Key to MidTerm Exam

Key to MidTerm Exam

Cpsc 873

Scales will vary but use case #1 has the lowest risk, use case 4 the highest, use case 2 & 3 equal amount of risk since both are errors, use cases 5,6,7 equal amounts of risk. So a relative rating would be:

#1 low

#2 high

#3 high

#4 very high

#5 medium

#6 medium

#7 medium

Test items AS STATED IN THE FIRST SENTENCE OF THE EXAM:

Player

HumanPlayer

State-based Test Strategy

5.1 Syntax – UML Statechart notation

5.2 Description of Functionality – the methods of Player and Human Player handle moves sent from the GameBoard and Game, notifications of having won/lost or tied the game, and interacting with the ORB

5.3 Arguments – We need to be able to create Move objects.

5.4 Expected Output – Each transition is expected to set the state of the object to the values representative of the target state.

5.5 Exclusions – All of the functionality of Player and HumanPlayer will be tested to an all transitions level of coverage

5.6 Dependencies – We must be able to register at least a dummy Game object with the ORB in order to complete these tests/The HumanPlayer class is a subclass of Player therefore, there is a dependency between the state machines.

5.7 Test Case Success criteria – A test case will be judged successful if the state of the OUT after the test is the expected state

Testing tasks for Player and Human Player

Inspect state machines against other design models
Select test cases by following transitions in the state diagrams
Code the tests into PACT classes including the automatic validation
Execute the tests
Evaluate the logged results

State-based test –

A Construct one test case using each technique.

<notYourTurn, setYourTurn(true), yourTurn>

Define each technique including:

What must be known to apply the technique – The state machine for the CUT must be defined
What type of faults the technique identifies – Detects when the implementation of the state machine does not match the its specification
Describe how levels of coverage are measured – either the number of states reached or the number of transitions that have been traveled

Give a justification for choosing this technique.- If Player objects can not successfully transition to certain states they will not forward on messages
Explain why the test case you constructed is a result of this technique. – The test was selected by choosing a transition from the state diagram. The initial values for the case were determined by looking at the state at the start of the transition arc. The expected result was determined by following the transition arc to its end point.
Explain where in the PACT hierarchy each test case should be implemented. This test case should be in the Player PACT class.

Specification –based test case of setMove using the post-condition –

Construct one test case using each technique. – Invoke setMove with a Move object in which position 3 is selected by Player0; the expected result is that the position in the GameBoard state now contains the id of Player0
Define each technique including:

What must be known to apply the technique – the specification for the method must be known
What type of faults the technique identifies – discrepancies between what is specified and what actually happens
Describe how levels of coverage are measured – the number of post-condition clauses that have been achieved

Give a justification for choosing this technique.- it provides a link between the design specification and the code
Explain why the test case you constructed is a result of this technique. – I selected data because it should result in the post-condition being true
Explain where in the PACT hierarchy each test case should be implemented. – This test should be in the HumanPlayerTest class because HumanPlayer overrides setMove

Interaction matrix test -

Construct one test case using each technique. Invoke setGameStatus followed by setCurrentStatus; check that result is correct
Define each technique including:

What must be known to apply the technique – the interaction analysis is given and the code is available to identify what the expected state is
What type of faults the technique identifies – This type of test identifies faults where the order of method invocations is not supposed to make a difference but does
Describe how levels of coverage are measured – The percentage of entries in the interaction matrix that have been covered

Give a justification for choosing this technique. – When a new method is added to a class, this type of test detects interactions between it and the inherited methods.
Explain why the test case you constructed is a result of this technique. The test was identified by use of the results of the interaction analysis
Explain where in the PACT hierarchy each test case should be implemented. In the HumanPlayerTest class since the setCurrentStatus method is not in Player