Fork me on GitHub
#rewrite-clj
<
2019-06-22
>
borkdude11:06:35

apart from the issue of republishing, are you also doing performance testing to see if performance is comparable to the original?

borkdude11:06:05

e.g. parse a giant code base, for example clojure and clojurescript themselves

lread12:06:01

I think I’ll release alpha before any perf tests. I expect clj will remain same/close to original but am curious about cljs as I favored, for an initial release, a single code base over preserving specific tunings (http://rundis.github.io/blog/2015/clojurescript_performance_tuning.html).

borkdude12:06:07

oh, moving away from multimethods is also something I wanted to do for my clone of rewrite-clj

borkdude12:06:09

very interesting link, thanks

lread12:06:01

interesting, I guess some real perf testing will help us to see the benefits such optimizations might bring.

borkdude12:06:46

I expect not much, but when I read that code the multi-methods weren't of any benefit afaik, it's using a static lookup table directly above those

lread12:06:44

ah, I see, thanks

sogaiu23:06:12

@lee your message to xsc seemed good to me. i think it's possible there are some folks around who might have had direct contact with him before, if we could find some of those folks, may be they'd have more luck establishing contact. perhaps this topic of reaching existing / past maintainers is worthy for clj-commons discussion? may be there could be some community-level action regarding putting out a call to find folks who might be able to reach someone?