News for hackers. Stuff that matters.

A comment on The Common Lisp Condition System – Upcoming Book. Parent comment

Re:The Common Lisp Condition System – Upcoming Book by aidenn0 on Wednesday July 15, 2020 @09:54AM (#23845237)
> The book also contains a description of a full implementation of a condition system that can (and hopefully will!) be used outside Common Lisp

Do you foresee any issues with other languages only having dynamic (i.e. THROW) non-local transfers of control rather than lexical as well (i.e. GO). I find it convenient for when I want to conditionally unwind the stack to have lexical transfers, and obviously it's been demonstrated you can implement GO with just THROW, so there's no loss of generality, but I wonder what the ergonomic impact will be.

[ Parent | on HN ]
Re:The Common Lisp Condition System – Upcoming Book by phoe-krk on Wednesday July 15, 2020 @10:00AM (#23845305)
None, other than for sheer performance. (THROW will perform worse than GO and RETURN since the list of catch tag will need to be traversed on each throw.)

As you mentioned, one can implement both GO and RETURN via THROW plus gensyms - see https://web.archive.org/web/20191008050814/http://home.pipel... for an implementation.

[ Parent | on HN ]