On Sat, 2 Dec 2000, David Alan Black wrote: > > If that were possible, would there be benefits in converting > > Emacs to use Ruby instead of (e)lisp? > > It would have to be not so much a conversion as a reimplementation. > The interdependence between Emacs and Emacs Lisp goes beyond language > choice or mode. For one thing, most of Emacs is written in Emacs That is why I thought it might be possible... > Lisp. If you were to rewrite Emacs, you'd really also have to create > Emacs Ruby, incorporating all the datatypes and primitives of Emacs > Lisp. You wouldn't be able to translate something like this: > > (defun list-buffers-other-window () > "Switch to buffer-list window upon listing buffers" > (interactive) > (list-buffers) > (switch-to-buffer-other-window "*Buffer List*")) > > without implementing buffers in Ruby, and that's just the > beginning.... OK. It sounds like it would be *possible*, but too steep a hill to be worth climbing. > > > I'm thinking of: you'd get more readable code; it might be possible > > to integrate it tightly with other ruby code, and to C using Ruby's > > API. Could it be faster (after tuning)? Could the Ruby syntax mode > > be REALLY intelligent if you did it this way? > > I can't comment on Point 1, because I've always found elisp pretty > readable. (Considering that it's a programming language :-) If you're I have always found lisp unreadable, but I have not come across a good introduction to it. I think I take your point though: any programming langauge of any power looks unreadble to the outsider. > interested in integrating Ruby into Emacs, in the sense of being able > to manipulate Emacs Lisp datatypes directly with Ruby scripts, then it > might make more sense to think about translating *from* Ruby *to* > elisp. (I'm not volunteering, though :-) I suppose if elisp is implemented in C then there could be a way, but I think you have successfully squashed my idea, :-) which is fine, and what I thought might happen. > > > David > Hugh hgs / dmu.ac.uk