POPL 2027
Sun 10 - Sat 16 January 2027 Mexico City, Mexico

One approach to reasoning about effectful programs is to use program logics that provide proof rules for various effects. However, developing program logics requires skills that are distinct from those needed for using program logics, making the development of new logics less accessible. Moreover, when developing new logics, it can be difficult to reuse components from prior logics or combine support for different effects.

In this paper, we propose an approach for building extensible program logics based on effect handlers. Our starting point is an expressive program logic for reasoning about programs written in a pure, sequential language with support for effect handlers. Within this language, we implement handlers that model concurrency, distributed execution, and crash-recovery behavior. Then, by proving properties of these handlers within the logic, we derive expressive rules for reasoning about these effects. In some cases, this approach leads to stronger reasoning rules than those found in prior program logics.

In addition, we develop a relational logic for proving contextual refinements between programs using effects. As with unary reasoning, handlers enable this relational logic to be developed in an extensible way.