• DocumentCode
    650693
  • Title

    Combining Static and Dynamic Analyses to Reverse-Engineer Scenario Diagrams

  • Author

    Labiche, Yvan ; Kolbah, Bojana ; Mehrfard, Hossein

  • Author_Institution
    Dept. SCE, Carleton Univ., Ottawa, ON, Canada
  • fYear
    2013
  • fDate
    22-28 Sept. 2013
  • Firstpage
    130
  • Lastpage
    139
  • Abstract
    This paper discusses a step towards reverse engineering source code to produce UML sequence diagrams, with the aim to aid program comprehension and other activities (e.g., verification). Specifically, our objective being to obtain a lightweight instrumentation and therefore disturb the software behaviour as little as possible in order to eventually produce accurate sequence diagrams. To achieve this, we combine static and dynamic analyses of a Java software, reducing information we collect at runtime (lightweight instrumentation) and compensating for the reduced runtime information with information obtained statically from source code. Static and dynamic information are represented as models and UML diagram generation becomes a model transformation problem. Our validation against a previous, correct approach shows that we indeed reduce the execution overhead inherent to dynamic analysis, while still producing useful diagrams.
  • Keywords
    Java; Unified Modeling Language; program diagnostics; reverse engineering; system monitoring; Java software; UML diagram generation; UML sequence diagrams; dynamic analysis; model transformation problem; program comprehension; reverse engineering source code; reverse-engineer scenario diagrams; software behaviour; static analyss; Data models; Flow graphs; Instruments; Reverse engineering; Runtime; Software; Unified modeling language; Dynamic analysis; Reverse-engineering; Scenario diagram; Sequence diagram; Static analysis;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Software Maintenance (ICSM), 2013 29th IEEE International Conference on
  • Conference_Location
    Eindhoven
  • ISSN
    1063-6773
  • Type

    conf

  • DOI
    10.1109/ICSM.2013.24
  • Filename
    6676884