Hi, Am 20.10.2011 um 19:53 schrieb Micah Martin:
> I recently tried to get Speclj running on Clojure 1.3 and came across the > following problem: > > (list > (declare ^:dynamic p) > (defn q [] @p)) > > (binding [p (atom 10)] > (q)) > >> java.lang.ClassCastException: clojure.lang.Var$Unbound cannot be cast to >> clojure.lang.IDeref > > Thanks to @cemerick for helping me condense the snippet, and thanks to both > @cemerick and @chouser for the lively discussion on IRC. Yet the discussion > was inconclusive. Is the above expected behavior? I'm not sure. Maybe it's undefined? Clojure compiles per top-level form. In this case this is the call to list. However in p does not exist before. It is created by the def inside this form. And maybe I'm just a weenie, but I wouldn't bet on some not very well documented peculiarities of def. >From the “one top-level form” rule do is an exception. And indeed this seems >to be for a reason: user=> (do (declare ^:dynamic p) (defn q [] @p)) #'user/q user=> (binding [p (atom 10)] (q)) 10 So maybe the rule of thumb is: never reference defs made in the same top-level form unless this form is a do. I'm sorry. I can't explain things further. I suspect that the effect you see is due to some accidental, implementation-dependent behavior of the compiler. Sincerely Meikel -- 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
