It would be fine if this only worked for specialized arrays, where the
elements *are* contiguous.  In that case you don't have to do all the
divisions and remainders: you look down at the storage-class body and
work from that.

On Thu, Apr 9, 2026 at 8:22 PM Bradley Lucier <xxxxxx@purdue.edu> wrote:
>
> On 4/9/26 14:59, John Cowan wrote:
> >
> > Common Lisp has an accessor (array-major-aref a n) which retrieves the
> > nth element of an array enumerated in lexicographical order.  As such,
> > it also comes with a setter:  (setf (array-major-aref a n) x) where x
> > is the new value.  Setf is equivalent to SRFI 17's version of `set!`,
> > and replaces essentially all mutator functions in CL (rplaca and
> > rplacd are still there for hysterical raisins).
> >
> > It seems partly possible to emulate this using array-body and
> > array-indexer, but then you have to know what kind of body you have
> > (vector, @vector, or something else).  I think it would be worth
> > adding these.  (Maybe there's a way to do them already, in which case
> > it would be worth pointing it out in 231bis.)
>
> I don't know Common Lisp.  I'll repeat, I don't know Common Lisp.
>
> I've looked at the Common Lisp spec for arrays, including "The Arrays
> Dictionary"
>
> https://www.lispworks.com/documentation/HyperSpec/Body/c_arrays.htm
>
> It appears to me from reading the spec that array elements are stored
> contiguously in increasing memory locations for CL arrays.  (CL arrays
> can be "displaced", which leaves elements in order.)  That makes
> row-major-aref useful in CL.
>
> "Sharing" specialized arrays as in Kawa (and, to be clear, I don't know
> Kawa) and SRFI 231 (and NumPy and Racket and ...) means that elements
> that are "adjacent" in row-major order can be widely dispersed in memory.
>
> And that doesn't even get into "generalized" arrays.
>
> So the only way to implement something like array-major-aref would be
> something like Kawa's rowMajorToEffectiveIndex in
> ./gnu/lists/Arrays.java (and speaking of things I don't know, I don't
> know Java) which would take your number n and compute a multi-index
>
> i_0 i_1 ... i_d-1
>
> by finding the remainder and quotient of n by successive dimension sizes
> ("width"s in SRFI 231).
>
> That is not an efficient way to access array elements.  To do something
> to all array elements in row-major order, it would be better to use a
> bulk array operation like array-for-each (or interval-for-each if you
> want to work at the multi-index level), which steps through the array
> elements in row-major order.
>
>