If the sets data structure is also not shared, then the paper I mentioned (link provided earlier) is one of the fastest to date. And it is very small & easy to implement.
On Wed, Dec 16, 2009 at 2:59 PM, Dragan Djuric <draga...@gmail.com> wrote: > Yes, the true/false for equality is not a problem. > > I am looking for a shortcut that finds different elements more > efficiently. So, the sets are different, but I want to get hold of > elements that are in s2 but not in s1. > > On Dec 16, 8:38 pm, Richard Newman <holyg...@gmail.com> wrote: > > > I imagine this is possible if you drill into the guts of > > > PersistentHaspMap, but I would strongly discourage the behavior in > > > user code. Perhaps as an upgrade to the object itself? There is a 1% > > > chance that this could be a language upgrade, assuming it works across > > > the board. I would tread cautiously. > > > > I could easily imagine an implementation of equality for two tree sets > > checking for intermediate node reference equality as a shortcut... but > > as you say, not in user code. > > > > Right now, APersistentSet.equals(Object o) only casts to Set, and does > > equality checking by iterating over the elements of the input set. > > > > I'd certainly benchmark this before doing the work, though -- I'd > > guess that most equal sets will hash-compare as equal, most non-equal > > sets will fail early on in a per-element comparison, and it's probably > > a minority of compared sets that will share structure. It's unlikely > > that adding an additional object identity comparison is worth the > > machine instructions. > > > > My hunch (from examining the performance of Clojure's sets in the > > past) is that it's not worth doing the work... > > -- > 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<clojure%2bunsubscr...@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
