On Fri, Oct 16, 2015 at 03:24:06PM +0200, Konrad Hinsen wrote:
Matthias Felleisen writes:
> > For me the strongest point of Racket is that it encourages linguistic
> > diversity while maintaining (nearly enforcing) interoperability. My
> > dream language environment would go one step further and provide a
> > second more low-level interoperability layer for performance-oriented
> > dialects (C/Fortran style).
>
>
> Does the existing FFI provide you with enough efficiency when needed?
Yes, but I have to write C code outside of Racket. I'd like to be able
to define a #lang in Racket that operates at the level of C
(i.e. machine-level data types, no GC, etc.), and I'd like to generate
specialized code in that low-level language from my standard Racket
code.
FYI, I'm a grad student at Utah with Matthew, and my current project
is a #lang pre-racket that compiles to C. It hasn't really gotten off
the ground yet because I've been busy with classes and fellowship
applications, but in the relatively near future we should have this.
William
P.S. On the point of keywords, I think having #lang colon-kw <lang> is
the right approach. If people worry about living the dream of Haskell's
Lens package, you could make another one that includes your common
subset of add-on reader features together.
--
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.
