So I thought that I'd post here to keep this particular meme going. 
I've looked at the Slime/Swank code.
Slime frontend -
 - written in Emacs Lisp and hosted in the Emacs editor
 - introduces a new editing mode in Emacs
 - delivers s-expressions to the Swank backend via a TCP port, I guess
that the s-exps are probably just plain text over the port (hmm, not
quite - there looks to be 6 digits at the start which is the length of
the s-exp)
 - processes the Swank return data

Swank
 - a remote TCP/IP server that listens on (defaultly) port 4005
 - executes the incoming s-expressions and sends the result back to Slime

I suspect that it ought to be pretty easy to talk to a running
Swank-enabled image, it looks do just be a matter of sending
s-expressions of the following form
"000015(format t "hi")".
It is probably a good idea to snoop the normal communications from
Slime<->Swank , I feel that this ought to be trivial to do via some
sort of port redirection, but I don't know exactly how to do this yet
- I'll look into that next probably.

As to the basic design of the Slime for Vim frontend, Larry mentioned
writing it in Perl (Vim supports Perl as a scripting language). 
Should we consider Python?  Lots of other applications embed Python,
would we be helping other potential Lisp environments by writing the
Swank frontend largely in Python?  I don't mind either way, just a
thought I had.
We should aim to provide all the features that the Emacs Slime does,
but with more Vim-like key access.

I'm sure I've missed at least a few things here though

Cheers
Brad
_______________________________________________
Gardeners mailing list
[email protected]
http://www.lispniks.com/mailman/listinfo/gardeners

Reply via email to