On Jan 19, 2009, at 12:39 PM, Michael Reid wrote: > > Forgive me I am probably missing something, but if your use case is in > a configuration file, why can't you just to a regular (ns .. (:require > ..)) so that your configuration file can declare its dependencies in > the same way other code does?
Configurations can be loaded at runtime, in my system. Loading a configuration file updates a *settings* Atom (might change to a Var, actually). Sure, the configuration file could (require ...) all the namespaces it needs, but isn't resolve* nicer? :) Also, due to the late-binding nature of using resolve* I gain some performance benefits of not evaluating top-level forms in the required namespaces until they're actually needed. > > > But I also wanted to add that this could be a very useful function for > an IDE/tools to use. > > I'm thinking of the case where the user is editing code and wants to > examine a reference to a var in their source file. This enhanced > "resolve*" could be called to load and resolve the reference and then > the IDE/tool can do whatever it might do for the result--examine it > for a docstring, or load up an inspector or whatever. > > /mike > > > --~--~---------~--~----~------------~-------~--~----~ 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 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 -~----------~----~----~----~------~----~------~--~---
