Een 8 RET 1 Black'box testing is a software tes PY | evaluates the functionality of an ae tethod Where | internal code, structure, or i an application without key input and output—h oF implementation detaiie tre sng sl 1 i put—how the system beh ails. The foous is \ internally. haves ratherthan hominy eee vow it sets ‘ ee se vs Black box testing identifies the following kind of errors : e Incorrect or missing functions e _ Interface errors e — Errorin data model e Errorin external database access Performance errors e Initialization and termination errors various test cases are designed to test the software During the black box testing, of test case consist of input value, techniques have been derived t me of black-box technique: expected output andactual systematically. Each 10 do the black box testing output. obtained. Various of any software product. Sol Jass testing s are: iy Equivalence © (i Boundary value analysis {iii Decision table testing (iv) Cause-effect graphs 2.2 When to Use Block, + rT 0S + Black box testing is useful 9 following scene" 1. During Functional Testing features wo expect ; -ation’s fea! nar an apptication® To check whet!ee ‘During Functional Testing fying imple? verifyin! ear accepts alod credential = valli id During System and © jects invall Testing ones: ste! 2, during sy For User Testing Acceptance Testing qo evaluate overall sySt0% = base For Regression pehavior Testing requirement example nsung n online payment For Security and gateway processes Performance Testing transact! ons correctly: 3, For user Acceptance Testing (VAT) To confirm the software meets business needs before deployment. Example :A banking apP undergoes UAT to ensure users can transfer funds securely. 4, For Regression Testing To make sure new, changes do not break existing functionality. Example : Testing anve-commerce checl 5, For Security and Performance Testing ies without accessing the internal logic. To test vulnerabl kout after adding a discount feature. ' E : ingi :xample : Checking if a website prevents unauthorized access to user data. Equivale! i quivalence class testing (or Equivalence Partitioning) Equivalence partitioni partitic is ofa neem i is ablack box testing method that divides the input dome!" lasses of data from which test cases can be derived. Thes? e derived. Thes classes are kr nown as equi juivi quivalence classes. In a equivalence class each membé! (2 of cl2® 10 an pelo" with Test clas inv nut Eqguides nouns pr moot the small, cequivale soonest, F range or 119 est 22808 10M th Rather than @ outputs v ‘= B Guidelines f es a «i J w Sao test cases should & re ps 2 numbers. Values jus r v fore N) pws Example et boundaries (e.g. an arta ‘an ATM 2 M4 seal ta structures tesa oT er foe pesadefna it 20 000) then design atest case to exercise the datz step |: Feast exotureat is boundary wwe F Example eo ep 2: ‘Avwebsite allows users 10 register if their age IS between 18 and 60 (inclusive) ()Age<18 : Registration fails Te (i) 18< Age < 60: Registration passes T T 1 (lil) Age> 60; Registration fails Step 1: Identify Boundaries he key boundary values are, 18 and 60 (since th fine the valid rang? I : (si ey defin 9 {26 eeint 20% _ the customer is a memt ember and spend 1 ( giscount _ ifthe customer is not a me member and spends $ 2 i) Feediseount: cathe customer is not a member ef and spends § 50 -T) giscount. w exemple (Login Authentication System) gystem has t! ifthe username i he following login rules is correct an ord rect, login suce \d the password is correct, login si , login success, 0 @ @ username i ifthe name is correct but the password is incorrect, | login fails. ifthe username is incorrect, login fails (regardless of pa: ss of password) Identify Conditions & Actions step 1+ ect Username is cor" Password is correct Login Success Login Fails e Decision Table Step 2: Create thi Test | Username (U) password (P) Case Yes e Yes Se No Yes No Mstem yeen hout sea le is tinct min tie entry portion of the tabi Masarul A aaiion On column are known e as inputs an sare Known AS outputs. Output ae a ot s are calculat , lated depondi ‘eal cations of the program. example cn an example of a typical Decis mn table shown in Table. Tt ‘able, The inputs in this ple derive th le} vat cond i Y the outputs depending on what ditions these al ese in Po table’, "—" represents a presents the don't care entries. TABLE Rulet | Rulez | Rules \ Ruled I Rules ; = : ; \ Rules F a 1. T T Bj iT E A x x a R x into a test case leis converted i the decision tabl ponding t° gix text cases correst entry portion of ‘six ‘columns | in ple, there are decision table. Each Golumn in the Therefore in this exam the entry portion of the Example An e-commerce website Pr2 * 400, they get 2 vo tnan § y (the customer Is member and puys Mm PA set W) gise? tess, they ge discount. ends g100 or A 104, 403" member put 8? ay ee qo _ Wthe customer isa gretnan s100:theY et aad 5% discount. porbut epon™ 8, i Pcs mem! ' qu) the customer is not® se 100 oF 88 —— 0) Woe discount. mber ‘and spe” 09 (wv) tthe customer IS not ame discount. exo step1: 1¢! Cyst o TO) Mf purchase? 7 $ a \ action 20% Discount .. e' 40% Discount step 5% Discount No Discount ision Table Create the Dec! step2: No Discount (D0) jiscount | 5% Discount Salen 20% Discount | 10%D Salen (p10) Step3: fep3: Execute and Verify (i) Tetsif the custor diseaintt mer is amember and spends $ 150 -Th -They should get a 20% [30]By testing right at the pouncarle® Correctly Handlos Himlt cond nor \ hips between yes ical od LE cos0* 191 ple-bas gw ued cases. without 1 of t8t Decison abiebaee I ia . jot comp! go we US input data yoo pul ® “got qroato 108 cases We USC a motte noe such a table A docision Table * tne ait rogram is ae oirsotl ai ues of 2P 2) sing the intemal an nthe | jon ni0 i 1 : | doois jour oat pit up int 18 FIGURE 2-2 .d two vertical parts resulting in four distinct ‘The table js divided into two porizontal an' The quadrants are usuall sled Conditon stub deals with the given. conditions and th oh e areas or quadrants: ly separted by double lines. The left and of upper portion cal ‘night hand of upper portion call condition, Ia condition is true, ther it , then we write Y or T against i xX OF Le itotherwise N or F. The lled Condition Entry deals with the truth value of \ r ore {eft hand side of I i io) ig of lower portion of the decision table is called Acti ib portion enlists all the pos i i t — eae Cae actions. The right hand side of lo is wer portion is c: alled Action Entry, if fan action is to be {able then a mark ) cae Vv} : enc. E upper portion of the ropriate colum in otherwise it is lef [28] oe ov 4 ongelect Test Cases. _Input (Age) 17 18 19 59 60 61 a Execute and Verity step 3? (i) Ifentering 17 then we get output as fall (ji) Ifentering 18 then we get etiotl a 5 (i) Hfentering 19 we get output as pass (iv) Ifentering 59, we get output as pass (v) Ifentering 60 we get output as pass (vi) Ifentering 61 we get output as fail ample (ATM Withdrawal Limit) Ei ls between Rs. 100 and Rs. 5000 (inclusive), An ATM allows withdrawal: Identify Boundaries Step 1: (i) Minimum withdrawal: Rs. 100 (i), Maximum withdrawal: Rs. 5000 Step 2: Select Test Cases Expected Output Input (Amount) Fail (Too low) Pass Pass Pass Fail (Too high)gfetass behaves in similar fashion, s Sotestir NING the cod n equivalence class is ag ¢ &S good to al pelonging to that equivalence cig 5 testing the So the , the output vith an within a equivalence Class will bo « ha similar Ot testing 1,” oF diferent v- > testcase design for equi ses for an rote re eattioning isbe cl wut condition, A, ased oy ‘ = . An @ Nan eva invalid states for input conditions ativalees lass op Valuation of equ i » Typically resents numeric value, a range of v: IY, AN ing set of haa alues, a put Condit Of valid of s » € Set of icy eee eS Equivalence classes may be defined a elated values, or at, either a specific Ccording * OF a boolean cong 4, _ Ifan input condition specifies a r: to the following guideline tion, ange, oni 2 ee defined. 1@ valid an SSeS are id and two invalid classe: Ae For example, if an i i ple, Hen input condition specifies a rar identify one valid equivalence class [1 99} ee fom 1 to 99, we and two invalid equivalence 2, If an input condi i y put condition requires a specific value, one valid and t equivalence classes are defined. 3 esate If an input condition specifies a member of a set, one valid and one invalid hs li li equivalence set are. defined. 4, Ifaninput condition is boolean, one valid and one invalid class are defined. delines for the derivation of equival lata item could be developed al st number of attributes of anequival Applying the gui lence classes, test cases for each input domain d selected so that the large ind executed. Test cases 27° lence class is exercised at once. fe tions: ; d gistration system with the following conditior re Awebsite has an age-base () Users below 18 en 18 and 60 yeal ars are not (i) Users betwe (231 (i) Users above 60 yean 15 characters : Invalid (Too tong) an Mere th 1 jaentity Equivalence Classes 8 step 1 ut Range Expect al uieseehe casas a ne pected Outcome | Equivalence Class Username <5 chars Invalid Invalid Class 5 < Username < 15 chars| Valid ~~ = Valid Class ysemame > 15 chars Invalid Invalid Class vq. gtep 2? select Test Cases —4 Input (Username) Expected Output —| “abe” Invalid “user123" Valid “thisisaverylongusername” Invalid Step 3: Execute and Verify et e Enter “abe”: Should display “Invalid Username.” e Enter “user123” = Should allow login. 1 e Enter “thisisaverylongusername” + Should display “Invalid Username” fect 2.4 Boundary Value Analysis A greater number of errors occurs at the boundaries of the input dom ae lary value analysis (BVA) has bee in the “center”. It is for this reason that bound developed as a testing techni mming errors occul ng valence rat the boundaries of different equivale! . (251 Some typical progral (e —_Testing or pehavioral Testing, ofthe output for which produces nas Function@! hich involves onl jalyZe gis also know! refers to testing, W nd there is neattempt to an ure of the program isignored? For this ich the content ofa black box testing in whit Jack box is understood completely in terms Black-box testin the code, reason, functional Functional testing certain inputvalues, al the output. The internalstruct testing is sometimes referred to as @ black box isnot known and function of bl of its inputs and outputs as shown in figure. Output domain Input domain Input test System under Output test data data test heater eee FIGURE 2.1 and what output the system “uh st Sarena can be ot cases in functional testing i euiiglve:in other words, th a ae ial ig is.the requirements or see i. pecifications of the system of [20] ee Cdpartitions) Where We ox,, oups (I : into a" Equivalence Class Invalid Class valid Class Invalid Class ses 4 step 2 eet select Test cm avery possible a2, we select representative value, dof test r LF ch equivalence class : Te Expected Output T 1 7 Tet iS 30. . Tc2 Step Tc3 65 Step3: Execute and Verify Run the test cases and check if the system behaves as expected. This technique reduces the number of test cases while ensuring effective coverage, E Pai 2.4 Bi Exe : ‘ample A login system a i ccepts us in Tee, PIS Usemames between 5 to 15 characters. es 'an 5 characters : Invalid cs () Sto15 ‘Invalid (Too short) Characters - s Valid username [24]xocute a9 rll steps ae a ect use e apasewor Login ne ot @ Tet a erect U0" arn Login should fait - Login should fail _ Login should fail arttioning and boundary value analysis equivalence class P phing is @ technigy Aweakness of they do not consider combinations of MPU; Cause-effect ort inations of input condi tions in a systematic manner that helps in selecting combi oftest cases do not become large. ifying causes and effects of th effect is a distinct output condition, that number: .e system under testing This technique starts: with identi distinet input conditions and an Acause is 2 ffects, for each effect we identify the causes tha ‘Atter identifying the causes and et can produce that effect and how the ‘conditions have to be combined to make the effect rue, Conditions are combined using the Boolean operators “and,” "or", an i ani ‘not,* which are represented in the graph ind ~{The mi 1 fo ie graph by &,!, and i Hh } basic symbols used fv dre rawing cause-effect graph are shown in figure. 8&8 Varnot we Boolean operators te AND: a example occ. dito hou Let the various causes (°F ine Sie > the various effects (of action§ re Ey" os table. causes Effect so the cause Finally, Advat Thec e ° e FIGURE 2.4ye S asily in table Y Converted ino he wihis cause-effect araph can i TABLE Rulet | Ruler | <* | Rutes | ° 9° Causes 9 oO aaadada | oO my m Effects m m Finally, each columns of the decision table is converted into a test case. Advantages : The cause effect graph technique has the following advantages Number of test cases reduced. e High yield test cases. e Helpsin understanding the functionality of the system. 7 Advantages of Block Box Testing : Advantages are : 1. Effective for large systems en i nits of code than white - More effective on larger UIe 29 [Block Box Testing vantages are: ¢ Not Identify Internat 1 code lssues Gannot detect memory joaks, logic errors, oF code inefficiencies. 2 pifficult 3 2 aa Without clear and concise 8% o neti o Unnes specifications, test cases = % \ Se innecesaary are very difficult to design Seen ane eA) a : a BemsUnnecessery Frepetition MiG col earner | eanenee | caments of Code there may be unnecessary repetition of test inputs if the Difficult to i Identify Edge Cases tester is not informed that programmer has already tried these test cases. 4, Not Cover All Segments of Code Code Cheats May leave many program paths untested. 5, _ Difficult to Identity Edge Cases ster only sees inputs and outputs, unexpected edge cases may since the te be missed. Repeated Testing If functionality changes, tes! After Code Changes 6. ent updates: t cases may need frequi White Box testing oF Structural Testing or Glass Box Testin Incase of fun details are being e code 1S nidden ane no internal im deals with testing the tional testing, ntrast to this, WI tested, inceHelps to expose Ambiguities | unne ~ Ther y Easy Test ep? E cases Design test! pro Helps Find Oy Missing Requirement | 4. No Me ee. Dl 4, Focuses on Use re done froma US Ambiguities 6 F Tests al er's point of view. 5, Helps to expose slack box testing wil help to expose any ambiguities or inconsistencies int specifications. 6. EasyTest cases Design Test cases can be designe as soon as the speci ications are complete +a 7. Helps Find Missing Requirement = Since tests are bi ased on specificati issil i pe specifications, missing or incorrect requireme d i eta can be detected early,Se a ane ate pa 7 ensures thorough Code coverage white box testing aims to test every Possible path, condition, and merenent within the code. This ensures that no part of the code remains untested, reducing the risk of hidden defects, identifying Hidden Bugs white box testing helps detect logical errors, jncorrect conditions, and missing j paths that may not be caught by black box testing. It ensures that all possible execution paths are reducing the ¢ of unexpected tested, chance! software failures. Early Bug Detection By exal helps identify and resolve. Improves Code Quality By identify! the softwar Strengthens Security Ituncovers security vull mining the code early in fix bugs before edundant code, Wi ing inefficient orr mance, reliability, e for better perfo 2 £ 3 & a Strengthens @ Security = roe = Validates Logic > and Design Facilitates Continuous Improvement. | Enh Be cates ) Collaboration Supports Automation the development process, white box testing they become more complex and costly to Ihite Box Testing helps optimize and maintainabilly points inthe code that could (39)roup of various struct ndopendent structures of the pre >on jules or the program. ” wr program 18 4.9! est cases are based on followin guidelines paths within a module haye chnique based Awnite box testing te g _ Atestcase should ensure thatall the independe ’ been visited at least once._, (i) Logical or Boolean conditions (or decisions) must be exercised on their true and false sides. i Loops must be executed for their poudary (edge) values as shown in figure. | Valid Boundary Values’ Invalid Boundary Values \ FIGURE 2.5 (iv) Internal nal data structures of cod VY, le must be exercised to ensure their valid 3 r validly. { 38)@ based on the control flow of th } of @ program is consid ered and ¢ Jet us first consider a control flow graph ( raph (or sir (OF simply tic W graph) o re oa oro! flow graph (or Flows graph) ie a, shows logical control fl no" 9 low using the following notation ae While-do Repeat until Switch (or case) | If-then-else FIGURE 2.7 We can make the flow graph from flowchart or from the source code of module (or program). Nodes rep! links between node. A path is a seque! path is a path whose first node independent path of a program Pis the pat one new processing condition. resent a single statement or group of statements. Edge represents the itions are referred to as predicate nodes. A node that contains conditio nce of nodes from a specific node i to node j. A complete is start node and last nod h that ensure that path co le is exit node. An ntains atleastyotdeheckEven(intnum) ( iyqnum % 2 == 0) { // Statement 1 print{(*Ever\n’);// Statement 2 1 \ } print(("Done\n"); // Statement 3 \ } int main() { checkEven(4); // Covers all statements \ \ return 0; ee: num =4 Covers all three statements in this example. umitatton* Does nott nt est both true and false branches. 03 Branch Coverage Criteria (or Testing) Inbranch coverage criteria, suitable 2 testa case ate designed to execute each branch cision at leasLonce: So, text case are designed such that the different ; branch conditions are given true and fal asbythename ofall-edge criteria. tis det auc ie arantees state! | penoted that branch coverage guara an the statement coverage: known of each de Ise values in turn. This criteria is also noted by the symbol Cy,anen It must also ment coverage, and hence this is a That is why itis symbolically stronger testing criterion th represented asfail — is that ANique ner so sting. dition. s that e the , and 2d for 32] FIGURE 2.3 Gause-effect graphing technique has following steps i (a) Various mocu" jodules in the system are identified Ve fic tions to be (b) For each of the module, identify various input conditions and th taken corres} : ponding to these input conditions. Various input conditions jons are called cause and actions corresponding to it is called effect. Hence, in this . Hence, in th step, causes and effects are listed for each module. (c) Draw cause-effect graph.. (d) Transform cause-effect graph into: decision table: (e) Convert decision table into fest cases) ig nodes are used to represent. In Gause-effect graph variou! () The input conditions or causes [33] (i) Effects of conditions+ protection against pote, 5, Thi logic, and do, algorithms, ss of aves as intended unde, °F a e correctnes a tenting PoE 16 software bel rations, ensuring f 0 Th implement improvement y of tests to source code, enabling traceabill ent cycles. future developm ation ing of the codebase, reducing pth understand 1d between developers and testers while ensuring ers gain an in-der communication overned ality goals. alignment on a! tion ‘supports Automal 10 the detailed knowledge of mated efficiently due t repetitive testing scenarios, Test cases can be autor and resources in intemal structures, saving time 2.11 Classification of White Box Testin, Classification of white box testing is shown in figure. White box Testing Control Flow- based Testing Data flow: based Testing Condition Covers ull ees or Path Testing moe Statement | |B: Co ranch [ neue. | | Coverage Coverage FIGURE 2.6 [40]ve figure f peso om, ae ve neuoeplicieese < Draw 8 step ( 2.12 FIGURE 2.8 F in figure is Jomatic complexity of flowgraph a Bed. step2 oye ule): Pade a va) = EN : = 10-842 a if fc Step3 As V(G)=4, therefore independent paths are 4 as shown below : Pant: 1-2-8-8 Pathe; 4-2-4-8 Path ; 1-5-6-8 Path4 ; 1-5-7-8 Step 4 Prepar i Pare test cases that will force executions of each path. Limitation : Path + Paths incre i crease, ©xponentially for large programs, (4 oOC—O —<_—— = ased testin, sonra tow 0d toting: Ont" Let us now consider the control flow-ba | (a) Statement coverage (b) Branch coverage | 1 (c) Condition Coverage \ (d)~ Path coverage Testing) iteria (or Tes & erage criteria Pe. 2.12.2 Statement Coveras ce. sik aie 5 | ee sais a iteria and is also known a8 all-nog, | J statos that each & na casos. Itis the simp! ee F ent coverage criteria I teria, Statement o¥e729 e 4 X= Example yelow : Consider the program segment givon be while (x > 50 && y < 10) | I | I | Sum=x+y; i | printf ("%d", Sum) ; ! | x=xt1 In this segment, the test case of x=51 and y=9 will cover all the statements. However this testing does not. guarantee that the test program will behave correctly " for all input values, [42]Example Consider the program seg i roar aN if(a> 0 && b> 0)( ! printf(Both Positive\n"); } } int main() { checkValue(1, 1); // True, True checkValue(1, -1); // True, False checkValue(-1, 1); // False, True checkValue(-1, -1); // False, False return 0; y eet () (1,1): True, True (i) (1) :Tue, False (iii) (1,1 : False, True : False, False dition is tested for both outcomes Does not check combinations. of multiple c (iv) Cr Each cont Limitation: ment given below : if (x > 50 && y < 10) { Sum=x+y> printf ("%d"; sum) + x=x+ts onditions.ample -— exam — —rydo 6088) 10) 3 condlton (> , 51 y=9 x25 i ‘a rogram all the branches of given prog } are enough to cove * fashio iter sting) 1 iteria (or Tes! a.12.5 Path Coverage C7 a . oe i that all independent Pa! Tn ey tetora no cated m ERED criteria and deny cons i feast once, That is why this cr with the symbol Cpa : y itis symbolically represe,, path coverage implies branch coverage. That is Wh) i Path coverage testing has the following steps + 1. Using the design or code as a foundation, draw a corresponding flow gray 2. 2. Determine the cyclomatic complexity of the resultant flow graph. x 4. ! Cyclomatic Ie 1 lomatic complexity, V(G), for a flowgraph (G) is defi ze Wee Noa (G) is defined as I fhere Eis the number of flowgraph ed, I Nis the number of flowgraph fades ny ¢ | Def 2: I GQ 'yclomatic complexity, V(G; , V(G), fe i / Wa) srs (G), for a flowgraph (G) is defined as Where P is the number o i K f predicate Nodes contained on a flowgraph G Se(a) Execute one pass through the looP i k (v) Execute two passes through the HooP Soseien " wi ci (c) Execute x passes through the (00P: E Execute n-1) net, passes through the (@) xect 1, (e) Skip the looP entirely. Example a -~ ( TT ginctude I I yoidloopTestcint n) { ! for (inti = 0; i< I printf(“%ed “is 7 ) | printf(“n"); 1 } | 1 ! int main() { JoopTest(0); // Zero iterations ° \ / loopTest(1); //One iteration . . ! JoopTest(3); // Multiple iterat ions ; return 0; | } | _ ee | Test cases: () N= (Loop never runs) N=1 (Loop runs once) N=3(Loop runs multiple times) Lim tation: Cannot test infinite loo (8ot tages of Path Testin, g Advantag 12: pdvantages ae: 4, Pathtesting wil detect majority o Ht ee | PREM Hh part of cose, 2. 3, _Allindependent paths deteg 4. The errors discovered are varying, Ey, may not get the same result, eryone 0 does b; 5, _ Itis helpful for testing of code thatis not yet ready § pseudocode or an algorithm Written for t IO 8xecution he program, 7 Loop Testing . st of the algorithms are implemented by using loops. There! oe be paid for testing the loops. Loop testing is a white box testin ‘hm i There are following four jn idity of loop constructs. g ly on the vali exalusive imple Loops 1. \ Figure of simple loop is ‘shown in figure. Ig!yy Example 2. TT inc | ( #include | | Nested Loops Ma loop contain of nested loo, MET loop, 4 od loop is sh . then thay teen gu ne FIGURE In case of nest = ted loop, the number Of possible tect tests would gr Set of tests as the level of nested increased. The followir ‘i wing nested loops. be applied to (a) Start at the innermost loop. Set all other loops to minimum vah (b) Conduct simple loop tests for the innermost loop, (c) Work outward, conducting tests for the next loop, but keepi ‘outer loops at minimum values and other nested loops to ty (d) Continue until all loops have been tested. jinamatnx alculate the sum of all elementsin@ . ea ols, int ations) C // Functio! ‘ ax(int rows: inte@ 4g, Determine a basis set oy thy provides the number oj, he structure. “AY indope, 4. Prepare test cases that wy, Data shoule be chosen gq a xc My 89 tha Not appropriately set as each | compared to expected ros, Path is tec, Uults, eae jrisimportantto note that some ind, pend fashion. In such cases, these path. me ‘Snt path Example tested as py, onsider the © Program to fin heroes: nu side est number tron f 4, | #include < studio, h> east | main () SEntey A int a,b,c; seanf ("%d %d%d", &a, &b, 2c); if (a>b) { A if (a>c) aan s printf (‘Largest no.=%d", a); else — 5 printf (Largest no.= %c!.c); else a alld if (b> °) 2%", b); printf ("Largestno-= " 4 1 = printf Largest adi (4— 3. Concatenated Loops Concatenated loops M! figure FIGURE 2.11 ‘ed loops, testing depends upon interdependence re independent of one another then they can be ps are interdependent, then they should be In the case of concatenate among the loops: If loops al tested one after the other. But if loo} tested in the same way aS nested loops. Example #include \ //Function with concatenated loops void il ! ProcessArray(intarr{], int size) { : \| | cases fe iZ tae P ia OP exec) lf Matrix Scution path if I : | ' elements nit i , a" {15} I | je Row Execution) ; } 4c3(5in9! fe, I inglecolumn Execution) | rea (sing (1,2), 81) | i. Execution) | rosqworme’ ) ((1,21,6,4)) {(1,2,3}, (4.5.6), 789)) 7c6 (Full Execution - Large Matrix) Coverage Analysis jon Case : Ensures that ifr iteration Case Ensures the smallest possible loop execu! ows = 0 or cols = 0, NO Joop runs. tion (1x1 Nested Loop (i) Zero Iterat (ii) Minimum t matrix). (ii) Partial Execution Cases -Test cases with 4petike execution (single ro" ies i ‘i ormal or large (iv) Normal & Large Execution: Covers all loop iterations inane 1" matrix. elin int sum = 0; int product = 1; 1/ First loop: Calcuta for (inti = 03 i< size; i44) SUM Of olomont, sum += ani}; } printf(‘Sum of array elements edn", sum) + Sum), 4 Second loop: Calculates produc for (inti = 0; i< size; iss) ( ie product *= afi}; elements } printf(“Product of array elements: } int main() { int size; // Taking input for array size printf(“Enter array size: “); scant(“%d", &size); intarr[size]; / Taking input for array elements printf(“Enter array for (inti = 0; i< sizes i++) ( Mop? &artlil)s elements:\n’); scanf( } edn", product);ypouter 1ooP yinner loop. 0; ‘ = 0; Fe rows zs cols: 4) { for(int ° for (inti = sum += matrixf0li : ) return sum ) int main() ( int rows, cols; zp Taking input for matrix size print(“Enter number of foWS: "; 96d", &rows); scant(' number of columns: “5 printf(“Ente scant(“%ed", &cols); int matrix[rows][cols]; // Taking input for matrix elements printf(“Enter matrix elements:\n’); for (inti = 0; is rows; i++) { for (intj = 0; | < cols; j++) { scanf("%d", &matrix{ilfj); } Calculating sumpairvotured LOOP H Fa toope ynich do not follow any of the above mont I typos) 2 led unstructured looy / ype ad FIGURE 2.12 lect the u Whenever possible this type of loops should be redesigned to ref of the structured programming constructs 2,13 Data flow Based Testing Data flow based testing focuses on the definitions of the variables and where the ¥ this testing scheme is to check the definitions are being used. The basic purpose Oo! chnique to detect human variables and their subsequent usage. This is a best te variable without defining it, a programm errors like programmer may use @ val defined a variable but don't initialised it and used it in predicate logic ner has [57] siaemmCalling function processArray(arr return 0; \ ) , | I 1 | | | I ee Coverage Test Cases for Concatenated 100P es and values. coverage, Wo test different array Siz To ensure full looP Loop Execution Test Case ID No loop ‘execution ¢1 (Zero Iterations) (gum & product not computed) gum = 6, Product = 5 Each loop rung 7¢2(Minimum Iteration) ce = 6, Product ='6 7ca{Normal Execution) | 3:{1.2 3} sum =6, Each loop runs 3 times TC4 (Negative Values) 3, {-1, 2, -3} Gum=-2,Product=6 | Each loop runs times TC5 (Large Input) 5, (1,2,,4,5} Sum = 15, Product = 120 | Each loop runs § times Concatenated Loop Coverage Analysis — (i) Zero Iterations Case : Ensures the function handles an empty array. (ii) Minimum Hteration Case : Ensures the loops run exactly once. numbers, (iv) Large Input Cases ; Ss : Ensures pet Tre performance and Correctness with largernt assignments | pesti@ ge 22 Ad yes 0) Data Flow re Bees corn 2.13.3 Advantages of Data FIOM god tes dl | _ ») petectsuninitialized VA ig ; pncy bY fos 4 . mproves test efficion’y ae ” code execution. incomion oP | : \ ical error™ . ‘ii) Hel ssidentify logic® a white Box Tes WHITE BOX 42. 13° itrequires i 45: 46 igses ON input-output functions: with reference to ment. 3. Itfocus values! 4, Testing is done user's functional require! end-users 5, Performed by testers or QA teams. itis done at early stage of " testing: Itfocusses on program 3. structure and data structure ea ; 4, Testing is done with referengg,, the code and standards use. 5, Performed by Developers Testers. ee Esecise ___@~7 a2 ih When to use Black-box testing? What do you mean by Black-box testing? Explain. Explain equivalence Class Testing by using example. Explain Boundary Value Analysis by using example. Write advantage; ind disadys - sting ages and disad ‘antages of black. box testi : 2 3 4. 5. Explain Decision Te fable-bi i i ; ased testing by using example. 7. What do you mean by white. 8 ft What are Needs of White-box box testing? Explain testing?oqutod peeuy 8 Busy) ‘WOH /) $9= de peay sind // ‘(nd)eed | | \((qu)yoozis)ooyjew = nd, IU! } (ougAoWwew plor i ajdwiexy | +(Miowaw pee,y “*6"e) pier seBuo} OU Ss} }! 48He ejqeea e Bush, (n/-p) palini Buea Jeyy pasn elqeHEA © { (enjen sayy Bujsn jnouyim)'e jo Zjed// OL = ejoljoay ig=e tut } (quewubissyjuepunpe: pIoA ajdwiexs uuBISsY -enjen snoinaid s}1 Bulsn e10jaq ejqewen op enien® 6 (p-p) esp nowy UoRUaPeH S14eHEA 4 { s)x OWNS // 4 ',P%,)HUNE ix quy y por op Buleg 1c ) Qajdurex=uoue pion ne bus Ugg jou sey yey elgeH?” pe! +a (98-9 |, 19Q) * (osp-a i. sea) (0 <9\ 0 ‘P%,)aund 939 989-9 17 '¢ + q=o1y | | I I ! I ‘(.u\@aneBeny, )yund } asye { ‘(,U\@AnIsog, auld 939 esN-d// }(O<4q)y qjoseay/ ‘z,e=qqui } (e 1uejduiexe pion ajdurex3 xa ‘Painsexe s} (ued na) Ye esn-oj-WoRIUYEp A1ane Jeu} sosnsuZ aBeseao0g syjed-Nd ily ssV yen ( z@ (esn-o) z.4=2 0 (esp) (01 ize uscg al i \ Initia = | 5; 1 abIeg eel I oid testFunotion(int x, int y){ ‘ a ! intz=x+¥ /] (Def of 2) ~ 1 8 Pr ! mn fa | f(z> ort // (P-Use of 2) \ 0, svi t printf(“Positive\n"); 1 I jelse { \ ! printf(‘Negative or Zero\n’); \ \ ations, | ) | resent i } pu Chain for 2: Feber 2 =X + ¥> g, P-Use: if (z> 0) (zis used for decision-making). le x} E Data FlowTesting checks if all variables are properly defined before use. Set ‘| | | 13.1 Types of Data F low Testing Coverage | 4, AlkDefinitions Coverage Ensures every variable definition is covered by at least one test case that reaches a usage point. Example : inta=10; // Def ofa printf(“2%q", a); // Use of a t where ais defined overt Test case should execute the statemen and used. 2 All-Uses Coverage Ensures every use of a variable is covered in test cases. r [591 58