org.apache.drill.exec.planner.index

Interface IndexCollection

All Superinterfaces:

Iterable<IndexDescriptor>

All Known Implementing Classes:

AbstractIndexCollection, DrillIndexCollection

public interface IndexCollection
extends Iterable<IndexDescriptor>

Interface used to describe an index collection

Nested Class Summary

Nested Classes

Modifier and Type	Interface and Description
static class	IndexCollection.IndexCollectionType Types of an index collections: NATIVE_SECONDARY_INDEX_COLLECTION, EXTERNAL_SECONDARY_INDEX_COLLECTION

Method Summary

Modifier and Type	Method and Description
boolean	addIndex(IndexDescriptor index) Add a new index to the collection.
void	clearAll() Clears all entries from this index collection
IndexGroupScan	getGroupScan() If this IndexCollection exposes a single GroupScan, return the GroupScan instance.
IndexCollection.IndexCollectionType	getIndexCollectionType() Get the type of this index based on IndexCollection.IndexCollectionType
double	getRows(org.apache.calcite.rex.RexNode indexCondition) Get the estimated row count for a single index condition
boolean	isColumnIndexed(SchemaPath path) Check if the field name is the leading key of any of the indexes in this collection
boolean	removeIndex(IndexDescriptor index) Remove an index (identified by table name and index name) from the collection.
boolean	supportsFullTextSearch() Whether or not the index supports full-text search (to allow pushing down such filters)
boolean	supportsIndexSelection() Whether or not this index collection supports index selection (selecting an appropriate index out of multiple candidates).
boolean	supportsRowCountStats() Whether or not the index supports getting row count statistics

Methods inherited from interface java.lang.Iterable

forEach, iterator, spliterator

Method Detail

addIndex

boolean addIndex(IndexDescriptor index)

Add a new index to the collection. Return True if index was successfully added; False otherwise

removeIndex

boolean removeIndex(IndexDescriptor index)

Remove an index (identified by table name and index name) from the collection. Return True if index was successfully removed; False otherwise

clearAll

void clearAll()

Clears all entries from this index collection

getIndexCollectionType

IndexCollection.IndexCollectionType getIndexCollectionType()

Get the type of this index based on IndexCollection.IndexCollectionType

Returns:

one of the values in IndexCollection.IndexCollectionType

supportsIndexSelection

boolean supportsIndexSelection()

Whether or not this index collection supports index selection (selecting an appropriate index out of multiple candidates). Typically, external index collections such as Elasticsearch already have this capability while native secondary index collection may not have - in such cases, Drill needs to do the index selection.

getRows

double getRows(org.apache.calcite.rex.RexNode indexCondition)

Get the estimated row count for a single index condition

Parameters:

indexCondition - The index condition (e.g index_col1 < 10 AND index_col2 = 'abc')

Returns:

The estimated row count

supportsRowCountStats

boolean supportsRowCountStats()

Whether or not the index supports getting row count statistics

Returns:

True if index supports getting row count, False otherwise

supportsFullTextSearch

boolean supportsFullTextSearch()

Whether or not the index supports full-text search (to allow pushing down such filters)

Returns:

True if index supports full-text search, False otherwise

getGroupScan

IndexGroupScan getGroupScan()

If this IndexCollection exposes a single GroupScan, return the GroupScan instance. For external indexes such as Elasticsearch, we may have a single GroupScan representing all the indexes contained within that collection. On the other hand, for native indexes, each separate index would have its own GroupScan.

Returns:

GroupScan for this IndexCollection if available, otherwise null

isColumnIndexed

boolean isColumnIndexed(SchemaPath path)

Check if the field name is the leading key of any of the indexes in this collection

Parameters:

path -

Returns:

True if an appropriate index is found, False otherwise