Inspired by
https://groups.google.com/d/msg/racket-users/H7vilh3KcD4/WoKcRUXe-PAJ, I
have a strange but simple workaround. Essentially, I added a module like:
#lang racket
(require syntax/parse/define)
(provide (rename-out
[child-module+ module+]
[child-module* module*]))
(define-syntax-parser child-module+
[(_ . stuff)
#'(module+ . stuff)])
(define-syntax-parser child-module*
[(_ . stuff)
#'(module* . stuff)])
The exported versions of `module+` and `module*` (for the `(module* child
#f ...)` case) carry the `#%module-begin` from `#lang racket` along with
them. This strikes me as very strange, and there are probably deeper
subtleties it doesn't address, but it solves my immediate issues.
-Philip
On Tue, Sep 4, 2018 at 6:19 PM Philip McGrath
wrote:
> Is there a way to control what version of `#%module-begin` is introduced
> for a `module+` submodule?
>
> I have a DSL that is implemented by doing some fairly strange things with
> `#%module-begin`, but I want to reuse `module+` in the DSL. The submodules
> should be able to see all of the bindings from the parent module as usual,
> but I want them to use the normal `#%module-begin` from `racket/base`, not
> the crazy one from the parent module. (The same goes for `(module* child #f
> ...)`, but I think it might be easier to add a wrapper around `module*`
> than to re-implement `module+`.)
>
> -Philip
>
--
You received this message because you are subscribed to the Google Groups
"Racket Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to racket-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
