Is there a right answer to whether this should be allowed?
I'm motivated to ask because when testing code based on Robert Brown's
list.lisp patch, clocc-ansi-test pointed out that LIST-REMOVE-DUPLICATES*
uses this style of call (to MEMBER) in its implementation. So if this
style of call is OK, I should change the patch, while if this style
of call is not OK, I should change LIST-REMOVE-DUPLICATES*.
I seem to remember someone (Kent Pitman?) writing that functions which
supported &OPTIONAL and &KEY arguments should try to treat explicit
NIL arguments as equivalent to unsupplied arguments, since that makes
it easier when you want to use one such function in the implementation
of another (like this). But I don't remember an authoritative answer
about what ANSI CL functions are supposed to do, and a little looking
in the spec didn't turn one up.
William Harold Newman <william.newman@...>
Furious activity is no substitute for understanding. -- H. H. Williams
PGP key fingerprint 85 CE 1C BA 79 8D 51 8C B9 25 FB EE E0 C3 E5 7C