GENERIC CODE CLONING METHOD FOR DETECTION OF CLONE CODE IN SOFTWARE DEVELOPMENT

GENERIC CODE CLONING METHOD FOR DETECTION OF CLONE CODE IN SOFTWARE DEVELOPMENT

• 1.
  http://www.iaeme.com/IJCET/index.asp 14 editor@iaeme.com InternationalJournal of Computer Engineering & Technology (IJCET) Volume 8, Issue 2, March-April 2017, pp. 14–22, Article ID: IJCET_08_02_003 Available online at http://www.iaeme.com/IJCET/issues.asp?JType=IJCET&VType=8&IType=2 Journal Impact Factor (2016): 9.3590(Calculated by GISI) www.jifactor.com ISSN Print: 0976-6367 and ISSN Online: 0976–6375 © IAEME Publication GENERIC CODE CLONING METHOD FOR DETECTION OF CLONE CODE IN SOFTWARE DEVELOPMENT Syed Mohd Fazalul Haque Maulana Azad National Urdu University, Hyderabad, Telangana, India V. Srikanth K L University, Guntur, Andhra Pradesh, India E. Sreenivasa Reddy Acharya Nagarjuna University, Guntur, Andhra Pradesh, India ABSTRACT The major part of risk the development of software or programs is existence of duplicate code that can affect the software maintainability. The main aim of Clone identification technique is to search and detect the parts of the software code which is identical. In the passed there are various techniques that are used to identify and reflect the code identity and code fragments. Code cloning reduces the time and effort of the software developer but it also decreases the quality of the software like readability, changeability and increases maintainability. So, code clone has to be detected to reduce the cost of maintenance to some extent. In this paper, a new Generic technique is purposed to detect code clone from various input source codes (from web, disk and etc.,) by segmenting the code into number of sub-programs or modules or functions. I propose a technique that can detect 1-type, 2-type, 3-type and 4-type clones efficiently. Key words: Generic technique, clone, threshold, code and methods. Cite this Article: Syed Mohd Fazalul Haque, V. Srikanth and E. Sreenivasa Reddy, Generic Code Cloning Method for Detection of Clone Code in Software Development. International Journal of Computer Engineering & Technology, 8(2), 2017, pp. 14–22. http://www.iaeme.com/IJCET/issues.asp?JType=IJCET&VType=8&IType=2 1. INTRODUCTION In software Development cycle, Many of the software Developers makes a copy of pieces of code and paste without making any changes like naming , updating or removing etc. to eliminate time and efforts. This type of work doing the process of copying and adding the request code in the software development is called as code Cloning. Reusability has become
• 2.
  Generic Code CloningMethod for Detection of Clone Code in Software Development http://www.iaeme.com/IJCET/index.asp 15 editor@iaeme.com common in recent software development by the developers, but it has various drawback[14]. It leads to increase in maintenance cost [16] which also decreases quality of the software in a system [1]. As the code is copied without alternations , there is a chance of increase in bugs in the software, because errors present in one module can be increase the errors in another , when they are linked with modules increase bugs in software system[11][8]. According to review, about 10% to 25% of code in the software development by the developers coping and pasting of code i.e. cloned code, [4] [2] and around 75% of cloned code is tailored clone code i.e. code copied and pasted with some changes like naming, updating or erasing of the instructions to the code[10]. The developers do this process regularly for rapid development of the software which can increase bugs and alter the execution. The developer usually do the process of cloning of code wontedly or unwontedly, A time the code which was written and copied intentionally by the software engineer by adding the code fragments with or without changes, can cause a bug ,a times the code cloning can serve as advantages in reusing the code which was efficiently written, can increase the effectiveness of the software very high. We come across two type of such cloning methods namely semantic and syntactic matching to find similarities. If the word or text of the code matches, then it is syntactic and if the module or procedure matches it is semantic identity. 1-Type: find the syntactic similarities, these allow only certain information’s like spaces and remarks 2-Type: These are naming changing clones with same syntactic, in this only changes is allowed in variable, constant, type of data type, spaces and remarks. 3-Type: clones with identical syntactic similarities, in this clone variation are allowed to rename literal, identifier and addition or deletion of statement to code. 4-Type: These are semantic clone with semantic similarities; these clones are semantically same but syntactically differenti. e. computation is same but implement by different syntactic variants. 2. CLONE DETECTION PRADIAGMS Text Based Approach- In this method the duplicate or copied code is detected line by line in the form of group of characters [4]. The approach is very easy to write, this approach only solves 1-type clone. Token Based Approach: In this method, the code source has to be transformed into smaller tokens by using lexer or parser[15]. The comparison is done line by line, but in the form of tokens[4]. This approach only detects 1-type and 2-type clones [15]. ASTB Approach: In this method first the code source is converted into Syntax tree using parsing [1][8]. The comparison in this approach is done using sub-tree of syntax tree. This approach is very difficult to write, it can detect only 1-type, 2-type and 3-type clones. Implementation of the ASTB is very tedious because to convert a source code into tree is difficult. PDG Approach: In this method when applied, converts code sources into direct graph structure. This methods solves only 1-type, 2-type and 3-type, Implementation of PDG is very difficult. Metric Based Approach: In the method, the code source is transformed into another form. After conversion it metric values are evaluated from the code, like number of classes, number of objects, number of lines, number of function calls, number of branching statements, number of loops, number of comments etc. Then the metrics of two source codes are
• 3.
  Syed Mohd FazalulHaque, V. Srikanth and E. Sreenivasa Reddy http://www.iaeme.com/IJCET/index.asp 16 editor@iaeme.com compared and clone will be detected [10]. This method is easy to write but very difficult when they are complex. 3. PROPOSED METHOD AND IMPLEMENTATION Proposed work is divided into two categories 1. Prepare an ontology Schema 2. Develop a Generic algorithm for Clone Detection using Ontology 3.1. Ontology Schema Table 1 Showing the Ontology used in Code Cloning We build a model table 1.Of ontology which captures clones, clone objects, clone sets like program variables, relationship between variables. The ontology elements include class instance, interface, functions, methods, variables and fields. The type of relationship provided between variables are implements, extends, declare and data types, association, aggregation , generalization , In added to the schema , there are other 3 specific clone type relations namely diffuse, reside in and contain. These associations describe the relationship between clones and clone fragments. Using the above ontology, code objects are detected and cloning relationship is found very easily. The figure 1. shows the Ontology Schemes relationship between cloned objects, objects sets, program variables, class instance, interface, functions, methods, variables, for loop, while, aggregation, associations and generalizations. 3.2. Ontology Schema Structure Ontology is a specification of a conceptualization that is designed for reuse across multiple applications and implementations. …a specification of a conceptualization is a written, formal description of a set of concepts and relationships in a domain of interest. It guides in forming relationship between the objects, clone objects and object sets. 3.3. Working of Ontology Schemas It helps to guide to form relations, associations, aggregation and generalization between objects and object sets in code Subject Relation Object Clone set Contain Clone instance Clone instant Reside_in Method Clone instance Diff_use Function, field, variable, class, interface, Clone instance Common_use Variable, loop, functions, class, interface Class Extend Class Class Implement Interface Class Declared_in Class Interface Extend Class Method Declared_in Interface Method has_return_type Class/interface Field Has_type Interface Field Has_type Class/interface Field Has_type Method, for, while , class, interface
• 4.
  Generic Code CloningMethod for Detection of Clone Code in Software Development http://www.iaeme.com/IJCET/index.asp 17 editor@iaeme.com Figure 1 Showing an Ontology Schema Structure 3.4. Algorithm for Clone Detection Generic approach is the combination of more approaches or methods. This approach can enhance the advantages of Generic code clone detection methods. The proposed work is based on Generic approach. Steps for the proposed tool are described below. Step 1: Open two files which has to be checked as source files Step 2: Modulate the files into smaller sub-programs based on class, interfaces and function in the program Step 3: Computer the modules based on metrics of both program based on like line numbers, number of classes , functions in the program , loops in the program , number of classes , number of interfaces, number of structures, number of arrays, number of variables etc. Step 4: Evaluate metrics to find out the potential of cloned code Step 5: Evaluate Template on latent renovation on potential clones in code Step 6: Perform word based comparison. If the quantity of line matched exceeds the threshold value, then it is called actual clone’s code.
• 5.
  Syed Mohd FazalulHaque, V. Srikanth and E. Sreenivasa Reddy http://www.iaeme.com/IJCET/index.asp 18 editor@iaeme.com Figure 2 Flow of the proposed Algorithms My proposed method mainly aims on modulating the code into number of classes, methods inside the classes. Then calculate the values of Metrics for number of lines of code, number of do-loop or for loop or while loops in a method, number of objects in a class, number of methods involved. Then the metric values of the two sources files are compared for identification of clones from the source with other source code. After that the parser or lexical transforms the clone into template convertor. The working of the parser or lexical is similar to the working in compiler design, instead it does not tokenize the code. After the template is converted, then char or string or words identification is done. Based on the number of lines matching the threshold percentage, value of the code done is detected. My technique detects 1-type, 2-type, 3-type and 4-type. The result of the work can be shown manually by selecting two source input files. An open dialog tool is used for selecting the input files. Any type of generic source file can be compared, but both the files should be similar type. After including the two files, Click on the icon of clone detection process, Then clone detection process is initiated, the below figure show the clone detection window figure 4. File selection window figure 3 and browsing of file window figure 2.
• 6.
  Generic Code CloningMethod for Detection of Clone Code in Software Development http://www.iaeme.com/IJCET/index.asp 19 editor@iaeme.com Figure 3 A dialog window to select files Figure 4 Partition Screen of code1 code 2 and detection window Figure 5 Code Clone is displayed 4. RESULT AND DISCUSSION The work done will detect 1-type, 2-type, 3-type and 4-type very efficiently. I developed a tool in .NET frame work using C# version 5.0 used. It provides user interface which is shown in figure 4. The tool that is developed to identify clones using metric approach like number of classes, number of variables, number of methods, number of objects , number of function calls, number of loops etc. The potential of the tool is to detect the cloned code effectively
• 7.
  Syed Mohd FazalulHaque, V. Srikanth and E. Sreenivasa Reddy http://www.iaeme.com/IJCET/index.asp 20 editor@iaeme.com based on converted template and char, textual or string based check is performed, Then the threshold is matched with the code of actual is done. 4.1. Performance Evaluation Precision and recall were the two important considered with comparing the clone detection techniques. below shows the description of the precision and recall factor equations. Precision: It finds out the good enough and find out less number of false, i.e. find duplicated code with high rate. Precision = Number of correct detected clones / number of clones detected. Recall: Calculates most and effective clones in a software Recall = number of correct detect clones / number of possible existing clones in software. The performance evaluation is calculated based on the code which is used for testing the various types of C, C++ and JAVA program code. The percentages are Table 2 Difference shown based on recall and precision of clone testing. Tool Number of clones Percentages of clones Generic Code Cloning 957 7.36% Hybrid Code Cloning 823 6.33% Parse based Cloning 783 6.02% Syntactic cloning 797 6.13% Table 3 Precision and Recall Values Technique Precision Recall Generic code cloning 0.48 0.60 Hybrid code cloning 0.57 0.55 Parse Tree cloning 0.44 0.49 Syntactic cloning 0.83 0.50 A technique or method with highest value in precision and Recall is treated as better tool, The table 3 show above, gives the various value of precision and recall calculated. From the above table, Generic code cloning method is considered as the better compared to others. Figure 6 Clone % graph for various clones’ methods 0 200 400 600 800 1000 1200 1 2 3 4 Percentages of clones Number of clones
• 8.
  Generic Code CloningMethod for Detection of Clone Code in Software Development http://www.iaeme.com/IJCET/index.asp 21 editor@iaeme.com The above figure 6. Shows the various techniques used of program code detect with percentage are show, from all the above , the height clones that can be detected accurately and efficiently is the Generic cloning techniques which can detect 1-type , 2 –type , 3-type and 4- type clones based on the syntax, semantic and etc. Figure 7 Show the precision and recall values of methods 5. CONCLUSIONS AND FUTURE SCOPE The method that has been developed used Generic method which combines various techniques for mapping and matching the code from source with respect to other. The developed tool can detect 1-type, 2-type, 3-type and 4-type very efficiently. This tool can be further can be further enhanced for better complexity and efficient detection. REFERENCES [1] Rajkumar Tekchandani, Rajesh Kumar Bhatia and Maninder Singh, “Semantic Code Clone Detection Using Parse Trees and Grammar Recovery”, pp.41-46, IEEE, 2013. [2] AmandeepKaur and Balraj Singh, “Study on Metrics Based Approach for Detecting Software Code Clones”, International Journal of Advanced Research in Computer Science and Software Engineering, Volume 4, Issue 1,January 2014. [3] KanikaRaheja, RajkumarTekchandani, “An Efficient Code Clone Detection model on java byte code using hybrid approach”,Page 16-21, IEEE, SEPT 2013. [4] Geetika, Rajkumar Tekchandani, “Detection of Potential Clones from Software using Metrics”, IJARCSSE, Volume 4, Issue 4, April 2014. [5] Deepak sethi, Manishasehrawat and Bharat BhushanNaib, “Detection of code clones using datasets”IJARCSSE, Volume 2, Issue 7, july 2012. [6] Yoshiki Higo, Yasushi Ueda, Minoru Nishino, Shinji Kusumoto, “Incremental Code Clone Detection: A PDG-based Approach”, Page 3-12, IEEE, 2011. [7] Mai Iwamoto, Shunsuke Oshima, Takuo Nakashima, “Token-based Code Clone Detection Technique in a Student’s Programming Exercise”, Page 650-655,,IEEE, 2012. [8] TahiraKhatoon, Priyansha Singh, ShikhaShukla” Abstract Syntax Tree Based Clone Detection for Java Projects” IOSR Journal of Engineering,Volume 2,Issue12,Dec 2012 [9] MikkelJonsson Thomsen, Fritz Henglein,”Clone Detection using Rolling Hashing, Suffix Trees and Dagnification:A Case Study.”IEEE,2012. 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1 2 3 4 Recall Precision
• 9.
  Syed Mohd FazalulHaque, V. Srikanth and E. Sreenivasa Reddy http://www.iaeme.com/IJCET/index.asp 22 editor@iaeme.com [10] PriyankaBatta, Miss Himanshi, “Hybrid Technique for software code clone detection” IJCT, Volume 2 no. 2 April 2012. [11] Kiranpreet, SushilGarg, “Detection and measuring similarity in code clone using ripley’s function Approach.”,IJAST,Volume 2,issue 4,dec 2014. [12] RubalaSivakumar,Kodhai.E, ”Code clone detection in website using approach ”, IJCA, Volume 48-No. 13,June 2012. [13] Balwinder Kumar, Dr. Satwinder Singh, “Code clone detection and Analysis using Software Metrics and Neural Network- A Literature Review”,IJCST, ,Volume 3,issue 2, mar-apr 2015. [14] Prajila Prem, ”A Review on code clone analsis and code clone detection ”, IJEIT , Volume 2, issue 12, june2013. [15] G. Anil kumar, Dr. C.R.K.Reddy, Dr. A. Govardhan, An Efficient Method-Level Code Clone Detection Scheme through Textual Analysis Using Metrics. International Journal of Computer Engineering and Technology (IJCET).3 (1), 2012, pp. 273-288. [16] Prof. Deepika Shukla and Apurva Desai, Review on Generic Object Recognition Techniques: Challenges and Opportunities. International Journal of Advanced Research in Engineering and Technology, 6 (12), 2015, pp. 104-133