org.apache.hadoop.hive.ql.index

Class IndexPredicateAnalyzer

java.lang.Object

org.apache.hadoop.hive.ql.index.IndexPredicateAnalyzer

public class IndexPredicateAnalyzer
extends Object

IndexPredicateAnalyzer decomposes predicates, separating the parts which can be satisfied by an index from the parts which cannot. Currently, it only supports pure conjunctions over binary expressions comparing a column reference with a constant value. It is assumed that all column aliases encountered refer to the same table.

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static interface	IndexPredicateAnalyzer.FieldValidator

Constructor Summary

Constructors

Constructor and Description
IndexPredicateAnalyzer()

Method Summary

Methods

Modifier and Type	Method and Description
void	addComparisonOp(String udfName) Registers a comparison operator as one which can be satisfied by an index search.
void	addComparisonOp(String columnName, String... udfs) add allowed functions per column
void	allowColumnName(String columnName) Adds a column name to the set of column names allowed.
ExprNodeDesc	analyzePredicate(ExprNodeDesc predicate, List<IndexSearchCondition> searchConditions) Analyzes a predicate.
void	clearAllowedColumnNames() Clears the set of column names allowed in comparisons.
static IndexPredicateAnalyzer	createAnalyzer(boolean equalOnly)
void	setAcceptsFields(boolean acceptsFields)
void	setFieldValidator(IndexPredicateAnalyzer.FieldValidator fieldValidator)
ExprNodeGenericFuncDesc	translateSearchConditions(List<IndexSearchCondition> searchConditions) Translates search conditions back to ExprNodeDesc form (as a left-deep conjunction).

Methods inherited from class java.lang.Object

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

Constructor Detail

IndexPredicateAnalyzer

public IndexPredicateAnalyzer()

Method Detail

setFieldValidator

public void setFieldValidator(IndexPredicateAnalyzer.FieldValidator fieldValidator)

addComparisonOp

public void addComparisonOp(String udfName)

Registers a comparison operator as one which can be satisfied by an index search. Unless this is called, analyzePredicate will never find any indexable conditions.

Parameters:

udfName - name of comparison operator as returned by either GenericUDFBridge.getUdfName() (for simple UDF's) or udf.getClass().getName() (for generic UDF's).

clearAllowedColumnNames

public void clearAllowedColumnNames()

Clears the set of column names allowed in comparisons. (Initially, all column names are allowed.)

allowColumnName

public void allowColumnName(String columnName)

Adds a column name to the set of column names allowed.

Parameters:

columnName - name of column to be allowed

addComparisonOp

public void addComparisonOp(String columnName,
                   String... udfs)

add allowed functions per column

Parameters:

columnName -

udfs -

analyzePredicate

public ExprNodeDesc analyzePredicate(ExprNodeDesc predicate,
                            List<IndexSearchCondition> searchConditions)

Analyzes a predicate.

Parameters:

predicate - predicate to be analyzed

searchConditions - receives conditions produced by analysis

Returns:

residual predicate which could not be translated to searchConditions

translateSearchConditions

public ExprNodeGenericFuncDesc translateSearchConditions(List<IndexSearchCondition> searchConditions)

Translates search conditions back to ExprNodeDesc form (as a left-deep conjunction).

Parameters:

searchConditions - (typically produced by analyzePredicate)

Returns:

ExprNodeGenericFuncDesc form of search conditions

setAcceptsFields

public void setAcceptsFields(boolean acceptsFields)

createAnalyzer

public static IndexPredicateAnalyzer createAnalyzer(boolean equalOnly)