Ken Tilton  <kentilton@gmail.com> wrote:
+---------------
| (position #\a "bbaa" :start 42) -> 76?
| 
| puh-lease.
| 
| Actually, I am whining about CLHS, doesn't say word about
| oddball values  for start.
+---------------

Well, actually, it sort of does, in that:

    Function POSITION, POSITION-IF, POSITION-IF-NOT
    ...
    Arguments and Values:
    ...
    start, end---bounding index designators of sequence.
        The defaults for start and end are 0 and nil, respectively.

And then if you go to the glossary entries for (all three of!)
BOUNDING INDEX DESIGNATOR, BOUNDING INDEX, and BOUNDED, you
eventually get to this bit:

     ...restricted to a subrange of the elements of S that...

and 42 is certainly *NOT* a designator for a :START that's restricted
to a subrange of "bbaa". So 42 is not a proper type for a :START arg
in this case.

And CLHS 1.4.4.3 "The ``Arguments and Values'' Section of a Dictionary
Entry" very clearly states:

    Except as explicitly specified otherwise, the consequences
    are undefined if these type restrictions are violated.

And we all know about the CLHS and undefined behaviors...  ;-}  ;-}


-Rob

p.s. Of course, any *decent* implementation will be more gentle,
and instead of starting WW3 or emitting snot monkeys will throw you
a handleable type error, e.g., CMUCL:

    > (multiple-value-list (ignore-errors (position #\a "bbaa" :start 42)))

    (NIL #<TYPE-ERROR {48970025}>)
    > (format t "~a~%" (second *))
    Type-error in KERNEL::INVALID-ARRAY-INDEX-ERROR-HANDLER:
	    42 is not of type (INTEGER 0 (4))
    NIL
    > 

-----
Rob Warnock			<rpw3@rpw3.org>
627 26th Avenue			<URL:http://rpw3.org/>
San Mateo, CA 94403		(650)572-2607