I think that you want drscheme/private/rep.ss's submit-predicate to reflect the current language. All of the stuff that is #lang-line specific is connected in via the module-language (ie the class implementing drscheme:module-langauge:module-language<%> interface) and there's no path that connects that class to the rep's submit predicate for now. I guess one would add something to drscheme:language:language<%> and then provide a default implementation in one of the functions that produces those things. This is, unfortunately, likely to be a backwards incompatible change tho. So maybe there is some other way things should be organized going forwards. Like perhaps the repl should start doing things like (is-a? ... drscheme:module-language:module-language<%>) to give special abilities to the module language.
Or maybe just wait until I find time. :( Robby On Tue, Apr 27, 2010 at 6:21 PM, Jay McCarthy <jay.mccar...@gmail.com> wrote: > Yes, I thought so. Where would be a place to start? I put my > repl-submit? in place with drracket:modes:add-mode, but that functions > is never called in the tree, so it must happen somewhere else and I'm > at a loss for where to start. > > Jay > > On Tue, Apr 27, 2010 at 5:00 PM, Robby Findler > <ro...@eecs.northwestern.edu> wrote: >> That'ms something that would have to be part of the (newish) >> read-language stuff. Ie, drracket would have to ask your language for >> a submit predicate and the use it in the repl, I guess. At least >> that's what would be the easiest change to make. >> >> Robby >> >> On Tuesday, April 27, 2010, Jay McCarthy <jay.mccar...@gmail.com> wrote: >>> Danny reported a bug in my datalog package, so I've been updating it >>> to the recent versions, while doing it I thought I would try removing >>> the need for the Dr... tool. The new color-lexer is really nice for >>> getting syntax highlighting in the module language, so I was able to >>> do that. The only two things I can't do are getting a different >>> comment character (%) and getting a different REPL submit predicate. >>> The first is easy to add. The second is more worrisome, because in my >>> current implementation it is part of the "Mode" rather than the tool >>> itself. >>> >>> Any pointers for implementing the second as a language info key? I'm >>> pretty much totally lost when I try to follow the DrRacket source. >>> Rather than just email Robby, I thought I'd email the list to get more >>> press for the q/a. >>> >>> Jay >>> >>> -- >>> Jay McCarthy <j...@cs.byu.edu> >>> Assistant Professor / Brigham Young University >>> http://teammccarthy.org/jay >>> >>> "The glory of God is Intelligence" - D&C 93 >>> _________________________________________________ >>> For list-related administrative tasks: >>> http://list.cs.brown.edu/mailman/listinfo/plt-dev >>> >> > > > > -- > Jay McCarthy <j...@cs.byu.edu> > Assistant Professor / Brigham Young University > http://teammccarthy.org/jay > > "The glory of God is Intelligence" - D&C 93 > _________________________________________________ For list-related administrative tasks: http://list.cs.brown.edu/mailman/listinfo/plt-dev
