lynx-dev
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: lynx-dev dev3/4 problem (progress!)


From: pg
Subject: Re: lynx-dev dev3/4 problem (progress!)
Date: Mon, 22 Feb 1999 19:21:25 -0700 (MST)

In a recent note, Philip Webb said:

> Date: Mon, 22 Feb 1999 20:11:22 -0500 (EST)
> 
> i've been continuing to try to track down the problem with dev.17 ,
> in which Lynx displays garbage & doesn't respond to key commands.
> 
>  #define EBCDIC 1  /* CF_EBCDIC */
> 
> i tried commenting out this line, but linking failed with a message:
> 
>  ld32: error 33: unresolved text symbol "FixCharacters"
>   -- first referenced in  LYMain.o ,
> 
Also comment out "#define NOT_ASCII 1"

> so why is configure assuming i have EBCDIC on this IRIX 5.3 ?!
>   
The test program is:
====================================================================
#line 9979 "configure"
#include "confdefs.h"

int main() {
 /* TryCompile function for CharSet.
   Treat any failure as ASCII for compatibility with existing art.
   Use compile-time rather than run-time tests for cross-compiler
   tolerance.  */
#if '0'!=240
#error "Character set is not EBCDIC"
#endif
; return 0; }
====================================================================

This should generate a compile-time error on an ASCII platform and
compile without error on an EBCDIC platform.

The hazards I can imagine are that somehow the "#if" branches the
wrong way, or that "#error" fails to generate a compilation error.

Perhaps a junk line within the #if ... #endif will force the intended
error:

    #if '0'!=240
    #error "Character set is not EBCDIC"
    boogaboogabooga is a syntax error
    #endif

Do any compilers quietly ignore unrecognized preprocessor directives?

-- gil

reply via email to

[Prev in Thread] Current Thread [Next in Thread]