spearman wrote on 07/06/2017 02:06 PM:
They claim to support the full Scheme standard so I guess there's a runtime and/or GC gets involved at some point, something I would like to avoid.
Correct, you do need GC, in the normal case. (Though a whole lot of RAM or swap would work in many cases. Even more, if you prioritize reducing garbage in your Racket style.)
If you're curious, there are some interesting ways of translating Scheme to C, such as: http://home.pipeline.com/~hbaker1/CheneyMTA.html
Also take a quick look at PreScheme, in case you someday find a need for it. https://en.wikipedia.org/wiki/PreScheme
I also like the idea of making a Racket `#lang` with s-expression syntax for C (just take a C parse tree or AST, and simplify it where you can), have the s-expression syntax forms ultimately emit C code with a very simple transliteration, and steal the existing Racket syntax transformers for use atop that syntax.
Nowadays, I like to just write pure Racket code for almost everything, and not flex my C muscles until I have to. Though I do occasionally enjoy trying to write perfect C. (Proper C is harder than most C programmers seem to think it is, or we wouldn't have a continuous stream of C memory exploits in mission-critical software.)
-- 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.
