From: Christophe R. <cs...@ca...> - 2003-10-29 19:27:06
|
Daniel Barlow <da...@te...> writes: > Christophe Rhodes <cs...@ca...> writes: > >> The specifics of what an unimplemented Unixoid function should do is a >> bit of an open question. I'm quite partial to something like this for >> the not found case: > [assorted style-warn/warn/error] > > i think I would like there to be some way for Lisp code to test > whether a given function is defined, preferably without having to call > it and handler-case the error. I haven't thought about this too hard; > it's just a gut reaction. This, I think, is a good point. What's in CVS at the moment (sbcl-0.8.5.15) has the behaviour that BOUNDP returns NIL for unfound constants, and FBOUNDP returns NIL for unfound functions. Also, because we detect availability of the call at sb-posix-compile-time, we no longer need to maintain #+linux or #-darwin conditionals in interface.lisp. On the downside, we now need RUN-PROGRAM to build sb-posix, which probably causes more pain for our MacOS friends :-/ Cheers, Christophe -- http://www-jcsu.jesus.cam.ac.uk/~csr21/ +44 1223 510 299/+44 7729 383 757 (set-pprint-dispatch 'number (lambda (s o) (declare (special b)) (format s b))) (defvar b "~&Just another Lisp hacker~%") (pprint #36rJesusCollegeCambridge) |