kamruzzamanasif / spl-1

Static Code Analyzer (SCA) analyzes the ‘C’ source codes without executing the program. It determines the software metrics (such as LOC and Halstead Complexity metrics) and detect syntactic code clones.

Software Project Lab 1

Static Code Analyzer

Static Code Analyzer (SCA) analyzes the ‘C’ source codes without executing the program. It determines the software metrics (such as LOC and Halstead Complexity metrics) and detect syntactic code clones.

Table of Contents

• Project Descriptioin
• Technologies Used
• Features
• Screenshots
• Setup
• Project Status
• Acknowledgements
• Contact
• License

Project Description

My software tool named Static Code Analyzer (SCA) analyzes the ‘C’ source codes without executing the program. It determines the software metrics (such as LOC and Halstead Complexity metrics) and detect syntactic code clones.

Static analysis of source code is very important and useful analysis for software quality and security. Software developers use static code analysis for better understanding of the software, maintaining coding standards, identifying potential bugs, detecting code clones and code smells, code simplification and sanitizing, improving application performance, better resource utilization, etc.

The whole project consists of two major parts software metrics generation and clone detection. In metrics part to calculate the LOC metrics(such as physical lines of code, number of comment lines and number of logical statements, etc.), I have followed the counting rules of SLOC according to SEI and the IEEE’s standards. At first, a single C source code file is taken as input from the user. The user can input the file path or choose a file to be processed. Following the counting rules and using several string manipulations (linear searching, substring generation, string matching) the LOC metrics are then generated.

For Halstead metrics (such as distinct operators and operands, the total number of operators and operators) and Halstead measures (such as program vocabulary, length, volume, difficulty, effort, time and number of delivered bugs) generation I have followed several steps. At first the preprocessing of source code (such as removal of blank lines, comments, extra spaces, function declarations and hash directives) is done. Then tokenize the preprocessed code and get tokens of all types (such as character tokens, string tokens, number tokens, identifier tokens, etc.) which are later categorized into two major types of tokens (Operator token and Operator token). Lastly, counting tokens and calculating the Halstead measures we get the Halstead metrics.

Clone detection is another core part of my project. To detect code clones, I have followed a token-based approach with the application of winnowing algorithm. At first the source code is preprocessed such as removal of blank lines, comments, hash directives, punctuations, extra white space, and capitalization of letters. Then from the preprocessed code k-grams are generated. The k-grams are then use for hashing and using rolling hash we get the hash values of the k-grams. Next the k-gram hashes are pushed into winnowing algorithm. The winnowing algorithm generates fingerprints from the hash values. Then comparing the fingerprints using statistical methods (Jaccard Similarity and Dice Similarity Coefficient), we get the clone result.

Through this project we get the static analysis of ‘C’ programs. We get software metrics of LOC and Halstead “software science” which provides objective information throughout the software organization. This reduces the ambiguity that often surrounds complex and constrained software projects. Metrics are the key to objectively representing the progress of project activities and the quality of associated software products across the project life cycle. We also can detect code clones which is very important in software engineering. Applications of code clone detection and management are restructuring of clone codes without any change in the external behavior of the code, duplicate code can be removed, bug detected in copied code leads to identifying bugs in original code or vice versa, code size can be reduced, reuse of existing code, etc.

Technologies Used

• Java - openjdk 11.0.8
• JavaFX - version 17.0.4
• CSS

Features

• LOC
• Halstead
• Code Clone
• GUI

Screenshots

Setup

• SCA -> without GUI
• SCA_GUI -> GUI added

• Requirements: Java (openjdk 11.0.8) and JavaFX - version 17.0.4

• Installation Instructions:

  1. clone this repository:

     $ git clone https://github.com/KamruzzamanAsif/SPL-1.git
     

  2. Open the project directory in an IDE (IntelliJ/NetBeans/VsCode/etc.)
  3. Go to SPL-1/SCA_GUI/src/main/java/GUI/Main.java and run

Project Status

Project is: complete & no longer being worked on

Acknowledgements

• This project was inspired by my supervisor
• Many thanks to my seniors and mates

Contact

Created by @KamruzzamanAsif - feel free to contact me!

License

Unlicensed: This project is open and available for all