Probing into code analysis tools: A comparison of C# supporting static code analyzers

摘要

Essence of a software system lies in the quality of source code and the degree to which it follows the underlying coding standard. The widely-used code analysis techniques focus upon examining the programs without actually executing them. The major purpose is the detection of complex code constructs and the potential defects that result in the decrement in quality of the code. The code complexity is often assumed to inflate the maintenance cost and lead to the unexpected system behaviour. A number of code analyzing tools are available and currently many researches are being conducted to improve software quality, bring down software complexity without affecting its external behaviour. A brief review of the existing code analyzers is presented in this paper; however the major focus is upon the tools that analyze the source code written in C#. Researchers round the globe have identified that the potential problems in the source code are: code smells, code clones, anti-detect pattern etc that often lead to the increase in system complexity and hence amplification in the system response time. The code analyzing tools have their own significance depending upon the domain where they are being applied. Code analyzers are of extreme value especially in the mission critical systems, where the system efficiency is one of the factors for mission success. The methodology adopted for the purpose of review of the existing tools is that first of all, the purpose of tool is described, secondly, the personal findings, along with reviews from the clients and technology analysts regarding the tool are taken into account, thirdly, a sample C# application is chosen and the outputs from the tools for this sample are taken for analysis and comparison. Upon the basis of analysis, reviews and comparison, the relatively best tool among the 8 analyzed tools has been figured out.