Inferring Region Types via an Abstract Notion of Environment Transformation

Ulrich Schöpp and Chuangjie Xu

Asian Symposium on Programming Languages and Systems (APLAS 2022), pp. 45–64

September 2022 · doi: 10.1007/978-3-031-21037-2_3

abstract

Region-based type systems are a powerful tool for various kinds of program analysis. We introduce a new inference algorithm for region types based on an abstract notion of environment transformation. It analyzes the code of a method only once, even when there are multiple invocations of the method of different region types in the program. Elements of such an abstract transformation are essentially constraints for equality and subtyping that capture flow information of the program. In particular, we work with access graphs in the definition of abstract transformations to guarantee the termination of the inference algorithm, because they provide a finite representation of field access paths.