[m-dev.] for review: big ints again

[Fergus Henderson]

On 08-Apr-1998, Bert Thompson <aet at hydra.cs.mu.oz.au> wrote:
> Implementation of an arbitrary precision integer type and
> operations on it.
> 
> library/integer.m
> test/general/bigint.m

The log message is not quite the usual format;
something like

	library/integer.m:
		Implementation of an arbitrary precision integer type and
		operations on it.
	 
	tests/general/bigint.m:
		Some test cases for library/integer.m.

would be better.

It would probably be better to name the test case integer_test.m
rather than bigint.m.  This is consistent with the names used for
various other test cases, e.g. relation_test.m.

> :- func integer:'/'(integer, integer) = integer.
> 
> :- func integer:'//'(integer, integer) = integer.
...
> integer:'/'(X1, X2) =
> 	big_quot(X1, X2).
> 
> integer:'//'(X1, X2) =
> 	big_quot(X1, X2).

You should delete the `/' function, or change it to return `rational'
rather than `integer'.

> :- func int_to_integer(int) = integer.
> int_to_integer(D) = i(signum(D), pos_int_to_digits(AD)) :-
> 	int__abs(D, AD).

This function does the wrong thing for MININT.
(See, isn't code review wonderful? ;-)

> :- func signum(int) = int.
> signum(N) = SN :-
> 	(N < 0 ->

Please s/(N/( N/

> :- pred pos_geq_rev(list(digit), list(digit)).
> :- mode pos_geq_rev(in, in) is semidet.
> pos_geq_rev(Xs, Ys) :-
> 	list__reverse(Xs, RXs),
> 	list__reverse(Ys, RYs),
> 	C = big_cmp(i(1, RXs), i(1, RYs)),
> 	( C = greaterthan ; C = equal).

Please s/equal)/equal )/

> :- func big_pow(integer, integer) = integer.
> big_pow(A, N) = P :-
> 	( N = integer(0) ->
> 		P = integer(1)
> 	; big_odd(N) ->
> 		P = A * big_pow(A, N-integer(1))
> 	; % even
> 		P = big_sqr(big_pow(A, N/integer(2)))
> 	).

You should use `//' here rather than `/'.

> :- func big_sqr(integer) = integer.
> big_sqr(A) = A * A.
> 
> :- pred big_odd(integer).
> :- mode big_odd(in) is semidet.
> big_odd(N) :-
> 	( N = integer(0) ->
> 		fail
> 	;
> 		N = i(_S, [D|_Ds]),
> 		D mod 2 = 1
> 	).

Isn't the explicit test for zero unnecessary?

	big_odd(N) :-
 		N = i(_S, [D|_Ds]),
 		D mod 2 = 1.

I'd like to see another diff please.

-- 
Fergus Henderson <fjh at cs.mu.oz.au>  |  "I have always known that the pursuit
WWW: <http://www.cs.mu.oz.au/~fjh>  |  of excellence is a lethal habit"
PGP: finger fjh at 128.250.37.3        |     -- the last words of T. S. Garp.