R. Koschke. Automatic Architectural Component Recovery for Program Understanding and Evolution. PhD thesis, Institut fur Informatik, Universitat Stuttgart, 2000.  Home/Search   Document Not in Database   Summary   Related Articles

....with C code yet. The operating system servers of the SawMill framework should be improved by building them from components. Components and component systems raise two important questions. The rst question is how to nd components in existing code, and how to evaluate the resulting components. Kos00] describes several clustering heuristics and the information they base on. He also evaluates the resulting clusters of well known programs, according to manually generated components. To evaluate the resulting clusters from legacy code, this method of evaluation is not feasible. It is not quite ....

....approach is similar to the re exion model and could be used for code analysis, but the presented tools are not yet sucient and they do not provide manipulation of the components. The second question, after code analysis and component extraction is, how to represent and manage these components. Kos00] suggests the Resource Usage Graph to represent structural information about source code. This graph is an abstract representation of this information. This graph is close to the AST, used by compilers. This allows simple representation of source code, and therefore of components in it. It ....

[Article contains additional citation context not shown here]

R. Koschke. Automatic Architectural Component Recovery for Program Understanding and Evolution. PhD thesis, Institut fur Informatik, Universitat Stuttgart, 2000.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC