On Thu, Dec 11, 2008 at 8:28 PM, Stephen C. Gilardi <squee...@mac.com> wrote: > > On Dec 11, 2008, at 7:24 PM, Rich Hickey wrote: > >> I am interested in the issues you are trying to address, and thanks >> for volunteering! > > Excellent. You're welcome. > >> I'd like to try to focus our efforts on release 1.0. > > Sounds good. > >> Towards that end, it would be nice if your repl code got more >> exercise, which it probably isn't given existing scripts which run >> clojure.lang.Repl. Maybe it would be useful to have that just call >> clojure.main/repl? > > I think clojure.lang.Repl should translate its args to the new format and > call clojure.main/-main: > > old: clojure.lang.Repl file1 file2 -- a b c > new: clojure.main -i file1 -i file2 -r a b c > > Similarly for clojure.lang.Script: > > old: clojure.lang.Script file1 file2 file3 -- a b c > new: clojure.main -i file1 -i file2 file3 a b c >
Ok, so calls to them won't change but they will delegate to main/repl. >> Other than that, I think we should hang on to these ideas post 1.0, >> since adding them now doesn't give us enough time to shake them out. >> Do you have a short list of things in this area that need to be fixed >> for 1.0? > > There are the two things I think need to be fixed for 1.0. (If you're > including [1] below in the set of things you want to defer until post-1.0, > please veto it.) > > [1] Stub launcher written in Java for when clojure.jar contains only source > code: > > - the new clojure.lang.Repl could fill this role, but I think that's not > ideal for 1.0 > > - Better would be to allow this java command line to work: > > java -cp clojure.jar clojure.main ... > > whether or not clojure.jar contains only source code. > > We could do this by writing a clojure/main.java that calls > clojure.main/-main and removing :genclass in clojure/main.clj. I think > that's a simple change that provides an important flexibility for > clojure.jar. Adding a "source_jar" target to build.xml would go hand in hand > with this. Ok. > > [2] A single "-e" or "--eval" option should be allowed to contain multiple > forms and it should evaluate them sequentially. There are cases where > wrapping multiple forms in a "do" does not work and writing > > -e (this) -e (that) -e (the-other) > > is tedious compared to > > -e "(this) (that) (the-other)" > This (-e) seems rather unimportant, but I wonder when does 'do' not work? 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 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 -~----------~----~----~----~------~----~------~--~---
