org.vanilladb.core.query.algebra

Class SelectPlan

java.lang.Object

org.vanilladb.core.query.algebra.ReduceRecordsPlan

org.vanilladb.core.query.algebra.SelectPlan

All Implemented Interfaces:

Plan

public class SelectPlan
extends ReduceRecordsPlan

The Plan class corresponding to the select relational algebra operator.

Constructor Summary

Constructors

Constructor and Description
SelectPlan(Plan p, Predicate pred) Creates a new select node in the query tree, having the specified subquery and predicate.

Method Summary

Modifier and Type	Method and Description
long	blocksAccessed() Estimates the number of block accesses in the selection.
static Bucket	constantRangeBucket(Bucket bkt, ConstantRange cRange) Creates a new bucket by keeping the statistics of records and values in the specified bucket falling within the specified search range.
static Histogram	constantRangeHistogram(Histogram hist, Map<String,ConstantRange> cRanges) Returns a histogram that, for each field, approximates the distribution of values from the specified histogram falling within the specified search range.
Histogram	histogram() Returns the histogram that approximates the join distribution of the field values of query results.
static Bucket	joinFieldBucket(Bucket bkt1, Bucket bkt2, double numRec) Creates a new bucket by keeping the statistics of joining records and values from the two specified buckets.
static Histogram	joinFieldsHistogram(Histogram hist, Set<String> group) Returns a histogram that, for each field, approximates the distribution of values from the specified histogram joining with other fields in the specified group.
Scan	open() Creates a select scan for this query.
static Histogram	predHistogram(Histogram hist, Predicate pred) Returns a histogram that, for each field, approximates the distribution of field values from the specified histogram satisfying the specified predicate.
long	recordsOutput() Returns an estimate of the number of records in the query's output table.
Schema	schema() Returns the schema of the selection, which is the same as in the underlying query.
String	toString()

Methods inherited from class org.vanilladb.core.query.algebra.ReduceRecordsPlan

syncHistogram

Methods inherited from class java.lang.Object

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

Constructor Detail

SelectPlan

public SelectPlan(Plan p,
                  Predicate pred)

Creates a new select node in the query tree, having the specified subquery and predicate.

Parameters:

p - the subquery

pred - the predicate

Method Detail

predHistogram

public static Histogram predHistogram(Histogram hist,
                                      Predicate pred)

Returns a histogram that, for each field, approximates the distribution of field values from the specified histogram satisfying the specified predicate.

Assumes that:

• Equality selection always finds matching records
• Values in a bucket have the same frequency (uniform frequency)
• Given values within two equal ranges (of two joinable fields), all values in the range having smaller number of values appear in the range having larger number of values
• Distributions of values in different fields are independent with each other

Parameters:

hist - the input join distribution of field values

pred - the predicate

Returns:

a histogram that, for each field, approximates the distribution of field values satisfying the predicate

constantRangeHistogram

public static Histogram constantRangeHistogram(Histogram hist,
                                               Map<String,ConstantRange> cRanges)

Returns a histogram that, for each field, approximates the distribution of values from the specified histogram falling within the specified search range.

Assumes that:

• Equality selection always finds matching records
• Values in a bucket have the same frequency (uniform frequency)
• Distributions of values in different fields are independent with each other

Parameters:

hist - the input histogram

cRanges - a map from fields to search ranges

Returns:

a histogram that, for each field, approximates the distribution of values from the specified histogram falling within the specified search range

constantRangeBucket

public static Bucket constantRangeBucket(Bucket bkt,
                                         ConstantRange cRange)

Creates a new bucket by keeping the statistics of records and values in the specified bucket falling within the specified search range.

Assumes that:

• Equality selection always finds matching records
• Values in a bucket have the same frequency (uniform frequency)

Parameters:

bkt - the input bucket

cRange - the search range

Returns:

a new bucket that keeps the statistics of records and values falling within the specified search range

joinFieldsHistogram

public static Histogram joinFieldsHistogram(Histogram hist,
                                            Set<String> group)

Returns a histogram that, for each field, approximates the distribution of values from the specified histogram joining with other fields in the specified group.

Assumes that:

• Values in a bucket have the same frequency (uniform frequency)
• Given values within two equal ranges (of two joinable fields), all values in the range having smaller number of values appear in the range having larger number of values
• Distributions of values in different fields are independent with each other

Parameters:

hist - the input histogram

group - the group of joining fields

Returns:

a histogram that, for each field, approximates the distribution of values from the specified histogram joining with other fields in the specified group

joinFieldBucket

public static Bucket joinFieldBucket(Bucket bkt1,
                                     Bucket bkt2,
                                     double numRec)

Creates a new bucket by keeping the statistics of joining records and values from the two specified buckets.

Assumes that:

• Values in a bucket have the same frequency (uniform frequency)
• Given values within two equal ranges (of two joinable fields), all values in the range having smaller number of values appear in the range having larger number of values
• Distributions of values in different fields are independent with each other

Parameters:

bkt1 - the input bucket 1

bkt2 - the input bucket 2

numRec - the total number of records in the histogram where bkt1 and bkt2 belong to

Returns:

a new bucket that keeps the statistics of joining records and values from the two specified buckets

open

public Scan open()

Creates a select scan for this query.

Returns:

a scan

See Also:

Plan.open()

blocksAccessed

public long blocksAccessed()

Estimates the number of block accesses in the selection.

Returns:

the estimated number of block accesses

See Also:

Plan.blocksAccessed()

schema

public Schema schema()

Returns the schema of the selection, which is the same as in the underlying query.

Returns:

the query's schema

See Also:

Plan.schema()

histogram

public Histogram histogram()

Returns the histogram that approximates the join distribution of the field values of query results.

Returns:

the histogram

See Also:

Plan.histogram()

recordsOutput

public long recordsOutput()

Description copied from interface: Plan

Returns an estimate of the number of records in the query's output table.

Returns:

the estimated number of output records

toString

public String toString()

Overrides:

toString in class Object