Package org.apache.druid.segment

Class CursorBuildSpec

java.lang.Object

org.apache.druid.segment.CursorBuildSpec

public class CursorBuildSpec
extends Object

Defines the plan for how the reader will scan, filter, transform, group and aggregate, and or order the data from a CursorFactory so that an appropriate CursorHolder can be constructed. The CursorBuildSpec includes physical and virtual columns will be read from the data, a Filter so that the Cursor and VectorCursor only provide matching rows, and details on how the scanned, transformed, and filtered data will be grouped, aggegated, and ordered if applicable to allow CursorHolder construction to provide optimized Cursor or VectorCursor such as providing cursors for pre-aggregated data with Projections.

See Also:

• getFilter()
• getInterval()
• getPhysicalColumns()
• getVirtualColumns()
• getGroupingColumns()
• getAggregators()
• getPreferredOrdering()

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static class	CursorBuildSpec.CursorBuildSpecBuilder

Field Summary

Fields

Modifier and Type	Field	Description
static final CursorBuildSpec	FULL_SCAN

Constructor Summary

Constructors

Constructor	Description
CursorBuildSpec(Filter filter, org.joda.time.Interval interval, Set<String> physicalColumns, VirtualColumns virtualColumns, List<String> groupingColumns, List<AggregatorFactory> aggregators, List<OrderBy> preferredOrdering, QueryContext queryContext, QueryMetrics<?> queryMetrics)

Method Summary

Modifier and Type	Method	Description
static CursorBuildSpec.CursorBuildSpecBuilder	builder()
static CursorBuildSpec.CursorBuildSpecBuilder	builder(CursorBuildSpec spec)
boolean	equals(Object o)
List<AggregatorFactory>	getAggregators()	Any AggregatorFactory which will be used by a query engine for the CursorHolder, useful to assist in determining if CursorHolder.canVectorize(), as well as specializing the Cursor or VectorCursor if any pre-aggregated data is available.
Filter	getFilter()	Filter to supply to the CursorHolder.
List<String>	getGroupingColumns()	Any columns which will be used for grouping by a query engine for the CursorHolder, useful for specializing the Cursor or VectorCursor if any pre-aggregated data is available.
org.joda.time.Interval	getInterval()	Interval filter to supply to the CursorHolder.
Set<String>	getPhysicalColumns()	Set of physical columns required from a cursor.
List<OrderBy>	getPreferredOrdering()	List of all OrderBy columns which a query engine will use to sort its results to supply to the CursorHolder, which can allow optimization of the provided Cursor or VectorCursor if data matching the preferred ordering is available.
QueryContext	getQueryContext()	QueryContext for the CursorHolder to provide a mechanism to push various data into Cursor and VectorCursor such as QueryContexts.VECTORIZE_KEY and QueryContexts.VECTOR_SIZE_KEY
QueryMetrics<?>	getQueryMetrics()	QueryMetrics to use for measuring things involved with Cursor and VectorCursor creation.
VirtualColumns	getVirtualColumns()	Any VirtualColumns which are used by a query engine to assist in determining if CursorHolder.canVectorize()
int	hashCode()
boolean	isAggregate()	Returns true if getGroupingColumns() is not null or empty and/or getAggregators() is not null or empty.
boolean	isCompatibleOrdering(List<OrderBy> ordering)	Returns true if the given ordering is compatible with getPreferredOrdering().
String	toString()

Methods inherited from class java.lang.Object

clone, finalize, getClass, notify, notifyAll, wait, wait, wait

Field Details

FULL_SCAN

public static final CursorBuildSpec FULL_SCAN

Constructor Details

CursorBuildSpec

public CursorBuildSpec(@Nullable
 Filter filter,
 org.joda.time.Interval interval,
 @Nullable
 Set<String> physicalColumns,
 VirtualColumns virtualColumns,
 @Nullable
 List<String> groupingColumns,
 @Nullable
 List<AggregatorFactory> aggregators,
 List<OrderBy> preferredOrdering,
 QueryContext queryContext,
 @Nullable
 QueryMetrics<?> queryMetrics)

Method Details

builder

public static CursorBuildSpec.CursorBuildSpecBuilder builder()

builder

public static CursorBuildSpec.CursorBuildSpecBuilder builder(CursorBuildSpec spec)

getFilter

@Nullable
public Filter getFilter()

Filter to supply to the CursorHolder. Only rows which match will be available through the selectors created from the Cursor or VectorCursor

getInterval

public org.joda.time.Interval getInterval()

Interval filter to supply to the CursorHolder. Only rows whose timestamps fall within this range will be available through the selectors created from the Cursor or VectorCursor

getPhysicalColumns

@Nullable
public Set<String> getPhysicalColumns()

Set of physical columns required from a cursor. If null, then a CursorHolder must assume that ALL columns are required.

getVirtualColumns

public VirtualColumns getVirtualColumns()

Any VirtualColumns which are used by a query engine to assist in determining if CursorHolder.canVectorize()

getGroupingColumns

@Nullable
public List<String> getGroupingColumns()

Any columns which will be used for grouping by a query engine for the CursorHolder, useful for specializing the Cursor or VectorCursor if any pre-aggregated data is available.

getAggregators

@Nullable
public List<AggregatorFactory> getAggregators()

Any AggregatorFactory which will be used by a query engine for the CursorHolder, useful to assist in determining if CursorHolder.canVectorize(), as well as specializing the Cursor or VectorCursor if any pre-aggregated data is available.

getPreferredOrdering

public List<OrderBy> getPreferredOrdering()

List of all OrderBy columns which a query engine will use to sort its results to supply to the CursorHolder, which can allow optimization of the provided Cursor or VectorCursor if data matching the preferred ordering is available.

If not specified, the cursor will advance in the native order of the underlying data.

getQueryContext

public QueryContext getQueryContext()

QueryContext for the CursorHolder to provide a mechanism to push various data into Cursor and VectorCursor such as QueryContexts.VECTORIZE_KEY and QueryContexts.VECTOR_SIZE_KEY

getQueryMetrics

@Nullable
public QueryMetrics<?> getQueryMetrics()

QueryMetrics to use for measuring things involved with Cursor and VectorCursor creation.

isAggregate

public boolean isAggregate()

Returns true if getGroupingColumns() is not null or empty and/or getAggregators() is not null or empty. This method is useful for quickly checking if it is worth considering if a CursorFactory should attempt to produce a CursorHolder that is CursorHolder.isPreAggregated() to satisfy the build spec.

isCompatibleOrdering

public boolean isCompatibleOrdering(List<OrderBy> ordering)

Returns true if the given ordering is compatible with getPreferredOrdering(). This means that, for every column in the preferred ordering, the supplied ordering must either:

use the same direction, or

use the exact opposite direction.

The supplied ordering may also include extra columns beyond those in the preferred ordering and still satisfy this condition.

equals

public boolean equals(Object o)

Overrides:

equals in class Object

hashCode

public int hashCode()

Overrides:

hashCode in class Object

toString

public String toString()

Overrides:

toString in class Object