Sorry about the late response, Gary. Would you mind taking the discussion over to cider’s issue tracker? (I tend to miss emails, but I don’t miss issues :-) ).
The second change seems totally reasonable. I guess ritz’s complete middleware returns the completion candidates in some odd format. Regarding the tooling-session - why can’t the completions be generated using the tooling session with ClojureScript (my knowledge of it is pretty basic). I guess this was something to do with piggieback, right? -- Cheers, Bozhidar On Saturday, January 18, 2014 at 8:56 PM, Gary Trakhman wrote: > Bozhidar, I had to slightly modify cider-interaction.el to make autocomplete > work for cljs. > > --- cider-interaction.el 2014-01-18 13:51:28.082131609 -0500 > +++ /home/gary/.emacs.d/elpa/cider-0.4.0/cider-interaction.el 2014-01-17 > 19:06:45.872591834 -0500 > @@ -469,12 +469,12 @@ > (let ((strlst (plist-get > (nrepl-send-request-sync > (list "op" "complete" > - "session" (nrepl-current-tooling-session) > + "session" (nrepl-current-session) > "ns" nrepl-buffer-ns > "symbol" str)) > :value))) > (when strlst > - (car strlst)))) > + strlst))) > > > > Here's the current implementing code: > https://github.com/gtrak/nrepl-complete/blob/master/src/nrepl_complete/middleware.clj > > Here's a screenshot: http://i.imgur.com/GmBJ6Fj.png > > It can't be implemented for cljs on the tooling session without sending the > real session id over somehow, it seems this would be wasteful and not a good > precedent for other middlewares. > > I'd appreciate your thoughts. > > I'm also curious what it would take to make the display pretty like ac-nrepl, > which I had to disable. > > As soon as this stuff is finalized, I'll package everything up nice and make > a first release. > > > > On Mon, Jan 13, 2014 at 10:21 AM, Gary Trakhman <gary.trakh...@gmail.com > (mailto:gary.trakh...@gmail.com)> wrote: > > Austin's lein-plugin already manipulates project middlewares, so that's an > > easy target. Onward! > > > > > > On Mon, Jan 13, 2014 at 10:10 AM, Bozhidar Batsov <bozhi...@batsov.com > > (mailto:bozhi...@batsov.com)> wrote: > > > Cider’s completion understands a `complete` op, so the middleware is the > > > best approach if you ask me. The only reason that there’s also an eval > > > based completion mechanism (the one used by default) is that > > > clojure-complete is present as a REPLy (which is used by lein) dependency > > > and many newcomers have absolutely no idea what an nREPL middleware is. > > > Unfortunately it’s hard to balance initial easy of setup and good design > > > decisions. > > > > > > -- > > > Cheers, > > > Bozhidar > > > > > > > > > On Monday, January 13, 2014 at 4:55 PM, Gary Trakhman wrote: > > > > > > > > > > > > > On talking to Chas, > > > > https://github.com/cemerick/piggieback/issues/22 > > > > > > > > it seems like the right approach is to reify ac-nrepl's use of eval > > > > into a real "complete" op, and reimplement it to use that, then a > > > > common middleware can either use clojure's environment > > > > (clojure-complete) or piggieback's compiler state to implement the > > > > appropriate auto-complete based on the active repl. > > > > > > > > The issue here is that clojure's auto-complete takes the JVM state as > > > > an implicit parameter, whereas cljs-complete requires an 'env' arg that > > > > has to come from somewhere (piggieback has a var that keeps track of > > > > repl session state). Ac-nrepl shouldn't be able to eval code, that > > > > means it's being coupled to the JVM state, which won't do for cljs or > > > > other sorts of repls-on-repls. > > > > > > > > > > > > > > > > > > > > On Mon, Jan 13, 2014 at 9:03 AM, Gary Trakhman <gary.trakh...@gmail.com > > > > (mailto:gary.trakh...@gmail.com)> wrote: > > > > > I've released a cljs port of clojure-complete: > > > > > > > > > > Here's the mailing list announcement, also inlined. > > > > > > > > > > https://groups.google.com/forum/#!topic/clojurescript/Dt1s4laHFXc > > > > > cljs-complete, A Clojure library designed to auto-complete > > > > > clojurescript based on cljs compiler state. > > > > > > > > > > - With leiningen: > > > > > [cljs-complete "0.1.0"] > > > > > > > > > > - Usage > > > > > > > > > > ;; env is pulled from cljs compiler state > > > > > => (completions @cljs.env/*compiler* "al" 'cljs.core) > > > > > ("alength" "alter-meta!") > > > > > > > > > > This is meant to hook into piggieback, that'll be the next thing I > > > > > try. I hope I can get some help with the hairy emacs bits :-). > > > > > > > > > > > > > > > On Tuesday, January 7, 2014 1:54:27 AM UTC-5, Bozhidar Batsov wrote: > > > > > > I'm cider's maintainer. The problem with code completion for > > > > > > ClojureScript is that the default mechanism is based on the > > > > > > Clojure-only library https://github.com/ninjudd/clojure-complete. > > > > > > As I don't use ClojureScript I haven't paid much attention to it so > > > > > > far. If there is a similar library for ClojureScript I might add > > > > > > support for it. Btw, cider will also pick up any nREPL middleware > > > > > > that provides a "complete" op - if there is a ClojureScript nREPL > > > > > > completion middleware it can be used with cider even now. Ideally > > > > > > at some point we'll have unified middleware supporting both Clojure > > > > > > & ClojureScript. > > > > > > > > > > > > On Sunday, January 5, 2014 10:30:44 PM UTC+2, Alexandru Nedelcu > > > > > > wrote: > > > > > > > > > > > > > > Hi, > > > > > > > > > > > > > > > > > > > > > I can’t get auto-completion or jumping to the definition of a > > > > > > > function in Emacs, while working with ClojureScript. Is this a > > > > > > > limitation of Emacs’ Cider > > > > > > > (https://github.com/clojure-emacs/cider) plugin? > > > > > > > > > > > > > > > > > > > > > I managed to get nRepl working by using the awesome > > > > > > > com.cemerick/austin (https://github.com/cemerick/austin) plugin > > > > > > > for lein (besides including the plugin in project.clj I’ve got no > > > > > > > other special configuration). So to work with ClojureScript, from > > > > > > > inside an nRepl session, I execute the following (not sure if > > > > > > > this is right): > > > > > > > > > > > > > > (cemerick.austin.repls/exec :exec-cmds ["google-chrome"]) > > > > > > > > > > > > > > In Emacs I do the same thing, by first starting the repl with > > > > > > > cider-jack-in and then I execute the above command. Compiling the > > > > > > > current buffer works (C-c C-k), I can also send expressions for > > > > > > > evaluation to the repl (C-x C-e). > > > > > > > > > > > > > > > > > > > > > However, as I said, auto-completion doesn’t work and neither does > > > > > > > jumping to the definition of a function. Is there any way to get > > > > > > > these features while working with ClojureScript in Emacs? Or > > > > > > > maybe in another IDE? > > > > > > > > > > > > > > > > > > > > > Thanks, > > > > > > > > > > > > > > > > > > > > > -- > > > > > > > Alexandru Nedelcu > > > > > > > www.bionicspirit.com (http://www.bionicspirit.com) > > > > > > > > > > > > > > PGP Public Key: > > > > > > > https://bionicspirit.com/key.aexpk > > > > > > > > > > -- > > > > > -- > > > > > 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 > > > > > (mailto: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 > > > > > (mailto:clojure%2bunsubscr...@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 unsubscribe from this group and stop receiving emails from it, > > > > > send an email to clojure+unsubscr...@googlegroups.com > > > > > (mailto:clojure%2bunsubscr...@googlegroups.com). > > > > > For more options, visit https://groups.google.com/groups/opt_out. > > > > > > > > -- > > > > -- > > > > 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 > > > > (mailto: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 > > > > (mailto: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 a topic in the > > > > Google Groups "Clojure" group. > > > > To unsubscribe from this topic, visit > > > > https://groups.google.com/d/topic/clojure/uXtd--Xv6vU/unsubscribe. > > > > To unsubscribe from this group and all its topics, send an email to > > > > clojure+unsubscr...@googlegroups.com > > > > (mailto:clojure+unsubscr...@googlegroups.com). > > > > > > > > For more options, visit https://groups.google.com/groups/opt_out. > > > > > > -- > > > -- > > > 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 > > > (mailto: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 > > > (mailto:clojure%2bunsubscr...@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 unsubscribe from this group and stop receiving emails from it, send an > > > email to clojure+unsubscr...@googlegroups.com > > > (mailto:clojure%2bunsubscr...@googlegroups.com). > > > For more options, visit https://groups.google.com/groups/opt_out. > > > > -- > -- > 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 > (mailto: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 > (mailto: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 a topic in the Google > Groups "Clojure" group. > To unsubscribe from this topic, visit > https://groups.google.com/d/topic/clojure/uXtd--Xv6vU/unsubscribe. > To unsubscribe from this group and all its topics, send an email to > clojure+unsubscr...@googlegroups.com > (mailto:clojure+unsubscr...@googlegroups.com). > For more options, visit https://groups.google.com/groups/opt_out. -- -- 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 unsubscribe from this group and stop receiving emails from it, send an email to clojure+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/groups/opt_out.
