I guess I was getting at how to do the airity overloading. My first instinct when developing is do something like this
(def distinct ([coll] (distinct identity coll)) ([f coll] (lot-of-code))) However, if you look around core, you'll see that Rich "repeats himself" a lot. This really obvious w/ fns like juxt & partial. It turns out that this is results in faster code. So I'd implement distinct like this. (def distinct ([coll] (lots-of-code)) ([f coll] (lot-of-similar-code))) That's all I was saying. Sean On Feb 23, 2:40 pm, Michał Marczyk <michal.marc...@gmail.com> wrote: > On 23 February 2010 20:26, Sean Devlin <francoisdev...@gmail.com> wrote: > > > Take a look at sort/sort-by in core. That would be starting point for > > fn style. > > Not sure what you mean by that. > > If anything, having an arity-overloaded distinct (optionally accepting > a binary predicate to use in place of =) and a separate distinct-by > (accepting an extra keyfn argument plus optionally the binary > predicate) would make for a more uniform standard lib. > > (I guess I suggested switching to the Haskell naming convention > before, but there's no real reason to do that if the same > functionality is available, plus I'd be confused by this myself, > having gotten used to the Clojure names already... I retract that > part.) > > Sincerely, > Michał -- 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
