Path: utzoo!attcan!uunet!crdgw1!underdog!volpe
From: volpe@underdog.crd.ge.com (Christopher R Volpe)
Newsgroups: comp.lang.c
Subject: Re: The Sins of K&R
Message-ID: <12221@crdgw1.crd.ge.com>
Date: 27 Sep 90 18:53:53 GMT
References: <31604@nigel.ee.udel.edu> <12180@crdgw1.crd.ge.com> <2487@lectroid.sw.stratus.com>
Sender: news@crdgw1.crd.ge.com
Reply-To: volpe@underdog.crd.ge.com (Christopher R Volpe)
Lines: 49

In article <2487@lectroid.sw.stratus.com>, jmann@angmar.sw.stratus.com
(Jim Mann) writes:
|>This still doesn't do everything you might want. It doesn't handle
|>the situation in which you do something for one case and then 
|>fall through to the next to do some additional stuff:
|>
|>switch(ch)
|>     {
|>     case 'a':
|>	   do_a();
|>     case 'b':
|>           do_a_or_b();
|>	   break;
|>	.
|>	.
|>	.
|>     }     

Ok, how about requiring the programmer to make it explicit when he/she
wants it to fall through, rather than making that the default? Like so:

switch(ch)
     {
     case 'a':
        do_a();
        continue;
     case 'b':
        do_a_or_b();
     }

or perhaps allow matching more than one case:
switch(ch)
     {
     case 'a':
        do_a();
     case 'a','b':
        do_a_or_b();
     }

This probably belongs in alt.lang.cfutures...
|>
|>Jim Mann
|>Stratus Computer
|>jmann@es.stratus.com
                                                      
==================
Chris Volpe
G.E. Corporate R&D
volpecr@crd.ge.com