Path: utzoo!attcan!uunet!cs.utexas.edu!yale!think!snorkelwacker!spdcc!ima!haddock!karl
From: karl@haddock.ima.isc.com (Karl Heuer)
Newsgroups: comp.lang.c
Subject: Re: multi-d arrays and types
Message-ID: <16289@haddock.ima.isc.com>
Date: 26 Mar 90 19:55:47 GMT
References: <1990Mar26.155319.23986@ccu.umanitoba.ca>
Reply-To: karl@haddock.ima.isc.com (Karl Heuer)
Organization: Interactive Systems, Cambridge, MA 02138-5302
Lines: 28

In article <1990Mar26.155319.23986@ccu.umanitoba.ca> rpjday@ccu.umanitoba.ca writes:
>  int calendar[12][31];
>  int (*monthptr)[31];
>  for (monthptr = calendar; monthptr <= calendar[11]; monthptr++) ...
>[Assignment and increment are okay, but the comparison has a problem.]

Since "calendar" has type "array of array of int", "calendar[11]" has type
"array of int", which decays into "pointer to int", which does not match
the "pointer to array of int" on the left side.  What you really want is
"&calendar[11]", but...

>Putting an "&" in front of "calendar[11]" makes it even worse as the compiler
>complains that I now have "& before array or function", tosses the "&", then
>generates the first warning.

Unfortunately, in pre-ANSI C it was not legal to explicitly take the address
of an array.  (This was a botch; the relationship between arrays and pointers
was not as well defined in the old days.)  Your options include: [0] use an
ANSI C compiler; [1] patch your compiler to remove (nothing to add!) the line
that makes this illegal; [2] rewrite your code with a dummy struct around the
array; [3] (recommended) change the expression from "&calendar[11]" to
"calendar+11", which is equivalent (due to the definition of "[]") but works
right even in pre-ANSI C.

Oh, I also recommend that you use 12 rather than 11 as the limiting subscript,
and change the relational operator to a strict "<".

Karl W. Z. Heuer (karl@ima.ima.isc.com or harvard!ima!karl), The Walking Lint