"J. Kim" wrote:
> 
> Dear Mark,
>  I have tried using the 'make' instead of 'gmake', since it would to take
> quite some time for me
> to get the 'gmake' program(or part of the programs) and install.
> After issuing 'make newunix' under the directory .... /source/unix,
> I got following messages:
> (I have not changed cc to gcc yet.)
> 
> Thanks in advance,
> Jong
> -------------------------------------------------
> /temp/povray31/source/unix >make newunix
> cc -O2 -ansi -c -DCOMPILER_VER=\".`uname`.cc\" -I. -I.. -I../libpng
> -I../zlib -I
> /usr/bin/X11  /include ../polysolv.c
> cc: Warning: ../polysolv.c, line 66: # not in column 1 is ignored, skipping
> to e
> nd of line.
>  #define FUDGE_FACTOR1 1.0e12
> -^
> cc: Warning: ../polysolv.c, line 69: # not in column 1 is ignored, skipping
> to e
> nd of line.
>  #define FUDGE_FACTOR2 -1.0e-5
> -^
> cc: Warning: ../polysolv.c, line 72: # not in column 1 is ignored, skipping
> to e
> nd of line.
>  #define FUDGE_FACTOR3 1.0e-7
> -^
> cc: Error: ../polysolv.c, line 1085: In this statement, "FUDGE_FACTOR1" is
> not d
> eclared.
>       if (fabs(biggest / x[i]) > FUDGE_FACTOR1)
> ---------------------------------^
> *** Exit 1
> Stop.
> *** Exit 1 (ignored)

Those errors are coming from cc, not make.  It looks like your C
compiler (or preprocessor) doesn't like the fact that some of the
#define's in this file don't start immediately at the beginning of the
line.  It looks like the author of this piece of code decided to make
the code a little more human-friendly by indenting his cpp directives,
but your compiler doesn't like it.  Just go in and erase the space from
the beginning of lines 66, 69, and 72, and let me know if it compiles
without complaining.

Of course, changing cc to gcc would probably fix that too. ;-)  I'd
still like to know whether that fixes the problem, since I'd prefer the
code be more portable.

-Mark Gordon

Post a reply to this message