org.evosuite.testsuite.localsearch

Class TestSuiteLocalSearchObjective

java.lang.Object

org.evosuite.testsuite.localsearch.TestSuiteLocalSearchObjective

All Implemented Interfaces:

LocalSearchObjective<TestChromosome>

public class TestSuiteLocalSearchObjective
extends Object
implements LocalSearchObjective<TestChromosome>

Represents the fitness functions (the objective) for a whole test suite such that we will apply local search to a single test case within the whole test suite. The TestSuiteLocalSearchObjective has an internal reference to the whole test suite and an index to the position of the test case that is being modified by means of local search.

Author:

Gordon Fraser

Method Summary

Modifier and Type	Method and Description
void	addFitnessFunction(FitnessFunction<? extends Chromosome> fitness) Since the fitness is governed by the underlying suite associated to this goal, this function should never be invoked.
static TestSuiteLocalSearchObjective	buildNewTestSuiteLocalSearchObjective(List<FitnessFunction<? extends Chromosome>> fitness, TestSuiteChromosome suite, int index) Creates a new TestSuiteLocalSearchObjective for a given list of fitness functions, a test suite and a testIndex for replacing an optimised test case (i.e.
List<FitnessFunction<? extends Chromosome>>	getFitnessFunctions() Since the fitness is computed by the underlying test suite associated to this local search objective, this function should not belong.
int	hasChanged(TestChromosome testCase) Overrides the test case at position testIndex with the individual.
boolean	hasImproved(TestChromosome testCase) Replaces the test case at position testIndex with the passed TestChromosome.
boolean	hasNotWorsened(TestChromosome testCase) Returns true if by replacing the test case at position testIndex with the argument testCase, the resulting test suite has not worsened the fitness
boolean	isDone() Returns true if all the fitness functions are minimising and the fitness value for each of them is 0.0
boolean	isMaximizationObjective() Returns true if all the fitness functions are maximising functions (Observe that it is not possible to store simoustaneously maximising and minimising fitness functions)

Methods inherited from class java.lang.Object

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

Method Detail

buildNewTestSuiteLocalSearchObjective

public static TestSuiteLocalSearchObjective buildNewTestSuiteLocalSearchObjective(List<FitnessFunction<? extends Chromosome>> fitness,
                                                                                  TestSuiteChromosome suite,
                                                                                  int index)

Creates a new TestSuiteLocalSearchObjective for a given list of fitness functions, a test suite and a testIndex for replacing an optimised test case (i.e. a test case over which was applied local search)

Parameters:

fitness - the list of fitness functions to be used on the modified test suite

suite - the original test suite

index - the index (between 0 and the suite length) that will be replaced with a new test case

Returns:

isDone

public boolean isDone()

Returns true if all the fitness functions are minimising and the fitness value for each of them is 0.0

Specified by:

isDone in interface LocalSearchObjective<TestChromosome>

Returns:

hasImproved

public boolean hasImproved(TestChromosome testCase)

Replaces the test case at position testIndex with the passed TestChromosome.

Specified by:

hasImproved in interface LocalSearchObjective<TestChromosome>

Parameters:

testCase - a Chromosome object.

Returns:

a boolean.

hasNotWorsened

public boolean hasNotWorsened(TestChromosome testCase)

Returns true if by replacing the test case at position testIndex with the argument testCase, the resulting test suite has not worsened the fitness

Specified by:

hasNotWorsened in interface LocalSearchObjective<TestChromosome>

Parameters:

testCase - a Chromosome object.

Returns:

a boolean.

hasChanged

public int hasChanged(TestChromosome testCase)

Overrides the test case at position testIndex with the individual. It returns -1 if the new fitness has improved, 1 if the fitness has worsened or 0 if the fitness has not changed.

Specified by:

hasChanged in interface LocalSearchObjective<TestChromosome>

Parameters:

testCase - a Chromosome object.

Returns:

a int.

getFitnessFunctions

public List<FitnessFunction<? extends Chromosome>> getFitnessFunctions()

Since the fitness is computed by the underlying test suite associated to this local search objective, this function should not belong. TODO: Why not simply returning the fitness functions of the suite?

Specified by:

getFitnessFunctions in interface LocalSearchObjective<TestChromosome>

Returns:

a FitnessFunction object.

addFitnessFunction

public void addFitnessFunction(FitnessFunction<? extends Chromosome> fitness)

Since the fitness is governed by the underlying suite associated to this goal, this function should never be invoked.

Specified by:

addFitnessFunction in interface LocalSearchObjective<TestChromosome>

isMaximizationObjective

public boolean isMaximizationObjective()

Description copied from interface: LocalSearchObjective

Returns true if all the fitness functions are maximising functions (Observe that it is not possible to store simoustaneously maximising and minimising fitness functions)

Specified by:

isMaximizationObjective in interface LocalSearchObjective<TestChromosome>

Returns: