Traditional parsing literature and design tends to focus on LL or LR grammars (Johnson et al., 1975; Parr and Fisher, 2011; Ford, 2004; Ortin et al., 2022; Leijen and Meijer, 2001) or have contexts that require minimal computation that can be reasonably expressed in formal notations not designed for general purpose computation (Woods, 1970; Laurent and Mens, 2016; Kurs et al., 2014). The architecture and tooling support that emerges from this research and the general community usually comes in the form of parser generators, parser combinators, or, usually, a fallback to recursive descent implemented directly in the host language. Context sensitive languages often feel like second class citizens in the software engineering parsing ecosystem (Laurent and Mens, 2016; Kurs et al., 2014). In our work on static, offline parsing of APL code, the complexity of the task has driven us to coin the term “perverse parsing environments” to describe the situation.