Good day,

what are your opinions on srfi 145? Looking at it, it could be a decent
r7rs portable way to provide some degree of static type checking. The work
would involve adding a syntax tree rewrite pass. For example, code such as

(define (foo a)
  (assume (integer? a))
  ...)

would be transformed to

(define (foo a ::int)
  ...)

and then the rest of it would be usual kawa behavior. Certainly making it
as powerful as a standard kawa syntax would be very difficult, but perhaps
it's not necessary? I think handling only function parameters + let
bindings, and only handling base types + user defined simple records would
pragmatically cover a lot of useful cases.

If there is a consensus that this would be useful, I'll give it a shot at
implementing it myself.

Arvydas