by-ref parameters, aggregate constants, etc

[Tainter]

> >In article <6229 at sun.uucp> guy at sun.uucp (Guy Harris) writes:
> >>This may, in fact, also be an argument for reference types a la C++; there,
> >>your routine "foo" would be written as
> >> 	foo(c) char &c; { ...
> I have no great love for this syntax either; but how else do you
> propose to add by-reference parameters?  (I believe that by-reference
> parameters are, in general, bad, at least if I cannot tell from
> the caller that the parameter is modifiable.  I would not add them
> at all.  Clearly Bjarne Stroustrup and I disagree.)
> In-Real-Life: Chris Torek, Univ of MD Comp Sci Dept (+1 301 454 1516)

The win for by-reference parameters is the syntax of their use in the callEE.
I do extensive personal programming in pascal and C.  When rereading pascal
with by-reference parameters I do at times lose track of what is a var
parameter at the callER level but never at the callEE level.  C fixes this
but at the expense of excessive clutter and confusion with pointers inside
the callEE.

I propose:

dowa(a)
int &a; 	/* this declares a to be a pointer to an whose use has
		    the syntax of an auto int. this is a pascal style var
		    parameter implementation of by-reference */
{
    int b;

    a = rand();
    b = rand() + a;
    return b;
}

main()
{
    int sr,fr;

    /* In the CALLER the pascal style var parameter implementation
	gets superceded by explicit address of and pointer syntax */

    sr = dowa(&fr);	/* still required to pass address of */
    printf("first random = %d, sum of two randoms = %d\n",fr,sr);

    /* this would also be legal: */
    pfr = malloc(sizeof (int));
    sr = dowa(pfr);
    printf("first random = %d, sum of two randoms = %d\n",*pfr,sr);
}

Basically it provides the same "pointer" syntax at the callER and
a friendly by-reference syntax at the callEE.

--j.a.tainter