Issues in clone classification for dataflow languages

Abstract

While clone detection and classification research for textual source code is well-established, clones in visual dataflow languages have only recently received attention. The accepted existing clone classification framework does not adequately capture the nature of clones in the latter kind of programs. In this article, we propose a new classification framework for clone types that may be found in dataflow programs. It parallels the scheme for textual languages but accounts for the differences in syntax and semantics present in graphical languages.