ECE453/SE465 Prof. Alencar University of Waterloo 1 Data Flow Testing Slice-Based Testing Winter 2007 Based on the tutorials by Prof. Kontogiannis, Winter

ECE453/SE465 Prof. Alencar University of Waterloo 1

Data Flow TestingSlice-Based Testing

Winter 2007Based on the tutorials by Prof. Kontogiannis, Winter 2006

Presented by Akhilesh Kumar, Aleh [email protected], [email protected]

ECE453/SE465 2University of Waterloo

AgendaAgenda

Slice-Based Testing DefinitionsSlice-Based Testing Definitions Slice-Based Testing ExamplesSlice-Based Testing Examples


AgendaAgenda



Slice-Based Testing DefinitionsSlice-Based Testing Definitions

Given a program P, and a program graph G(P) Given a program P, and a program graph G(P) in which statements and statement fragments in which statements and statement fragments are numbered, and a set V of variables in P, the are numbered, and a set V of variables in P, the slice on the variable set V at statement slice on the variable set V at statement fragment nfragment n, written S(V,n), is the set node , written S(V,n), is the set node numbers of all statement fragments in P prior to numbers of all statement fragments in P prior to n that contribute to the values of variables in V n that contribute to the values of variables in V at statement fragment nat statement fragment n

The idea of slices is to separate a program into The idea of slices is to separate a program into components that have some useful meaningcomponents that have some useful meaning



We will include CONST declarations in slicesWe will include CONST declarations in slices Five forms of usage nodesFive forms of usage nodes

P-use (used in a predicate (decision))P-use (used in a predicate (decision)) C-use (used in computation)C-use (used in computation) O-use (used for output, e.g. writeln())O-use (used for output, e.g. writeln()) L-use (used for location, e.g. pointers)L-use (used for location, e.g. pointers) I-use (iteration, e.g. internal counters)I-use (iteration, e.g. internal counters)

Two forms of definition nodesTwo forms of definition nodes I-def (defined by input, e.g. readln())I-def (defined by input, e.g. readln()) A-def (defined by assignment)A-def (defined by assignment)

For now, we presume that the slice S(V,n) is a slice on For now, we presume that the slice S(V,n) is a slice on one variable, that is, the set V consists of a single one variable, that is, the set V consists of a single variable, vvariable, v



If statement fragment n (in S(V,n)) is a defining If statement fragment n (in S(V,n)) is a defining node for v, then n is included in the slicenode for v, then n is included in the slice

If statement fragment n (in S(V,n)) is a usage If statement fragment n (in S(V,n)) is a usage node for v, then n is not included in the slicenode for v, then n is not included in the slice

P-uses and C-uses of other variables are P-uses and C-uses of other variables are included to the extent that their execution affects included to the extent that their execution affects the value of the variable vthe value of the variable v

O-use, L-use, and I-use nodes O-use, L-use, and I-use nodes are excludedare excluded from slicesfrom slices

Consider making slices compilableConsider making slices compilable


AgendaAgenda



Slice-Based Testing ExamplesSlice-Based Testing Examples

Find the following Find the following program slicesprogram slices

S(commission,48)S(commission,48) S(commission,40)S(commission,40) S(commission,39)S(commission,39) S(commission,38)S(commission,38) S(sales,35)S(sales,35) S(num_locks,34)S(num_locks,34) S(num_stocks,34)S(num_stocks,34) S(num_barrels,34)S(num_barrels,34)



S(commission,48)S(commission,48) {1-5,8-11,13,14, {1-5,8-11,13,14,

19-30,36,47,48,53}19-30,36,47,48,53} S(commission,40), S(commission,40),

S(commission,39), S(commission,39), S(commission,38)S(commission,38) {{ØØ}}

S(sales,35)S(sales,35) {{ØØ}}



S(num_locks,34)S(num_locks,34) {1,8,9,10,13,14,19, {1,8,9,10,13,14,19,

22,23,24,26,29,30, 22,23,24,26,29,30, 53}53}



S(num_stocks,34)S(num_stocks,34) {1,8,9,10,13,14,20, {1,8,9,10,13,14,20,

22-25,27,29,30,53}22-25,27,29,30,53}



S(num_barrels,34)S(num_barrels,34) {1,8,9,10,13,14, 21-{1,8,9,10,13,14, 21-

25,28,29,30,53}25,28,29,30,53}



Find the program slice on Find the program slice on FinalUse(low). Use the FinalUse(low). Use the Program Dependency Program Dependency Graph approachGraph approach

int binSearch(int x, int v[], int n) { {1}int binSearch(int x, int v[], int n) { {1}

int low = 0;int low = 0; {2} {2}

int high = n – 1;int high = n – 1; {3} {3}

int mid; {4}int mid; {4}

while (low <= high) { {5}while (low <= high) { {5}

mid = (low + high) / 2; {6}mid = (low + high) / 2; {6}

if (x < v[mid])if (x < v[mid]) {7} {7}

high = mid – 1; {8}high = mid – 1; {8}

else if (x > v[mid])else if (x > v[mid]) {9} {9}

low = mid + 1; {10}low = mid + 1; {10}

elseelse {11} {11}

return mid; {12}return mid; {12}

} {13}} {13}

return -1; {14}return -1; {14}

} {15}} {15}



Program Dependency GraphProgram Dependency Graph



Slice based on the criterion FinalUse(low)Slice based on the criterion FinalUse(low)




int low = 0;int low = 0; {2} {2}





if (x < v[mid])if (x < v[mid]) {7} {7}

high = mid – 1; {8}high = mid – 1; {8}


low = mid + 1; {10}low = mid + 1; {10}

elseelse {11} {11}

return mid; {12}return mid; {12}

} {13}} {13}

return -1; {14}return -1; {14}

} {15}} {15}


int low = 0;int low = 0; {2} {2}





if (x < v[mid])if (x < v[mid]) {7} {7}

high = mid – 1; {8}high = mid – 1; {8}


low = mid + 1; {10}low = mid + 1; {10}

} {13}} {13}

} {15}} {15}

→→Slice on FinalUse(low)Slice on FinalUse(low)


AgendaAgenda



ReferencesReferences

Software Testing A Craftsman's Approach Software Testing A Craftsman's Approach 2nd edition, Paul C. Jorgensen, CRC 2nd edition, Paul C. Jorgensen, CRC Press (Chapter 10 (10.2))Press (Chapter 10 (10.2))

Documents

ECE453/SE465 Prof. Alencar University of Waterloo 1 Data Flow Testing Slice-Based Testing Winter 2007 Based on the tutorials by Prof. Kontogiannis, Winter