DocumentCode :
188969
Title :
VizMe: An Automatic Visualization Tool for Source Codes
Author :
Junseok Cheon ; Kyungeun Jo ; Gyun Woo
Author_Institution :
Dept. of Electr. & Comput. Eng., Pusan Nat. Univ., Busan, South Korea
fYear :
2014
fDate :
11-13 Sept. 2014
Firstpage :
592
Lastpage :
597
Abstract :
As the scale of software is getting large and complex, the development cost of software scales up to lots of man-months. The maintenance cost of complex software is also increasing and usually more than the development cost. Understanding the underlying logic and modifying the actual code become the large part of the maintenance. However, understanding the code can be extremely hard even for professional developers especially when the code to modify is written by others. Though several visualization methods such as flowcharts have been proposed up to now, it is still hard to adopt such methods in analyzing existing code because the complexity of the code results in complex visualization results. This paper introduces VizMe, the automatic visualization tool for helping the developers to understand the source codes. VizMe accepts the source code written in C, Java, or Scheme and produces the visual output including flowcharts, PADs (problem analysis diagrams), and pseudo codes. According to the experimental results, the complexity of PADs is less than the flowcharts, the graph density of PADs is 0.024 and that of flowcharts is 0.027, the cyclomatic complexity of PADs is 1.0 and that of flowcharts is 8.6, and the graph entropy of PADs is 0.469 and that of flowcharts is 0.356. The last result is much interesting that it seems to imply that, for the same code, the corresponding PADs are more irregular than the flowcharts. However it seems that this controversial result comes from the diversity of graph nodes of PADs.
Keywords :
data visualisation; software maintenance; PAD; VizMe tool; automatic visualization tool; code modification; code understanding; flowcharts; problem analysis diagram; pseudo codes; software development cost; software maintenance cost; source codes; visualization methods; Complexity theory; Entropy; Equations; Java; Software; User interfaces; Visualization;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computer and Information Technology (CIT), 2014 IEEE International Conference on
Conference_Location :
Xi´an
Type :
conf
DOI :
10.1109/CIT.2014.33
Filename :
6984717
Link To Document :
بازگشت