Lattice Based Modularization of Static Analyses
Abstract: Static analyses which compute conceptually independent information, e.g., class immutability or method purity are typically developed as standalone, closed analyses. Complemen- tary information that could improve the analyses is either ignored by making a sound over-approximation or it is also computed by the analyses but at a rudimentary level. For ex- ample, an immutability analysis requires field mutability in- formation, alias/escape information, and information about the concurrent behavior of methods to correctly classify classes like java.lang.String or java.util.BigDecimal. As a result, without properly supporting the integration of in- dependently developed, mutually benefiting analysis, many analyses will not correctly classify relevant entities. We propose to use explicitly reified lattices that encode the information about a source code element’s properties (e.g., a method’s purity or a class’ immutability) as the sole interface between mutually dependent analyses. This en- ables the composition of multiple analyses. Our case study shows that using such an approach enables highly scalable, lightweight implementations of modularized static analyses.
Resources
BibTeX
@inproceedings {EKHR+,
title = {{Lattice Based Modularization of Static Analyses}},
author = {Eichberg, Michael and Kübler, Florian and Helm, Dominik and Reif, Michael and Salvaneschi, Guido and Mezini, Mira},
booktitle = {{Proceeding ISSTA '18 Companion Proceedings for the ISSTA/ECOOP 2018 Workshops}},
series = {SOAP 2018},
pages = {113-118},
year = {2018},
doi = {10.1145/3236454.3236509},
url = {http://dx.doi.org/10.1145/3236454.3236509},
}