach, spliterator

Methods inherited from interface org.apache.drill.exec.planner.PartitionDescriptor

createTableScan, getBaseTableLocation, getIdIfValid, getMaxHierarchyLevel, getPartitionHierarchyIndex, getVectorType, isPartitionName, populatePartitionVectors

Field Details

locationSuperList

protected List<List<PartitionLocation>> locationSuperList

A sequence of sublists of partition locations combined into a single super list. The size of each sublist is at most PartitionDescriptor.PARTITION_BATCH_SIZE For example if the size is 3, the complete list could be: {(a, b, c), {d, e, f), (g, h)}

sublistsCreated

protected boolean sublistsCreated

Indicates if the sublists of the partition locations has been created

Constructor Details

AbstractPartitionDescriptor

public AbstractPartitionDescriptor()

Method Details

createPartitionSublists

protected abstract void createPartitionSublists()

Create sublists of the partition locations, each sublist of size at most PartitionDescriptor.PARTITION_BATCH_SIZE

iterator

public Iterator<List<PartitionLocation>> iterator()

Iterator that traverses over the super list of partition locations and each time returns a single sublist of partition locations.

Specified by:

iterator in interface Iterable<List<PartitionLocation>>

supportsMetadataCachePruning

public boolean supportsMetadataCachePruning()

Specified by:

supportsMetadataCachePruning in interface PartitionDescriptor

createTableScan

public org.apache.calcite.rel.core.TableScan createTableScan(List<PartitionLocation> newPartitions,
 org.apache.hadoop.fs.Path cacheFileRoot,
 boolean isAllPruned,
 MetadataContext metaContext)
                                                      throws Exception

Description copied from interface: PartitionDescriptor

Create a new TableScan rel node, given the lists of new partitions or new files to scan and a path to a metadata cache file

Specified by:

createTableScan in interface PartitionDescriptor

Parameters:

newPartitions -

cacheFileRoot -

isAllPruned -

metaContext -

Throws:

Exception