Meta Garbage Collection: Using OCaml's GC to GC Rust

38 points by annieversary

veqq

Why don't they use Rust for these tools?

tmcgilchrist

Maybe because the authors are more familiar with Rust? Soteria founders have this paper https://dl.acm.org/doi/10.1145/3808306 (haven't read it yet). IMO I don't see what Rust provides here that is a major selling point over OCaml.

Why is OCaml a bad choice here? There are many many similar tools for program analysis written in OCaml (Infer, Semgrep, Trust-in-Soft and so on).
- veqq
  
  OCaml's awesome, but it seems like a sensible default to write a language's tooling in that language, so I'd expect a particular reason but couldn't find one reading around. (People often call Rust an OCaml (or ML?) though too, complicating matters.)
  - char
    
    soteria aims to target many languages, not just rust
  - aidenfoxivey
    
    I thought they called Rust an ML? (Not to be pedantic - I just haven’t heard anyone conflate OCaml and Rust before.)
    
    ssokolow
    
    Back when /r/rust/ had RSS feeds, I called Rust an ML-lineage language disguised by the addition of ownership and borrowing and its use of C++ syntax on multiple occasions.
    
    Heck, basically every bit of syntax that couldn't be borrowed from C++ for familiarity traces back to Ocaml.
    
    Some examples:
    
    let for variable declaration and postfix type annotations separated with colons
    
    'a (Ocaml's equivalent to <T> for generics, because, on an abstract/theoretical level, Rust's lifetimes are a special type of generic parameter, right down to "a" apparently being the habitual default type variable name, just like "T" is in C++)
    
    Option (option in Ocaml), Some, and None
    
    match
    
    fn (fun or function in Ocaml)
    
    -> and => (just -> in Ocaml) for function returns and match arms
    
    ()
  - orib
    
    Garbage collection makes it a lot easier to deal with graphs in memory.