Returning the symbol seems odd at least to me. I think the best we could do is have a compiler flag so that defs can return nil for REPLs.
On Tuesday, May 15, 2012, kovas boguta wrote: > Yeah, that is sort of what I was implying in "changing the semantics of > def" > > Though I wasn't ready to totally committed to that, since I don't > understand the properties of symbols in clojurescript. > > Like, how do go from the symbol to the javascript object we've just > bound to the symbol? > > > On Tue, May 15, 2012 at 2:26 AM, Laurent PETIT <laurent.pe...@gmail.com> > wrote: > > Clojurescript doesn't have vars, so why not have def return the symbol ? > > > > > > > > Le 15 mai 2012 à 06:14, kovas boguta <kovas.bog...@gmail.com> a écrit : > > > >> I think this is a pretty valid feature request. > >> > >> The main question is, can this be done without having vars in > clojurescript. > >> > >> One way to do it is to surpress output somehow, under certain > conditions. > >> > >> Either as a token at the end of a repl input, or in the semantics of > def itself. > >> > >> I don't have an ideal specific solution here, but I've noticed this > >> problem as well and found it pretty annoying. > >> > >> There should be a way to solve it that is a reasonable compromise. > >> > >> > >> > >> On Mon, May 14, 2012 at 11:13 PM, Mark Engelberg > >> <mark.engelb...@gmail.com> wrote: > >>> On Mon, May 14, 2012 at 4:41 PM, David Nolen <dnolen.li...@gmail.com> > wrote: > >>>> > >>>> On Mon, May 14, 2012 at 7:27 PM, Mark Engelberg < > mark.engelb...@gmail.com> > >>>> wrote: > >>>>> > >>>>> (def tree (function-that-produces-an-enormous-tree 2)) > >>>> > >>>> > >>>> Isn't doing this at the top level bad form? > >>> > >>> > >>> The purpose of a REPL is for interactive experimentation. I want to > give > >>> names to the things I'm building so I can play with them in the REPL. > >>> Nothing bad form about that. > >>> > >>>> > >>>> Also I don't see how this isn't solved by modifying some habits. > >>>> > >>>> (defn test-tree [] (function-that-produces-an-enormous-tree 2)) > >>>> (time (test-tree)) > >>>> > >>> > >>> time prints out the value that is computed. The above example would > suffer > >>> the same problem of printing out the tree at the REPL. You could do > >>> something like (time (do (test-tree) nil)) to suppress printing, but > if you > >>> want to do further interactive manipulations to the tree, you'd end up > >>> recomputing it. So then, you get into workarounds involving delay. It > >>> starts to get ugly, I think. > >>> > >>> Clojure's ability to give names to things without printing the values > is a > >>> feature I use every day in my interactive explorations. > >>> > >>> > >>> -- > >>> 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 > >> > >> -- > >> 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 > > > > -- > > You r -- 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
