On Mon, Feb 15, 2010 at 3:56 AM, Konrad Hinsen <konrad.hin...@fastmail.net> wrote: > On 14.02.2010, at 22:48, Mark Engelberg wrote: > >> Actually, the more I think about it, the more I feel like deftype's >> "specify clojure.lang.IPersistentMap as an interface with no >> implementation, and you'll get a default implementation" seems like a >> weird exception, > > I agree. The current deftype is meant to be both the primitive construct for > implementing data types, and a high-performance substitute for defstruct. It > would be conceptually cleaner to separate these two roles. > >> Has their been any community brainstorming about what the syntax could >> look like for a general mechanism to specify within a deftype default >> implementations or default partial implementations of various >> interfaces? > > I don't remember. The issue of default implementations comes up now and then, > but only in the form of the question "is it possible?". > > In my own applications, I have only used deftype as a primitive construct, > not as a substitute for defstruct. The special status of IPersistentMap thus > never bothered me. However, I do see other cases where default > implementations would be useful to have. And then IPersistentMap could become > just one of them. >
I just wanted to chime in and say I agree - I am not totally happy with how getting the map/default implementation works, and think that possibly the two roles deftype/defstruct-replacement should be more distinct. So, I'm working on it. Rich -- 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
