On Wed, Dec 22, 2010 at 1:29 PM, Nicolas Buduroi <nbudu...@gmail.com> wrote:
> On Dec 22, 1:03 pm, David Nolen <dnolen.li...@gmail.com> wrote: > > (Foo.) is a Java method call determined at compile-time. If you change > the > > definition of the Foo type/record, create-foo will be out of date. > Default > > constructor fns could alleviate this but I think some design/modularity > > issues need to be hashed out. > > Yes, I've realized this when developing this project. At first, it was > less modular and I encountered much more of such issues. I also used > `immigrate` previously and that was causing troubles not only with > protocols, but with multimethods also. In the end there still appears > to be some problems while developing protocol based code in a REPL- > driven fashion. I was hoping I did something wrong that somebody could > point out. > > P.S.: It makes me wonder if there's some improvements that can be done > to make that type of coding more REPL friendly. Hmm. Actually now that you mention it I do recall I did run into some issues when I was interactively developing types/records across namespaces. Probably worth coming up with small reproducible example. David -- 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
