Black-box testing TDDD04 Lecture 2

Transcription

1 Black-box testing TDDD04 Lecture 2 Ola Leifler

2 Unit testing Different kinds of unit testing Code inspections Code walkthroughs Black-box testing White-box testing Determines what the inputs are Determines how inputs are constructed Input Unit under test (Test object, SUT) Output PASS Oracle FAIL 2 TDDD04 Software Testing

3 Efficiency of fault detection methods For requirements defects, no empirical evidence exists at all, but the fact that costs for requirements inspections are low compared to implementing incorrect requirements indicates that reviewers should look for requirements defects through inspection. For design specification defects, the case studies and one experiment indicate that inspections are both more efficient and more effective than functional testing. For code, functional or structural testing is ranked more effective or efficient than inspection in most studies. Some studies conclude that testing and inspection find different kinds of defects, so they are complementary. Results differ when studying fault isolation and not just defect detection. P. Runeson, C. Andersson, T. Thelin, A. Andrews, and T. Berling. What do we know about defect detection methods? IEEE Software, 23(3):82 90, May TDDD04 Software Testing

4 Black-box Testing 1. Exhaustive testing 2. Equivalence class testing (Equivalence Partitioning) (chapter 3) 3. Boundary value analysis (chapter 4) 4. Decision table testing (chapter 5) 5. Pairwise testing (chapter 6) 6. State-Transition testing (chapter 7) 7. Domain analysis testing (chapter 8) 8. Use case testing (chapter 9) 4

5 Equivalence class testing Ola Leifler Equivalence Class (EC) testing is a technique used to reduce the number of test cases to a manageable level while still maintaining reasonable test coverage Each EC consists of a set of data that is treated the same by the module or that should produce the same result. Any data value within a class is equivalent, in terms of testing, to any other value Choose one point within each EC to test Discuss: Problems? Weaknesses? 5 TDDD04 Software Testing

6 Identifying equivalence classes Ola Leifler Look for conditions and tables in the specification that partition the input space. Input ExamplesResult 0-1 Infant 1-12 Child Teen 20+ Adult If all three input values are the same, then return EQUILATERAL. If two input values are the same, and the third is different, then return ISOSCELESE. If all three values are different from each other, then return SCALENE. Also look for invalid equivalence classes inputs outside the specification! 6 TDDD04 Software Testing

7 Ola Leifler Example: NextDate Problem statement: NextDate is a function of three variables: month (M), day (D) and year (Y). It returns the date of the day after the input date. The month, day and year variables have integer values subject to these conditions: C1: 1 <= month <=12; C2: 1<= day <= 31; C3: 1850 <= year <= 2050 Valid ECs: M1 = {month: 1 <= month <= 12} D1 = {day: 1 <= day <= 31} Y1 = {year: 1850 <= year <= 2050} Invalid ECs: M2 = {month: month < 1} M2 = {month: month > 12} D2 = {day: day < 1} What about combinations of M,D & Y? 7 TDDD04 Software Testing

8 Guidelines Ola Leifler 1. If an input condition specifies a range of values; identify one valid EC and two invalid EC. 2. If an input condition specifies the number (e.g., one through 6 owners can be listed for the automobile); identify one valid EC and two invalid EC (no owners; more than 6 owners). 3. If an input condition specifies a set of input values and there is reason to believe that each is handled differently by the program; identify a valid EC for each and one invalid EC. 4. If an input condition specifies a must be situation (e.g., first character of the identifier must be a letter); identify one valid EC (it is a letter) and one invalid EC (it is not a letter) 5. If there is any reason to believe that elements in an EC are not handled in an identical manner by the program, split the equivalence class into smaller equivalence classes. 8 TDDD04 Software Testing

9 Applicability and limitations Ola Leifler Most suited to systems in which much of the input data takes on values within ranges or within sets. It makes the assumption that data in the same EC is, in fact, processed in the same way by the system. The simplest way to validate this assumption is to ask the programmer about their implementation. EC testing is equally applicable at the unit, integration, system, and acceptance test levels. All it requires are inputs or outputs that can be partitioned based on the system s requirements. 9 TDDD04 Software Testing

10 Black-box Testing 1. Exhaustive testing 2. Equivalence class testing (Equivalence Partitioning) (chapter 3) 3. Boundary value analysis (chapter 4) 4. Decision table testing (chapter 5) 5. Pairwise testing (chapter 6) 6. State-Transition testing (chapter 7) 7. Domain analysis testing (chapter 8) 8. Use case testing (chapter 9) 10

11 Boundary value testing Ola Leifler Boundary value testing focuses on the boundaries between equivalence classes, simply because that is where so many defects hide. The defects can be in the requirements or in the code Method 1. Identify the equivalence classes. 2. Identify the boundaries of each EC. 3. Create test cases for each boundary value by choosing one point on the boundary, one point just below the boundary, and one point just above the boundary. 11 TDDD04 Software Testing

12 Example Ola Leifler Input Result 0 <= x < 1 Infant 1 <= x < 13 Child 13 <= x < 20 Teen 20 <= x <= 120 Adult Boundary Below On Above Add to each: Identity Expected value 12 TDDD04 Software Testing

13 Applicability and limitations Ola Leifler Boundary value testing is equally applicable at the unit, integration, system, and acceptance test levels. All it requires are inputs that can be partitioned and boundaries that can be identified based on the system s requirements. Most suited to systems in which much of the input data takes on values within ranges or within sets. It makes the assumption that the program contains no extra boundaries. These can be found through inspection or white-box texting. Boundary value testing subsumes equivalence class testing 13 TDDD04 Software Testing

14 Black-box Testing 1. Exhaustive testing 2. Equivalence class testing (Equivalence Partitioning) (chapter 3) 3. Boundary value analysis (chapter 4) 4. Decision table testing (chapter 5) 5. Pairwise testing (chapter 6) 6. State-Transition testing (chapter 7) 7. Domain analysis testing (chapter 8) 8. Use case testing (chapter 9) 14

15 4. Decision Table Testing Decision tables are an excellent tool to capture certain kinds of system requirements and to document internal system design. They are used to record complex business rules that a system must implement. In addition, they can serve as a guide to creating test cases. 15

16 Decision tables The general format of a decision table: Condition-1 Rule 1 Rule 2 Rule p Conditions Condition-2 Condition-m Action-1 Actions Action-2 Action-n 16

17 Example Decisions for bank transactions (totally imaginary): R1 R2 R3 R4 R5 R6 R7 R8 Sufficient funds Y Y Y Y N N N N Conditions Good credit Y Y N N Y Y N N Bank is creditor Y N Y N Y N Y N Execute transaction X X X X X X Actions Charge overdraft X X Send warning X X X Send to collections X 17

18 Decision tables R1-R4 R5 R6 R7 R8 C1 Y N N N N Conditions C2 - Y Y N N C3 - <= 2 >2, <=3 >3, <=4 >4 A1 X X X Actions A2 X X A3 X X X Limited entries Extended entries Don t-care-entry condition is irrelevant or not applicable; any value is OK Limited entry table conditions are binary only Extended entry table conditions may take on multiple values Decision tables are declarative order between entries does not matter 18

19 Testing using decision tables Rules become test cases Conditions become inputs Actions become expected results: Condition-1 Test case 1 Test case 2 Test case p Inputs Condition-2 Condition-m Action-1 Expected results Action-2 Action-n 19

20 Triangle program The program accepts three integers, a, b, and c as input. The three values are interpreted as representing the lengths of sides of a triangle. The integers a, b, and c must satisfy the following conditions: a, b, c forms a triangle, i.e. a < b + c b < a + c c < a + b 20

21 Decision table for the triangle program R1 R2 R3 R4 R5 R6 R7 R8 R9 a,b,c forms a triangle N Y Y Y Y Y Y Y Y a = b - Y Y Y Y N N N N Conditions a = c - Y Y N N Y Y N N b = c - Y N Y N Y N Y N Not a triangle X Actions Scalene Isosceles Eqiulateral X If it is not a triangle, the other conditions don t matter If one side equals the other two, then the other two must also be the same Some conditions are therefore impossible

22 Decision table for the triangle program R1 R2 R3 R4 R5 R6 R7 R8 R9 a,b,c forms a triangle N Y Y Y Y Y Y Y Y Co ndi a = b - Y Y Y Y N N N N tio ns a = c - Y Y N N Y Y N N b = c - Y N Y N Y N Y N Not a triangle X Ac X tio Impossible Impossible Impossible ns Isosceles X X X Eqiulateral X

23 Testing the triangle program a b c Exp Conditions Actions a < b + c b < a + c c < a + b a = b a = c b = c Not a triangle Scalene Isosceles Eqiulateral T1 T2 T3 T4 T5 T6 T7 T8 T9 T10 T11

24 Testing the triangle program Conditions a < b + c N b < a + c - c < a + b - a = b - a = c - b = c - a b c Exp T NaT T2 T3 T4 T5 T6 Not a triangle X T7 Actions Scalene Isosceles Eqiulateral T8 T9 T10 T11

25 Testing the triangle program Conditions Actions a < b + c F T T T b < a + c - F T T c < a + b - - F T a = b T a = c T b = c T Not a triangle X X X Scalene Isosceles a b c Exp T NaT T NaT T NaT T Eq T5 T6 T7 T8 T9 Eqiulateral X T10 T11

26 Testing the triangle program a < b + c F T T T T T T t t t T b < a + c - F T T T T T t t T T Co ndi c < a + b - - F T T T T T T T T tio ns a = b T T T F T F F F a = c T T F F F T T F b = c T F T F F F T T Not a triangle X X X Act ion Scalene X s Isosceles X X X Eqiulateral X a b c Exp T NaT T NaT T NaT T Eq T5 T6 T Sc T Iso T Iso T10 T Iso

27 Testing the triangle program a < b + c F T T T T T b < a + c - F T T T T Co ndi c < a + b - - F T T T tio ns a = b T T T a = c T T F b = c T F T Not a triangle X X X Act ion Scalene s Isosceles a b c Exp T NaT T NaT T NaT T Eq T T T7 T8 T9 Eqiulateral X T10 T11

28 Testing the triangle program a < b + c F T T T T T T T T T T b < a + c - F T T T T T T T T T Co ndi c < a + b - - F T T T T T T T T tio ns a = b T T T T F F F F a = c T T F F T T F F b = c T F T F T F T F Not a triangle X X X a b c Exp T NaT T NaT T NaT T Eq T T T Iso Act ion Scalene s Isosceles X X X X T T Iso Eqiulateral X T Iso T Sca

29 How many test cases If there are n conditions, there must be 2 n rules Each don t care makes that rule count double If there are m don t care entries in a rule, that rule counts as 2 m rules a < b + c F T T T b < a + c - F T T Co ndi c < a + b - - F T tio ns a = b T a = c T b = c T Rule count?? 8 1

30 The NextDate Function Triangle program: relationships between inputs and correct outputs. NextDate function: logical relationships among the input variables. Problem statement: NextDate is a function of three variables: month, day and year. It returns the date of the day after the input date. The month, day and year variables have integer values subject to these conditions: C1: 1 <= month <=12 C2: 1<= day <= 31 C3: 1850 <= year <=

31 Complexity The rule that determines when a year is a leap year: A year is days long. Declaring a leap year every fourth year, a slight error would occur. The Gregorian Calendar (after Pope Gregory) resolves this by adjusting leap years on century years. Thus, a year is a leap year if it is dividable by 4, unless it is a century year. Century years are leap years only if they are multiples of 400 (Inglis, 1961); so 1992, 1996, and 2000 are leap years, while the year 1900 is not a leap years. 31

32 The NextDate function Conditions Actions month in M30 T month in M31 - T - - month = T - month = T 1 <= day <=27 T T - T day = T - day = day = day = leap year F F F F Increment year Increment month X Increment day X X X Reset year Reset month Reset day X How about when you generate automated tests based on these entries?

33 Mutually exclusive conditions C1 T - - Conditions C2 - T - C3 - - T There should be eight rules total Rule count is 4 for each rule Gives total of 12 rules! Why? A1 X Actions A2 X A3 X Which actions apply to T,F,T A1 or A3 or both? Solution Be explicit to avoid overlaps Include impossible pseudo-action Use extended entries if possible

34 The NextDate function month in M30 M30 M30 M31 M31 M12 M12 M2 M2 M2 M2 M2 M2 Conditions day in D27 D28 D29 D30 D31 D27 D28 D29 D30 D31 D27 D28 D29 D30 D31 D27 D28 D28 D29 D29 year in YL YN YL YN - Increment year Increment month X X X X X D30 D31 Increment day X X X X X Actions Reset year Reset month X Reset day X X X X X Error X X X M31 = { 1,3,5,7,8,10 }; M30 = { 4,6,9,11 }; M2 = { 2 }; M12 = { 12 } D27 = { 1, 2,, 27 }; D28 = { 28 }, D29 = { 29 }, D30 = { 30 }, D31 = { 31 } YN = not a leap year; YL = leap year

35 Test cases for NextDate Case ID Month Day Year Expected Output 1-3 April April 16, April May 1, April Error 6-9 January January 16, January February 1, December December 16, December January 1, February February 16, February February 29, February March 1, February March 1, February Error 21, 22 February Error 35

36 Applicability and Limitations Decision table testing can be used whenever the system must implement complex business rules when these rules can be represented as a combination of conditions and when these conditions have discrete actions associated with them. 36

37 Pairwise Testing 1. Exhaustive testing 2. Equivalence class testing (Equivalence Partitioning) (chapter 3) 3. Boundary value analysis (chapter 4) 4. Decision table testing (chapter 5) 5. Pairwise testing (chapter 6) 6. State-Transition testing (chapter 7) 7. Domain analysis testing (chapter 8) 8. Use case testing (chapter 9) 37

38 Pairwise testing 20 parameters, 10 values each = 10^20 combinations Test pairs only, and ensure each pair of parameter values exists at least once: All-pairs algorithm Latin square Orthogonal array

39 Black-box Testing 1. Exhaustive testing 2. Equivalence class testing (Equivalence Partitioning) (chapter 3) 3. Boundary value analysis (chapter 4) 4. Decision table testing (chapter 5) 5. Pairwise testing (chapter 6) read on your own 6. State-Transition testing (chapter 7) 7. Domain analysis testing (chapter 8) 8. Use case testing (chapter 9) 39

40 State-transition testing When a system must remember something about what has happened before or when valid and invalid orders of operations exist. State-Transition diagrams direct testing efforts by identifying the states, events, actions, and transitions that should be tested. 40

41 State transition diagrams Start giveinfo /startpaytimer Current state Event Action Next state ReservationMade paymoney usercancelled Paid CustomerCancelled Start giveinfo startpaytimer Reservation Made Reservation Made paymoney usercancelled Reservation Made Paid Customer Cancelled Ticketed Model-based testing may rely on state transition diagrams Used

42 Creating test cases Create a set of test cases such that all states are visited at least once. May miss important transitions. Create a set of test cases such that all events are triggered at least once. May miss both states and transitions. Create a set of test cases such that all transitions are exercised at least once. Subsumes (includes) all-states and all-events. Stronger: cover all possible pairs of transitions Even stronger: cover all possible triplets of transitions Create a set of test cases such that all paths are executed at least once. Subsumes all others. Can be infeasible consider loops. 42

43 Wait For Pin State transition testing pinentered Second pinbad First Events Expected states pinbad pinok Third pinok Cover all states Cover all events Cover all transitions pinbad pinok Eat card Unlock

44 Applicability and Limitations Excellent to capture certain system requirements. Not applicable when the system has no state Frequently inapplicable if the system does not need to respond to outside events 44

45 Black-box Testing 1. Exhaustive testing 2. Equivalence class testing (Equivalence Partitioning) (chapter 3) 3. Boundary value analysis (chapter 4) 4. Decision table testing (chapter 5) 5. Pairwise testing (chapter 6) read on your own 6. State-Transition testing (chapter 7) 7. Domain analysis testing (chapter 8) 8. Use case testing (chapter 9) 45

46 Domain analysis testing Testing of multiple variables simultaneously. It builds on and generalizes equivalence class and boundary value testing. Is usually applied to one input variable or two simple combinations of two variables, based on specifications. 46

47 Test cases For each relationship (>=, >, <=, or <) choose one on point and one off point. For each strict equality condition (=) choose one on point and two off points, one slightly less than the conditional value and one slightly greater than the value. 47

48 Applicability and Limitations Applicable when multiple variables should be tested together either for efficiency or because of a logical interaction. Best suited to numeric values. 48

49 Black-box Testing 1. Exhaustive testing 2. Equivalence class testing (Equivalence Partitioning) (chapter 3) 3. Boundary value analysis (chapter 4) 4. Decision table testing (chapter 5) 5. Pairwise testing (chapter 6) read on your own 6. State-Transition testing (chapter 7) 7. Domain analysis testing (chapter 8) 8. Use case testing (chapter 9) read on your own 49

50 Black-box Testing 1. Exhaustive testing 2. Equivalence class testing (Equivalence Partitioning) (chapter 3) 3. Boundary value analysis (chapter 4) 4. Decision table testing (chapter 5) 5. Pairwise testing (chapter 6) read on your own 6. State-Transition testing (chapter 7) 7. Domain analysis testing (chapter 8) 8. Use case testing (chapter 9) read on your own

51 Testing foo: add one and divide by int foo(int j) { j = j-1; // Error, should be j+1 j = j/30000; return j; } 51 TDDD04 Software Testing