Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!elroy.jpl.nasa.gov!usc!jarthur!nntp-server.caltech.edu!suncub.bbso.caltech.edu!EYCHANER
From: eychaner@suncub.bbso.caltech.edu (Amateurgrammer)
Newsgroups: comp.lang.c
Subject: Re: A quick question...
Message-ID: <1991Mar12.062941.2369@nntp-server.caltech.edu>
Date: 12 Mar 91 06:29:41 GMT
References: <1991Mar12.030759.26698@nntp-server.caltech.edu>
Sender: news@nntp-server.caltech.edu
Reply-To: eychaner@suncub.bbso.caltech.edu
Organization: Big Bear Solar Observatory, Caltech
Lines: 45

eychaner@suncub.bbso.caltech.edu (Amateurgrammer) writes:
*Note* A followup question follows....
>Is this legal...
>	unsigned char *pointer1;
>	short short_value;
>	...
>	*((short *) pointer1) = short_value;
>	...
>And does it do what I think it does, that is, assign short_value to the
>storage pointed to by pointer1?  I hope you understand what I mean...
>								-Glenn
Summary so far:  In short, yes it is legal.  However:
1) pointer1 must point to something large enough to hold a short.  Duh. :-)
2) Unless pointer1 is something like:
	pointer1 = (unsigned char *) (&something_short);
   alignment problems may result on some machines.  I figured that.
3) The exact effect is not machine portable (depending on your machine's
   order of storage for short, I guess).
Followup question:
The reason for this is I am writing a subroutine which stores some values
in an array of short OR unsigned char.  I thought the solution would be:
(this is a HIGHLY simplified version; the real routine is complex enough
that I don't need two nearly identical copies floating about...)

int do_array (void *array, int size, int type)
{
unsigned char *arrayptr;
int i;

arrayptr = array;   /* If only void * arithmetic were legal in VAX C */
for (i = 0; i < size; i++) {
    if (type == UNSIGNED_CHAR) {
        /* My precedence rules say this is right */
        *arrayptr++ = some_value();  /* Declared properly elsewhere */
        }
    else if (type == SHORT) {
        /* My precedence rules ALSO say THIS is right */
        *(short *)arrayptr++ = othervalue();  /* Declared elsewhere */
        }
    }
}
******************************************************************************
Glenn Eychaner - Big Bear Solar Observatory - eychaner@suncub.bbso.caltech.edu
"...this morning's unprecedented solar eclipse is no cause for alarm."
                                                               -_Flash Gordon_