With some minor alterations it should be just as fast as the Java implementation.
David On Fri, Apr 20, 2012 at 7:35 PM, Casper Clausen <casp...@gmail.com> wrote: > Excellent! Any idea how this implementation would compare performance > wise to the java implementation if imported to Clojure? > > On Apr 20, 8:36 pm, Michał Marczyk <michal.marc...@gmail.com> wrote: > > It's pure ClojureScript. Hopefully a step towards CinC, yes. :-) > > > > Sincerely, > > Michał > > > > On 20 April 2012 20:32, Brent Millare <brent.mill...@gmail.com> wrote: > > > > > > > > > > > > > > > > > Quick question, so does this mean we have clojure's persistent data > > > structures implemented in clojurescript or js? This would mean we are > one > > > more step closer to C-in-C right? > > > > > On Friday, April 20, 2012 1:38:17 PM UTC-4, Michał Marczyk wrote: > > > > >> Since the latest PHM patch has now been merged to master (thanks, > > >> David!), I wanted to take this opportunity to note that porting all > > >> that Java code (including the transient support for PHM -- a working > > >> version of which is available for testing in its own ticket [1] -- and > > >> now the PersistentTreeMap [2]) has been completely smooth sailing. > > >> Some additions have been made to the implementation to improve > > >> performance while maintaining clarity of the code (here some excellent > > >> suggestions from David were very helpful), but the initial > > >> implementation already worked without them and client code could > > >> absolutely replicate them (by providing the requisite compiler macros > > >> in its own namespace). It's not that I expected insurmountable > > >> difficulties, but experiencing just how complete ClojureScript already > > >> is in the context of this sort of non-trivial data structure > > >> implementation task has been amazing. > > > > >> For those interested in how PHM's performance compares to that of the > > >> previously used ObjMap and HashMap copy-on-write implementations, > > >> there are some jsPerf tests linked to from the ticket [3]. There's > > >> also a TransientHM vs. PHM comparison linked to from [2]. > > > > >> Sincerely, > > >> Michał > > > > >> [1]http://dev.clojure.org/jira/browse/CLJS-181 > > >> [2]http://dev.clojure.org/jira/browse/CLJS-187 > > >> [3]http://dev.clojure.org/jira/browse/CLJS-178 > > > > > -- > > > You received this message because you are subscribed to the Google > > > Groups "Clojure" group. > > > To post to this group, send email to clojure@googlegroups.com > > > Note that posts from new members are moderated - please be patient > with your > > > first post. > > > To unsubscribe from this group, send email to > > > clojure+unsubscr...@googlegroups.com > > > > > For more options, visit this group at > > >http://groups.google.com/group/clojure?hl=en > > -- > You received this message because you are subscribed to the Google > Groups "Clojure" group. > To post to this group, send email to clojure@googlegroups.com > Note that posts from new members are moderated - please be patient with > your first post. > To unsubscribe from this group, send email to > clojure+unsubscr...@googlegroups.com > For more options, visit this group at > http://groups.google.com/group/clojure?hl=en > -- You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com Note that posts from new members are moderated - please be patient with your first post. To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en
