Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!rutgers!princeton!allegra!ulysses!sfmag!sfsup!mpl
From: mpl@sfsup.UUCP
Newsgroups: comp.lang.c
Subject: Re: What can I get away with
Message-ID: <1578@sfsup.UUCP>
Date: Wed, 1-Jul-87 09:48:08 EDT
Article-I.D.: sfsup.1578
Posted: Wed Jul  1 09:48:08 1987
Date-Received: Sat, 4-Jul-87 06:47:38 EDT
References: <608@zen.UUCP> <2299@hoptoad.uucp> <21211@sun.uucp> <830@omepd> <8190@linus.UUCP>
Organization: AT&T-IS, Summit N.J. USA
Lines: 49
Summary: yes

In article <8190@linus.UUCP>, jfjr@mbunix.UUCP writes:
> Suppose I declare an array of pointers to functions in one module.
	[deleted stuff]
> pointers) and call the functions pointed to even if they are not
> declared externally (to avoid further confusion all these functions

YES, you can!  The only thing is you don't get any sanity checking on the
return value of the function (with K&R flavored C).  Consider

file1.c:

int	(*a[10])();

static next;

install(f)
int	(*f)();
{
	a[next++] = f;
}

gleep()
{
	int	n;
	.
	.
	n = (*a[0])(3);
	.
	.
}

file2.c

char *
foobar()
{
	return NULL;
}

main()
{
	install(foobar);
	gleep();
	.
	.
}

"n" in gleep holds an "int", but (*a[0])(3) returns a char *.
The compiler will not pick up this bug, although lint will.