Program Testing & Maintenance (Cambridge (CIE) A Level Computer Science): Exam Questions

A program includes the following assignment statement:

Result ← STR_TO_NUM(x) / STR_TO_NUM(y)

When the program evaluates the expression in the statement, it performs a calculation.

Variable Result is of type real and variables x and y are of type string.

Two checks are required before the calculation is performed:

Identify the type of error that could occur if these checks are not carried out and state a cause of this error.

A software developer follows a program development life cycle. The life cycle divides the development process into various stages.

The program contains a validation function.

(i) The function will:

• take an integer value as a parameter .

• return TRUE if the value is within the range 24 to 37, inclusive

• otherwise return FALSE.

Complete the table to define a test plan to thoroughly test the operation of the function.

(4)

(ii) The function is to be tested on its own. When it is shown to work correctly the function will be combined with other modules and testing will continue.

Identify the type of testing that this represents.

(1)

A program contains a global 1D array Data with 100 elements of type INTEGER.

The program contains a function Process() expressed in pseudocode as follows:

FUNCTION Process(Number : INTEGER, Label : STRING) RETURNS STRING

DECLARE Index, Count : INTEGER

DECLARE ReturnValue : STRING

Count INT(100 / Number)

Index Data[Number]

CASE OF (Index MOD 2)

0 : ReturnValue TO_UPPER(RIGHT(Label, Count))

1 : ReturnValue "****"

ENDCASE

RETURN RetVal

ENDFUNCTION

Run-time errors can be generated in different ways. For example, a run-time error will be generated if a function is called with invalid parameters.

The pseudocode contains three statements that could generate a run-time error.

Write the three statements and explain how each could generate a run-time error

A software developer follows a program development life cycle. The life cycle divides the development process into various stages.

A software developer has written modules Test_A() and Test_B(). These have been written but contain errors. These modules are called from several places in the main program and testing of the main program (integration testing) has to stop.

Identify a method that can be used to continue testing the main program before the errors in these modules have been corrected and describe how this would work.

A program is being designed in pseudocode.

The program contains a global 1D array Data of type string containing 200 elements.

The first element has the index value 1.

A procedure Process() is written to initialise the values in the array:

PROCEDURE Process(Label : STRING)

DECLARE Index : INTEGER

Index 0

INPUT Data[Index]

WHILE Index < 200

Index Index + 1

CASE OF (Index MOD 2)

0 : Data[Index] TO_UPPER(Label)

1 : Data[Index] TO_LOWER(Label)

OTHERWISE : OUTPUT "Alarm 1201"

ENDCASE

NEXT Index

OUTPUT "Completed " & Index & " times"

ENDPROCEDURE

The pseudocode contains two syntax errors and one other error.

Identify the errors.

After correcting all syntax errors, the pseudocode is translated into program code which compiles without generating any errors.

When the program is executed it unexpectedly stops responding.

Identify the type of error that has occurred.

An algorithm is designed to find the smallest numeric value from an input sequence and count how many numeric values have been input.
An example of an input sequence is:

23, AB56, 17, 23ZW, 4, 10, END

Numeric input values are all integers and non-numeric input is ignored, except for the string "END" which is used to terminate the sequence.

The algorithm is expressed in pseudocode as shown:

DECLARE NextInput : STRING
DECLARE Min, Count, Num : INTEGER

Min ← 999
Count ← 0

REPEAT
INPUT NextInput
IF IS_NUM(NextInput) = TRUE THEN
Num ← STR_TO_NUM(NextInput)
IF Num > Min THEN
Min ← Num
ENDIF Count ← Count & 1
ENDIF
UNTIL NextInput ← "END"

OUTPUT "The minimum value is ", Min, " and the count was ", Count

The pseudocode contains three errors due to the incorrect use of operators.

Identify each error and state the correction required.

The operator errors are corrected and the algorithm is tested as follows:

The input sequence:

18, 4, ONE, 27, 189, ERIC, 3, 65, END

produces the output:

The minimum value is 3 and the count was 6

The algorithm is tested with a different test data sequence. The sequence contains a mix of integer and non-numeric values. It is terminated correctly but the algorithm produces unexpected results.

(i) Explain the problem with the algorithm

[2]

(ii) Give a sequence of four test data values that could be input to demonstrate the problem.

Value 1 .............................................................................................................................. Value 2 .............................................................................................................................. Value 3 .............................................................................................................................. Value 4 ............................................................................................................................

[2]

A procedure Count() will:

1. input a value (all values will be positive integers)
2. count the number of odd values and count the number of even values
3. repeat from step 1 until the value input is 99
4. output the two count values, with a suitable message.

The value 99 must not be counted

The procedure Count() is to be tested.

Typical test data would consist of odd and even values, for example:

23, 5, 64, 100, 2002, 1, 8, 900, 99

The purpose of this test would be to test a typical mix of even and odd values and check the totals.

Give three test data sequences that may be used to test different aspects of the procedure.

Do not include invalid data.

A program is being developed in pseudocode before being converted into a programming language.

The hardware that runs the program is changed and the program needs to be modified so that it works with the new hardware.

Identify the type of maintenance that this represents and give one other reason why this type of maintenance may be needed.