let-to-define tricky case

[Metaxal]

https://github.com/jackfirth/resyntax/blob/7fc36a29268f1210efa840c587881ad3fb4286e5/default-recommendations/let-binding-suggestions.rkt#L40

A recent example by Alexis King demonstrates that there may be more tricky cases where a let cannot be turned into a define: https://groups.google.com/g/racket-users/c/XkGnOMQKi2g/m/3AJJGaC4AgAJ?utm_medium=email&utm_source=footer&pli=1

Not sure how to detect or handle this one :-/

[jackfirth]

Things seem fine here actually. I took both the examples Alexis gave and wrapped them in functions (so that resyntax would see that they're in tail position) and manually applied the let-to-define migration myself. Here's what I ended up with:

(define (f)
  (let ([x 'outer])
    (define-syntax-rule (m a)
      (let ([a 'inner]) x))
    (m x)))

(define (f2)
  (define x 'outer)
  (define-syntax-rule (m a)
    (let ([a 'inner]) x))
  (m x))

(define (g)
  (let ([x 'outer])
    (define-syntax-rule (m a)
      (begin
        (define a 'inner)
        x))
    (m x)))

(define (g2)
  (define x 'outer)
  (define-syntax-rule (m a)
    (begin
      (define a 'inner)
      x))
  (m x))

Results:

• (f) and (f2) both produce 'outer, so no problems there
• The definition of g2 fails to compile, as it introduces a duplicate binding. The current let-to-define rule isn't able to detect this, because it doesn't actually locally expand the body of the let when it tries to determine if the rewrite would make duplicate bindings. So fixing #10 would address this issue.

[jackfirth]

Closing this as a duplicate of #10.