Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!pacific.mps.ohio-state.edu!linac!att!cbnews!cbnewsm!lfd
From: lfd@cbnewsm.att.com (Lee Derbenwick)
Newsgroups: comp.lang.c
Subject: Re: Source File Organization
Summary: correction on trick - wrong macro definitions
Message-ID: <1991Feb27.185509.1452@cbnewsm.att.com>
Date: 27 Feb 91 18:55:09 GMT
References: <1991Feb26.045242.23453@rfengr.com> <1991Feb27.182525.29758@cbnewsm.att.com>
Organization: AT&T Bell Laboratories
Lines: 37

In article <1991Feb27.182525.29758@cbnewsm.att.com>,
I gave a trick for keeping stuff in sync; but I left commas out of
the two macro definitions (SORRY ABOUT THAT!):
> 
> letters.h contains:
> 
>   LETTER(A, 'A')
>   LETTER(B, 'B')
>   LETTER(C, 'C')
>   LETTER(D, 'D')
>   #undef LETTER
> 
> Then, to get your two examples:
> 
/*
>   #define LETTER(A,B) A
*/
    #define LETTER(A,B) A,	/* should have been */
>   typedef enum {
>   #include "letters.h"
>   } CMD;
> 
/*
>   #define LETTER(A,B) B
*/
    #define LETTER(A,B) B,	/* should have been */
>   char ltrs[] = {
>   #include "letters.h"
>   };

This trick relies on C allowing a trailing comma in an enum or
initializer list; this and other machine-generated C code were
the reasons that the trailing comma _is_ allowed.

 -- Speaking strictly for myself,
 --   Lee Derbenwick, AT&T Bell Laboratories, Warren, NJ
 --   lfd@cbnewsm.ATT.COM  or  <wherever>!att!cbnewsm!lfd