software Development Proces 1.3.4 MYTHS mim P yths «bout software Joper should b ind software engineering in so There are many 1) 0 4 it, before go development community. Dev! ave to clear about go al sted below. | some of them J 4 : chanical hardwa «The cost of computers 1s lowe devices or manual -ystem. em Computer reduces sisk over mech: intcal systems. it is flexible. = Software is easy to change becauie oie of project but it requires proper care but the impact of change varies with th ithout introducing crop 4 lete re-verification anda op thin that, of electrom It is casy to edit .he source © Software requirements change, tis introduced. So making changes time at which i omp! is extremely difficult. Each changes required ct testing. | | = Computers provide greater reliability tha. ; It is true that software docs not fail in no unlimited time before it ‘wear out’. In previous days manual system errorig days we have software error as well. 4 the devices they replace. rmal situaiion and can execute was normal but nowat = Increasing software reliability will increase safety. It is true that as increasing software reliability, it increases test cases. Reliability is the probability of failure-free operation of software in ay fied environment. Finally it increases safety. igs of the software projeci to a third [Link] re relax you. But if an organization does nol re projects internally, it will rojects. You should have 10 speci = Outsourcin, lax the customers. t It is right that outsourcing understand how to manage and control softwar always struggle when it outsources software p: take care of cach phase of project. + Tesiing software and formal verification of software can remove all the errors. : ‘ojet! er Ol 1 > { i [Testing software or tools definitely find out presence of errors in pr d out maximum numb So design such effective test cases which can fin possible érrors) More we test, more we can trust of design and functionalitf | \ \ \ \ | ‘opment —— irements ts mance iterpret tive to ciness tform cation prised forms :s for for a ercial oject. n the Software Development Process 4 SOFTWARE ENGINEERING Software is there bef roblems i ere before software engineeri p ms in the cost, timeliness. ering. But there ar Software engineering solves the: 1.4.1 DEFINITION [Software engineerin i ig iS systematic approacl evel eee pproach to the development, operation and a son ¢) The term software engineering was defined at NATO science sponsored conference at Germany i i mn y in 1968, to discuss the growi ° « ; 5 ing softwa crisis and the need to focus on software development. . es At first conference on software engineering, NATO chairman Prof, [Link] defined the sofiware engineering as “The establishment and use of sound engineering principle in order to obtain economically developed software that is reliable and works efficiently on real machines.” Stephen Schach defined “A discipline whose aim is fo produce quality software that is delivered on time, within budget and satisfies requirements of sofiware.” [Software engineering body of knowledge defines software engineering as “Mis an application of systematic, disciplined and quantifiable approach to the development, operation and maintenance of software and the study of these approach." ] ogramming, {Ihe role of software engineering is often confused with pr eds and specify the “blueprints” the customer’s business ne plement it.) Software engineering is to capture for the system so that programmers can im, 1.4.2 NEED OF SOFTWARE ENGINEERING eady described in introduction section that 1? Software engineering 1S meth developing the software. Thi : as requirements analysis, Iding software Jementation why software engineering is odology provides the is framework defines Je is alr required? What its requiremen| framework that guides engineers 10 ae vcog of software development suc nent different phases of soft Se af and imp! i intenance etc. Th Sv amentation, maintenance ¢ ; testing, implementatl Od et «representation igni g, not th specification, designin rey wang, Stag, vuuruiaung, OUdgeting and directing software development activit 5 , 1.4.3 LAYERED TECHNOLOGY APPROACH a (software engineering is a layered technology. Software engineering encompasses process, methods and tools that enable complex computer system to be built in a timely manner with quality. To build any software product engineer should use and are off these four layers: Quality, Process, Methods and Tools, Figure 1.6 shows aw ( layered approach.rent | software Development Process =] = what dlete 5 TooLs that ing j METHOD are } are 5 PROCESS is) (QUALITY FIGURE 1.6 : SOFTWARE ENGINEERING LAYERS [The main focus of sofiware engineering is to develop quality product. Any engineering approach, including software engineering, must rest on quality attribute, Quality of product refers to the characteristics that engineer specify for the product, [In software development, quality of product encompasses the degree to which the output meets the functions and features specified in the requirement model] implementation follows the design, and resulting system meets its requirements andl performance. £ The foundation layer is process layer. It is the heart of the software engineering approach. It is intermediate layer which combine and together the technology layers! Process defines framework that establish the foundation for a complete engincering § process by identifying small activities. The intent is to deliver software in timely manner with quality to satisfy users./It deals with technical and management of software development. "| : The layer after process is method. 'It is a prescription for how to build the § software product.) Method includes different tasks such as user communication, a requirement analysis, design modeling, coding, testing and maintenance. Method gives the exact way to build the software! Method is a way to execute processes in propet below layers; method and Process.)Tool is any software or hardware device which ig Funda_Software Development \ 2014 \ 3vs seuavare, Few well-known software life eycle models || are discussed here, I 1.5.1 LINEAR SEQUENTIAL MODEL/WATERFALL MODEL re i The first process model which wa 5 z 2 3 oo g s g é = a g & S 7 must complete each phase before the next phase begins. That’s why it is also known as waterfall model, Sometimes it is also called classical life eyele model or conventional Process model, (a this approach software development begins with communication with customer and requirement gathering through planning, modeling, construction and end with deployment with successful customer delivery and feedback:JNo one phase can neither overteke nor overlap another one. In this model each phase has well defined starting and ending point, with identifiable deliveries to the next Phase,(This is only possible When requirements for a problem are well understood in advaned) (his approach cnforce “defines before design” and “design before code"\[It expects complete and accurate requirement early in the process, which is unrealistic. ") Evidence of failure in applying the waterfall comes from one of its most frequent users, the U.S. Department of Defense (DoD). Throughout the 1980s and into the 1990s, most DoD Projects were follow a waterfall cycle of development. DoD project's report concludes that 75 percent of the projects failed or were never used.Panning estimation & ‘scheduling re quivmen > adhe a c spalays? te 2) Ealeetae Spec dice AiO} . . a 3) desiguir uy Codi ! sy fests pethytin” oH The model consists of five dis given below. @ Communication phase : » Modeling (Des!9") truction (coding & Tostna) NTIAL MODEL maintenance) : LINEAR SEQUE! The working approac! (FIGURE 1-11 th of each tinct phases. = Requirements are gathered. . A a — F; * The problem is specified along with the desired service objectives (goals The constraints are identified. The system specification is produced from the requirement analysis. Thi document should clearly define the product function. ® Planning phase : Planning is a critical activ: the requirements) detailed require; ity in software dev of the system and should be done ‘ment is not necessary for pl, Cost estimation to find best affor Task scheduling and timeline mai Work Organization and team stru clopment. A good plan is based 9 before later phases begin. Howeve fanning) Tt includes activities like dable cost, ® yi . . nagement, cturelopmentl comware Development Process =) of each ig _-* Coding is done. es (goals! coding effort: sis, THS the software meets its required specification, $ based gy n. Howeigg After construction the software is delivered to the customer, The system is handed ike @ Modelling phase (the system specifications are translated into a software Tepresentation) Th, | software engineer at this stage is concerned with: “4 = Data structure F, Software architecture = Algorithmic detail and /® Interface representations { Cithe hardware requirements are also determined)at this stage along with a picture } of the overall system architecture. By the end of this stage the software engineer } should be able to identify the relationship between the hardware, software and the { associated interfaces. Any faults in the specification should ideally not be passed | down-stream’, 1 @ Construction phase } .-" The implementation and testing is done. The designs are translated into the software domain. 7 ® Detailed documentation from the design phase can significantly reduce the Testing at this stage focuses on making sure that errors are identified and All the program units are integrated and tested to ensure that the compleie system meets the software requirements, ® Deployment phase : over to the customer and will be used the first time by them. {All the problems which 4id not arise during the previous phases will be solved\in this last phase. Maintenance Part is largest and inever ending phase of software development, In this phase the Software is updated to: “< Meet the changing customer nec —* Adapted to accommodate changes in the external environment. j 1 Fonda_sonwareDevelopnient \ 2014 \ 4 ‘26 Fundamentals of Software D: _-* Correct erroi Versi; viously ing phases. ' 1s und oversights previously undetected in the testing ph = Enhanci ¢ efficienc; oftware. ing the efficiency of the s : 7 are Observe that feedback loops ops allow for correcti i into the | 4,5, : ctions to. be .incorporated i : model, When changes are made at any phase, the relevant documentation ae { : updated to reflect that change. @ Advantages : : ‘oft _™ Easy to explain to the user. eS _-* Stages and activities are well defined. and _s Helps to plan and schedule the project than haphazard approach. ie «Verification at each stage ensures early detection of errors: / \2 misunderstanding. 1 * Testing is inherent to every phase. { the | j the documentation is produced at every stage. + int | ee It is documentation driven, that is, @ ‘Disadvantages : » [Iris difficult to er must be able to express their requirem define all requirements at the, beginning of @ project}: ents completely, correctly * Custom: : and clearly. Delays can occur in designing, coding and testing, entation, and at the end of project large red, so it is time consuming. ntil the end of project, so the cl has been coded. jivered in one shot at the } Each phese produces docum integration and testing is requi ustomer can ations can occur = No demonstr ersion of the product after it only sees a working ¥ «It follows ‘big-bang’ approach; entire system is del sion. It last. It produces heavy risk. y follow this approach. Changes can cause confu we Real projects rarcl orporate any kind of risk assessment. | does not inc jted for Short duration and routine type OF ¢ clear idea is present, derstood)in advance, ti Tf the | ding to next phas ! @ Application + The linear sequential requirements are of each phase is | model is well su well un frozen before procee’ projects where all and deliverablespata teks 4) bY F ; ot Soe ! G Prototype is a worki ‘odel that is functi | of the product | Actually Pro ing m lolype is a usable Proj (iis actually a pilot or test , mode! ionally equivalent to @ component ram but not a final software Product. \n goal of Prototyping model js -Mnstead of freezi irements before any designing, firs; Working Prototype ii ie actual product. And this Prototype Teview to help understand the i \. tequiremen: the Processing needs, and the Outpt ts, the Prototyping model may be em; 0 increase the flexibility of the development pr Communication (Requirements) Planning (Estimation Refinement of Rogitiromont as por Suggestion Modeling (Design) Construction (Couing & testing) Deployment (Dalivery, foodbook, maintenance) FIGURE 1,12 ; PROTOTYPING MODELFIGURE _ Business Moding i ily 28 2 5 2 a i3 {i oj wind ————____ereeeerereseewrrrrm_n—ss—— 4 Cumulative cost Ts Progress through stops: Evaluate alternatives, Identify, resolve risks Determine objectives, ailernatives, constraints, Risk analysis Risk analysis Risk analysis / sk [i Inger | Protos Protaype, \ Proraype Review Commune! i prototype partition T [i at =] Simulations, models, Benchmarks p \ \ Mo-oycle “wlan Concept of ce ae ; Sottwaro , NN requirements / Sottwore / Delated ! = product lesign Devens Bagutomenty a “To any Intagrasei Dusicn validation fol test | cag ¢otieation imal. tan acon Plan noxt phases Dovetop, verily suxt-leval product PIRAL MODEL + URE 1.15Software Development Process | 13 4 Product and Process : } (what is delivered to the customer is c:lled a product. It may include source code, specification document, manuals, [Link] etc. Basically, it. is nothing but a set of deliverables only.] [Process is the way in which we produce software or product. It is the collection of activities that leads to a product. An efficient process is required to produce good quality products) If the process is weak, the end product will undoubtedly suffer, but an obsessive overreliance on process is als) dangerous. Ue? Measures, Metrics and Measurement : The terms measures, metrics and measurement are often used interchangeably. I is interesting to understand the difference amongst these. [Measure provides a quantitative indication of the level, dimension, size, capac’ efficiency, productivity or reliability of some attributes of a product or process.) [Measurement is the act of evaluating a measure, (Metric is a quantitative measure of the degree to)which a system, component, of process possesses a given attribute. sie alk: dew bon] ld Geeteislanioaieie tial ices: | how@ measure of de, modules with highly coupled are g other, Two modules with low Coupling are not is measured by number of interconnection bet with high to low is shown and desc: BICE Of interde ; Pendency be; trongly intere by onnected and dependent on Ween modules, tibed below. | High coupling between module: understand but also increase development effort as they another modules, so low coupled modules are best, Data Coupling Lowest Coupled (Best) Stamp Coupling te ees Control Coupling External Coupling Common Coupling Content Coupling Highest Coupled (Worst) TABLE-2.2 : LEVEL OF MODULE COUPLING Data Coupling :\The dependency is said to be data is based on the fact they communic modules are independent coupled if their dependency ate by only passing of data. \Other than data, two Int Float hud Stamp Coupling data structure is Passed fr he dependency is said to be stamp coupled when complete ‘©m one module to another module,)It is not necessary to Send all data structure, If calling module requires only part of data structure, called Module just Pass that part Control Coupling Tie dependency is said to be contro! coupling when modules ate communicate by pass: ing of control information) This is usually accomplished by Means of flags that are set by onc module and reacted upon by, the dependent module. External Coupling mal to the software b, “sically related to the co Common Coupling ; ommon data affects Un this, a module has a dependency to other module, cing developed or to particular types of hardware) [This is mmunication to external tools and devices.) eXtey ao ing change In this two or more module has shared data. Tae a all modules who share this common data] It is didetermine whi ch module 87 Content Couptin ie onsible for having set a value to a particular variable. Passed from s B *LOne module changes data of other module and control is : modul - coupling, -) le to’ the middle of another. Content coupling is highest ® COHESION (Cohesion j iS 2 meacuc. . functionally related of the degree to which the elements of the module are together Con Uted.|Cohesion may be viewed as glue that keeps the internal modul * SOvesion is strength within a module. It represents how tightly bound the internal cleme: t oe ents of the module are to one another, There are several leveis of Ohesions. Each are describe the elements of the module, d and shown below with degree of relationship between Functional Cohesion High (Best) ee | Sequential Cohesion Communication Cohesion Procedural Cohesion Temporal Cohesion Logical Cohesion Coincidental Cohesion Low (Worst) TABLE-2.3 : LEVEL OF MODULE COHESION Functional Cohesion : jit is strongest cohesion, In a functionally bound module, all the elements of the module are related (o performing a single function) All modules accomplishing a single goal by mean of function. Sequential Cohesion 1 When the elements are together in u module because tie output of one form is the input to another, we get sequential cohesion.) However, sequential cohesion does not provide any idea on how to combine them into modules. [Module may contain several functions or part of function. ) * Communication Cohesion :{When multiple modules operate on same input data or same output data, it is called communication cohesion modules on those modules are separate. This module may perform more’ 3) But the procedures an one function.Fur damentals of Software D velopment 68 Procedura’ Coh sion :/A procedural [Link] contains elements that belong toa mon _proce'lural unit.\This module often occurs when modular structure is comi ———ae determined from some form of flowchar!. A msdule with procedural cohes:on contains jeter i only part of a complete fur tion or parts of several functions, Temporal \ohesion : l[t is same as log'cal cohesion, except that the elements are also felated in time and are executed together] Modules that perform activities are usually tem orally bound, That means it performs al" activities in same time span. Logical Cohesion :[A moduli has logical cohesior. if there is some logical relationship be-ween the elements of module, and the elements that fall in the same logical class) In Beneral, logi avoided, if possible, Coineiden Cohesion : {It is low. folationship other than shared data: Thi relationship among the elements ofa far as possibl 20 MATAMONETInA A perform functions Shesion modules should be est level than the rest, There is no conceptual is cohesion occurs when there is no meaningful module. [Coincident cohesion is avoidable asFundamentals of Software Development ate | Tremory size, and defects reported. Indirect measures include functionality, quality, complexity, efficiency, reliability, and maintainabilit The contrast between direct measurement and ct or derived measurement is between a (direct) metric function whose domain is only one variable and a (derived) function whose domain is an n-tuple. For example, density is a function of mass and volume. Direct measures are generally casier to collect than indirect measures. Size- oricnted metrics are used to collect direct measures of software engineering output and quality. -Function-oriented metrics provide indirect measures. ¢ @ SIZE-ORIENTED METRICS - 1 7,9 - @3 {si oriented metrics are direct measure of software and the process by which it was developed. These metrics can include effort (time), money spent, KLOC (1000s lines of code), pages of documentation created, errors, and people on the project.) For example, two projects have been developed and organization maintains simple size oriented record. Table of size-oriented metrics can be created as below: Project] LOC | Effort] Cost (Rs.)| Page Documents| Errors | Defects! People ABC {17,907 38. | 19,225 795 ao | ot | 3 xyz fosais] 60 | 15,775 1212 325 Be 5 TABLE-3.1: SIZE-ORIENTED METRICS From these data some simple size-oriented metrics can be generated. Productivity = Errors / KLOC, Productivity = LOC / person-month Quality = Defects / KLOC, Quality = Frrors / person-month Cost = Cost / LOC, Cost = Cost / pages of documentation Documentation = Pages of documentation / KLOC The oldest metric for software projects is that of “lines of code” (LOC).)This : was first introduced circa 1960 and was used for economic, productivity, and quality studies. The definition of LOC by most researchers is }“A line of code is any line of i ; ‘Program text that is not a comment or blank line, regardless of the number of Statements or ftagments of statements on the I That means all lines containingsoftware Project Management ion and executable and non. ‘executable Source program header, declar: statements”. [Size-oriented metrics are not universally accepted.\The use of LOC a, measure is the center of the conflict Supporter of the LOC measure Claim; * = It is a man-made measurement of all software engineering Processes can easily be counted. ich * Many existing metrics exist which use LOC as an input, * A large body of literature and data exist which is predicated °n Log, Opponents of the LOC measure claim: * That it is programming language dependent, = They penalise well-designed but shorter programs. "They do not work well with non-procedural languages, ® Their use in planning is difficult because the Planner must estimate Loc before the design is completed, @ FUNCTION-ORIENTED METRICS Fp Function-oriented metrics are indirect measures of Software) which focus on functionality and utility, Functionality cannot be measured directly; it must be derived indirectly using other direct measures, IBM in 1979s, He developed a technique which appeared to. be a solution to the size measurement problem. He suggested/a productivity measurement approach called the im function point method.}Function points (FPs) are derived from countable measures a and assessments of sorfware complexity: | | \ The first function-oriented metric Was proposed by Albrecht while working for i ! i 1 | ues are number of inputs, number of outputs, number of inquiries (on-line inputs), number of files, number of external interfaces (machine Teadable interfaces - tape, disk),118 _2/ Number of inputs: |These are Project input that provides distinct application. oriented data to the software. Inputs as inquiries are counted separately, ) J Number of outputs:(These are project Output that provides application. Fundamentals of Software Development | oriented information to the 1 cer} In this context output refers to reports, screens, TOF Messages, ete These are on-line inputs that results in the generation Number of ingui of some immediate software response in the form of aa online output) Number of files: These are logical master files)e.g., a logical grouping of data that may be one part of a large database or a separate file. Number of external interfa “These are all machine readable interfaces ) (eg., data files on Storage media) that are used to transmit information ‘9 another system “software this mea accomm time, pr produce Orn olaim:") a i) OpponeSV AS the estimation variable (either LOC or FP) ® The COCOMO Masa: Man4? Taal Oona Con, WO’: Dr, Berry Boehm, has found this model oe aii be ‘Bochm’s book (COCOMO). This model gained rapid popularity after pul i Q (variable) ¥ ‘ic, single-valued i 2 ies? model is a static, si ‘Software Engineering Economies [This mo124 _—__ Fundamentals of Software Development —____ model that computes software develo oe eemeninls of Gofware Development: size expressed in eda feet effort (and cost) as a Tuneton of program fine ue ines of code (LOC), Dr. [Link] has classified Pirieal estimation model. [This model is widely used in industry.) COCOMO may be applied three modes of software project. J Organic Mode: software in Project is aj In this sma team of experienced developers develops st very familiar e wvironment, It is relatively simple and size of P sroximately 2-5(+ KLOC. e.g,, payroll. inventory project, etc. f) Semidetached Mode: 1 js intermediate mode of organic and embedded system mode. Team size is medium. It contains average experienced person and mixed type of requirements, It is medium size project. approximately 50-300 KLOC. e.g., compiler, database system, editors, etc. a Embedded Mode: In this project constraints are tight, which might be related to target processor and its interface with the hardware. The size of project is large and it is real time system. Ver for thi little previous experience is required | The size is of approximately over 300 KLOC, e.g., any real time | system, aircraft control system, automatic medical instruments etc. | 1 { | \ { | {COCOMO is hierarchy of software cost estimation model Which includes basic, intermediate and detailed sub models. Basic COCOMO takes an equation form: ( Effort E = a x (KLOC)? Per month; Duration D = ¢ x (E)4 Per month; Number of person N = Effort/Duration; Productivity = KLOC/Effort. ode Project Size | Nature Innovation| Development | Environment Organic | 2-50 KLOC | Small size project, Little Familiar for Experienced developer, In-home. Familiar environment, For ex. payroll, inventory Semi 30-300 KLOC| Medium size project, Medium | Medium Detached Medium size project team, Average experience required, For ex. utility system, compile:7? @@ Risk Mitigation, Monitoring and Management (RMMM) Plan ey a“ Lita software team follows a proactive approach to risk, avoidance is always the best strategy. This is achieved by developing a plan for risk mitigation, To mitigate this risk, project management must develop a strategy for reducing losses. Among the possible steps to be taken are; wy a 4 “a a Meet with current staff to determine causes for loss (e.g., poor working conditions, low pay, and competitive job market). Mitigate those causes that are under our contro] before the project starts. Once the project begins, assume loss will occur and develop techniques to ensure continuity when people leave the project. Organize project teams so that information about cach development activity is widely distributed. Define documentation standards and establish mechanisms io be sure that documents are developed in a timely manner. Conduct peer reviews of all work (so that more than one person is “up to speed”). Assign a backup staff member for overy critical technologist. As the project proceeds, risk_monitoring activities commence, (he project manager monitors factors that may provide an indication of whether the risk is becoming more or less likely) In the case of high staff turnover (staff loss), the following factors can be monitored:Software Project Management tm Gen attitude of team members based on project pressur: » The degree to which the team has jelled, ae + Z* Interpersonal relationships among team members. "Potential problems with compensation and benefits. the availability of jobs within the company and outside it. | w In addition to monitoring these factors, the project manager should monitor the effectiveness of risk mitigation steps. The project manager should monitor documents carefully to ensure that each can stand on its own and that each suggests information that would be necessary if a newcomer were forced to join the software team somewhere in the middle of the Project. [Risk management and contingency planning assumes that mitigation efforts have failed and that the risk has become a reality.| If the mitigation Strategy has been followed, backup is available, information is documented, and knowledge has been distributed across the team. In addition, the project manager may temporarily refocus ‘sources (and readjust the project schedule) to those functions that are fully staffed, enabling newcomers who must be added to the team to “get up to speed” (Those individuals who are!leaving are asked to Stop all work and spend their last weeks in “knowledge transfer mode”, This might include video-based knowledge capture, the development of “commentary documents,” and/or meeting with other team members who will remain on the project. Ut is important to note that RMMM plan increase additional Project cost cause of these activities,Testing Testing : TYPES OF TESTING | FIGURE 4. 4.4.1 WHITE BOX TESTING Out of two approach of testing, [white box testing is concerned with testing the implementation of program, The intent of this testing is not to exercise all the different input and output condition but to exercise/the different internal programming structure and data structure used in program. That’s why it is also called structural testing. It is also known as glass box testing.\Using white box testing, tester can derive test cases that,) * Guarantee that all independent paths within a module have been exercised at least once. * Exercise all decision whether it is true or false. «Execute all paths within their bounds and their boundaries. * Exercise internal data structure to ensure their validity. ‘Tf you ave. WIND A cadets MAE Eoyd Working of Outpus wes cabeulWhite Box Testing = To-hwieuay i Contro} ey Site en Independent Cyclomatic Graph Condition Data Foy > Path Fiow Graph 0D Falh Complexity Matix Testing sTeau Notation FIGURE 4.3 : WHITE BOX TESTING METHODS @ BASIS PATH TESTING ®€ | Basis path testing is a white-box testing technique first proposed by Tom Moc: The basis path method enables the test case designer to derive a logical Complery measure of a procedural design and use this measure as a guide for defining a bs set of execution paths. Test cases derived to exercise the basis set are guaranteed » execute every statement in the program at least one time during testing. ! It requires complete knowledge of the program’s structure and used by develoges to unit test their own. (This type of testing involves (i) generating a set of paths tt will cover every branch in the program, (ii) finding a set of test cases thal wil execute every path in this set of program. \These two steps are not necessary to exe: in sequence. ® Flow Graph Notation (the Control flow of the program can be analyzed using a graphical represent 2 OO } pee swential — Itthen-else or L00P FIGURE 4.4 ; While Loop switch Statement FLOW GRAPH NOTATIONof Software Development ‘he flow graph is a directed graph in which nodes are either entire statement or fragments of statement. The flow graph depicts logical control flow using the notation illustrated in Figure 4.4. Each structured construct has a conesponding flow graph symbol, The bellowed Figure 4.5 illustrates how the flow chart maps into flow graph using specific notation. Each circle, called a flow graph node, represents procedural statements. Number of processes and a decision procedure can map into a single node. The arrows on the flow graph, called edges, represent flow of control and are equivalent to flowchart arrows. Areas bounded by edges and nodes are called regions. Area outside the graph is also counted as a region. FIGURE 4.5-: FLOW GRAPH CORRESPONDING FLOW CHART ? DD Path Graph : In path testing afier flow graph, next step is to draw a DD path graph from the fev graph. DD path means decision fo decision path graph.) The nodes of an aa : . a s ) bh, which are in sequence is combined into a single node. So, /DD path aa Mircted graph in which nodes are sequences of statement and edges represent Col fevbetween node.ae 175 se Software Coding And Testing E46: DD PATH GRAPH FIGURI The DD path graph is used to find Tadependent paths. The flow graph nodes which are in sequence are forming on independent hat each routed at least once. ath, It ensures U independent path has been exe Wo i : 8% @ Cyclomatic Complexity 15 °° fu in W-8-T Iso known as structural complexity peeause it ity is al d to find the number of ‘The cyclomatic complex! uted by one of view of the code. gran his approach is use! complexity is comp! gives internal independent paths through a pro! three ways: (a) The number of regions in any flow graph. exity VG) = B-N +3 where E is number of edges Lb) Cyclomatic Compl and N is number of nodes atic Complexity VIG) =P * 1s ow graph G, (Diamonds ar ec cyclomatic complexity ca where P is number of predicates (c) Cyclom ¢ decision nodes) (decision) node in Ml If we refer Figure 4.5, th n be computed using each of three ways is: (a) Total numbers ®) Cyclomatic Com of regions are 4. plexity (6) = exity V(G) = P+ 1 = 3 predicate N+2=11 edges - 9 nodes +2 =4 (decision) nodes + 1 = 4 VO Cyclomatic Compl Several properties of cyclomatic complexity are as below: ) wor)Fundamentals of Software Development 276 CC ii) V(G) is maximum number of independent path in graph G. iii) Inserting and deleting functional statement does not affect V(G). iv) Inserting a new row in graph increases V(G) by unity. v¥) V(G) depends only on the decision structure of G. @ Graph Matrices As a tester we are interested to find out independent paths. The objective of ing is not easy task and is subject testing is to trace all paths at least once. Path tra to error, If the size of graph is very large, it is difficult to tracing manually. For that it is advisable to use tool. To develop such a tool, a data structure, called graph matrix can be very useful tool. (A graph matrix is a square matrix whose size is equal to the number of nodes on the flow graph?\Each row and column corresponds to an identified node, and matrix entries correspond to connections between nodes. A simple example of a flow graph and its corresponding graph matrix is shown in Figure 4.7. Connection Connection Node 4G Nog dB 1 8 1 1 1-1-0 2| bla 2 1]4 24=1 3 efe 3 a [4 fester 4 4 5 ' afr 5 1 1 [4 [atez ViGeaH=5 Flow Graph Graph Matrix Connection Matrix FIGURE 4.7 : GRAPH MATRIX Each node of the flow graph is identified by numbers, while each edge is identified by letters, A letter entry is made in the matrix to correspond to a connection between two nodes. The graph matrix is redrawn as connection matrix, Each letter has been replaced with a 1, which is called link weight. The link weight provides additicnal information about control flow. Simply, the link weight is 1 if connection exists or 0 if connection does not exist. @ CONTROL STRUCTURE TESTING [Although basis path testing is simple and highly effective, it is not sufficient in itself.’ Here, other variations on control structure testing)are discussed, These expand