We present Spt, a tool that helps programmers write low-level data-structure manipulations by combining various forms of insights such as abstract and concrete input-output examples as well as implementation skeletons. When programmers write such manipulations, they typically have a clear high-level intuition about how the manipulation should work, but implementing efficient low-level pointer manipulating code is error-prone. Our tool aims to bridge the gap between the intuition and the corresponding implementation by automatically synthesizing the implementation. The tool frames the synthesis problem as a generalization of an abstract-interpretation based shape analysis, and represents the problem as a set of constraints which are solved efficiently by the Sketch solver. We report the successful evaluation of our tool on synthesizing several linked list and binary search tree manipulations. © 2012 Springer-Verlag.
CITATION STYLE
Singh, R., & Solar-Lezama, A. (2012). SPT: Storyboard programming tool. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 7358 LNCS, pp. 738–743). https://doi.org/10.1007/978-3-642-31424-7_58
Mendeley helps you to discover research relevant for your work.