Tony Martinez <tonyms@...> writes:
> Shouldn't datum then hold the value of PLACE here, rather than the
> place itself?
> I haven't been able to find a clearer spot to back up my expected
> behaviour, though, which is that the type-error-datum hold the object
> which failed the type test, rather than the /place/ which does so.
> Lispworks and CLisp share the SBCL behaviour, which makes me wonder
> whether my expectations are leading me astray.
I don't think your expectations are wrong. I'm slightly surprised at
the commonality of behaviour, but I've changed sbcl's in 0.8.12.2; if
there are howls of protest from hordes of heretofore silent users,
then we'll be able to examine the case more closely. But in general,
it seems logical to me that
(typep (type-error-datum c) (type-error-expected-type c))
is the complaint that the type-error is reporting.
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)