Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!watmath!clyde!rutgers!ames!amdcad!amd!intelca!oliveb!sun!gorodish!guy
From: guy@gorodish.UUCP
Newsgroups: comp.lang.c++
Subject: Re: Functions as arguments ("cfront" bug?)
Message-ID: <25710@sun.uucp>
Date: Fri, 14-Aug-87 22:38:25 EDT
Article-I.D.: sun.25710
Posted: Fri Aug 14 22:38:25 1987
Date-Received: Sun, 16-Aug-87 08:58:03 EDT
References: <297@pvab.UUCP>
Sender: news@sun.uucp
Lines: 47

> I have trouble taking the address of a function, sending that address
> as an argument to a function, declaring that argument as a function
> address in the called function and then call the function pointed-to
> by the address. What is the correct way to do this? The usual C way
> doesn't work. We have C++ 1.1.

I tried the following:

1	#include <stdio.h>
2
3	static int
4	func1(int x)
5	{
6		return x + 1;
7	}
8
9	static int
10	func2(int (*funcp)(int x))
11	{
12		return (*funcp)(33);
13	}
14
15	int
16	main(int argc, char **argv)
17	{
18		(void) printf("%d\n", func2(func1));
19	}

with C++ 1.1 and "cfront" got very upset, giving no less than 3 syntax errors
on line 10, one on line 11, three on line 12, and 1 on line 13.  However,
replacing "func2" with

	static int
	func2(funcp)
	int (*funcp)(int x);
	{
		return (*funcp)(33);
	}

replaced all those complaints with one warning of an "old style definition of
func2()", and yielded a program that correctly printed "34".

I can see nothing syntactically wrong with the first definition of "func2".
Looks like a "cfront" bug to me....
	Guy Harris
	{ihnp4, decvax, seismo, decwrl, ...}!sun!guy
	guy@sun.com