If this situation is common enough, shouldn't defprotocol support optional implementations which are implicitly merged?
On Feb 9, 6:01 am, Konrad Hinsen <konrad.hin...@fastmail.net> wrote: > On 09.02.2010, at 02:14, Stuart Sierra wrote: > > > On Feb 8, 6:13 pm, aria42 <ari...@gmail.com> wrote: > >> (defprotocol Span > >> (start [self]) > >> (stop [self]) > >> (span-length [self])) > > >> Now I know I can just make span-length a function on Span as opposed > >> to part of the protocol. Is that what one should do? > > > Yes. > > I would say "it depends". > > I have a similar situation in my multiarray package > (http://code.google.com/p/clj-multiarray/). In the multiarray protocol, I > have two functions, "shape" and "rank", with the latter being by definition > the same as (comp count shape). However, I still have "rank" in the protocol, > because for some implementations it is more efficient to compute the rank > directly, rather than construct a shape vector just for computing its length > afterwards. > > In such situations it is useful to provide a default implementation and leave > it up to each type to implement a more efficient alternative or not. With > extend and the maps that go with it, this is easy to achieve: make a map with > the default implementations, and merge this with the type-specific > implementations fed to extend. > > Konrad. -- 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
