No, it actually compiles down to a pile of ifs, and try/catch/throws (for
backtracking). It's pretty fast, but it won't be as fast as a jump table
for complex matches, that's for sure.
Timothy
On Fri, May 15, 2015 at 5:53 PM, Surgo wrote:
> core.match compiles down to case, does it not? So the
core.match compiles down to case, does it not? So the comparisons here are
similar: http://insideclojure.org/2015/04/27/poly-perf/
-- Morgon
On Friday, May 15, 2015 at 2:57:23 PM UTC-4, tbc++ wrote:
>
> One big thing to consider is that core.match is closed dispatch. If you
> write a function t
One big thing to consider is that core.match is closed dispatch. If you
write a function that uses core.match, I can't extend it inside my code.
This is something that is possible with both multi-methods and protocols.
Timothy
On Fri, May 15, 2015 at 12:49 PM, Sam Raker wrote:
> The discussion/
The discussion/post-linked-to
in https://groups.google.com/d/msg/clojure/eoAp6QVimYI/iipmEJNKdrIJ have
got me thinking about protocols & multimethods, which I admittedly have
possibly never actually used, now that I think about it. I'm wondering how
they differ from core.match[1]. I realize pro
