[climacs-devel] A little on rings

[Robert Strandh]

Hello Elliott, 

Elliott Johnson writes:
 > 
 > Hope everyone's holidays have been going well.  :)

Sure, lots of hacking.  

 > I've hardly had time but I've worked up a ring.lisp file.  It's 
 > fairly simple.   Classes for generic ring & kill-ring, plus generic 
 > functions for ring-push, ring-pop, rotate-ring, initialize-kill-ring, 
 > and kill-in-region.

Sounds good. 

 > The kill-ring structure is nothing outstanding right now either, 
 > simply a slot-value for an array, a max size fixnum, and a offset 
 > fixnum.  The kill-ring being a vector can handle any valid lisp 
 > structure, so weird cuts and inserts are no problem. 
 > 
 > If anyone else has cool ideas or better implementations for rings, 
 > please speak up.

Yes, in fact, if you are going to rotate it and/or do arbitrary
inserts and deletes, a flexichain is particularly useful.  And if you
want an offset into the kill ring, you might do that with a
flexicursor into a cursorchain.  All of this is in the flexichain
library already used for the buffer data structure, so no extra
libraries necessary. 

 > Things I still need to do: create a few helper functions (inspired 
 > by hemlock) and create the CLIM commands, keybindings, etc, so I'll 
 > have it down before leaving for LA.  Cl.net guys havn't replied to 
 > my account email yet, so hopefully that comes togeather as well.  
 > If not I'll email it before I go :)

No rush.  I'd say wait for the account. 

Have a nice trip.
-- 
Robert Strandh

---------------------------------------------------------------------
Greenspun's Tenth Rule of Programming: any sufficiently complicated C
or Fortran program contains an ad hoc informally-specified bug-ridden
slow implementation of half of Common Lisp.
---------------------------------------------------------------------