Matthew D Swank wrote:
Matthew D Swank wrote:
Would it be possible to have destructure-case match against arbitrary, non-nil atoms (patch attached)?
Forget it. That patch obviously won't work as is, and I can't seem to find the patched version of slime I was using at the moment.
This one should work. Matt -- "You do not really understand something unless you can explain it to your grandmother." — Albert Einstein. --- slime/swank.lisp.orig 2006-06-18 13:21:54.000000000 -0500 +++ slime/swank.lisp 2006-06-30 22:07:03.000000000 -0500 @@ -293,19 +293,22 @@ A cross between `case' and `destructuring-bind'. The pattern syntax is: ((HEAD . ARGS) . BODY) +or: + (ATOM . BODY) The list of patterns is searched for a HEAD `eq' to the car of -VALUE. If one is found, the BODY is executed with ARGS bound to the -corresponding values in the CDR of VALUE." +VALUE (or ATOM `eq' to VALUE as the case may be). If one is +found, the BODY is executed with ARGS bound to the corresponding +values in the CDR of VALUE." (let ((operator (gensym "op-")) (operands (gensym "rand-")) (tmp (gensym "tmp-"))) `(let* ((,tmp ,value) - (,operator (car ,tmp)) - (,operands (cdr ,tmp))) + (,operator (or (and (listp ,tmp) (car ,tmp)) ,tmp)) + (,operands (or (and (listp ,tmp) (cdr ,tmp))))) (case ,operator ,@(loop for (pattern . body) in patterns collect - (if (eq pattern t) - `(t ,@body) + (if (and pattern (atom pattern)) + `(,pattern ,@body) (destructuring-bind (op &rest rands) pattern `(,op (destructuring-bind ,rands ,operands ,@body)))))