Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!cornell!rochester!rutgers!att!ulysses!andante!princeton!jonlab!jon
From: jon@jonlab.UUCP (Jon H. LaBadie)
Newsgroups: comp.lang.c
Subject: Re: C inconsistency?
Summary: Oh the memory still hurts
Message-ID: <772@jonlab.UUCP>
Date: 12 Apr 89 06:33:51 GMT
References: <1989Mar30.192903.28426@gpu.utcs.toronto.edu> <15@masscomp.UUCP>
Organization: 4455 Province Line Rd., Princeton, NJ 08540
Lines: 47

Rajiv's question, and Farrell's response recall painful memories of a
similar "bug".

In article <15@masscomp.UUCP>, ftw@masscomp.UUCP (Farrell Woods) writes:
> In article <1989Mar30.192903.28426@gpu.utcs.toronto.edu> sarathy@gpu.utcs.UUCP (Rajiv Sarathy) writes:
> 
> >	struct element {
> >	    char *key;
> >	    int level;
> >	    struct element **fpointer;
> >	}
> >	struct element *head[LEVELCAP];     /* array of pointers to struct */
> 
> Indeed, you're missing a semi-colon.  (after the closing brace j.l.)

I developed some code for a client with the beginning of the program
organized like this:

	struct foo {
		...
		...
	}

	/*
	**
	**   long ( > 1 page ) preamble of comments
	**
	*/

	main(...)
	{
		...
	}

On my development machine, the code worked fine and I delivered it
and recompiled it on the clients system.  The latter had a different
cpu.  It seemed to run fine ... except that it core dumped when it
exited.  Ran fine until it exited, just that d****d core file each
time you used it.  Try to convince someone paying the bill

	"It works fine, just dumps core each time you use it."

Yeah, sure.

Took me quite a while to realize main was "trying" to return a
68 byte structure, not an int.  Whatever was receiving the return
value of main was surely surprised!