[Groff] Re: signedness problems

[Thomas Klausner]

On Thu, Jul 03, 2003 at 09:12:29AM +0200, Werner LEMBERG wrote:
> Since `char' is always signed in groff, isn't there another
> possibility to fix this?

I believe the correct way is to just explicitly spell out
this assumption in the source code.

To quote from the gcc man page (-fsigned-char also exists):
`-funsigned-char'
     Let the type `char' be unsigned, like `unsigned char'.

     Each kind of machine has a default for what `char' should be.  It
     is either like `unsigned char' by default or like `signed char' by
     default.

     Ideally, a portable program should always use `signed char' or
     `unsigned char' when it depends on the signedness of an object.
     But many programs have been written to use plain `char' and expect
     it to be signed, or expect it to be unsigned, depending on the
     machines they were written for.  This option, and its inverse, let
     you make such a program work with the opposite default.

     The type `char' is always a distinct type from each of `signed
     char' or `unsigned char', even though its behavior is always just
     like one of those two.

> I've thought of a configure test which sets
> the appropriate compiler flag.  autoconf provides the macro
> AC_C_CHAR_UNSIGNED, defining __CHAR_UNSIGNED__; would the following do
> the right thing in src/include/lib.h (and some other places for code
> written in C instead of C++)?
> 
>   #ifdef __CHAR_UNSIGNED__
>   typedef signed char char
>   #endif

I added #define __CHAR_UNSIGNED__ to config.h and the paragraph above
(with an additional ';') to lib.h.

Compilation fails:
In file included from /usr/src/gnu/dist/groff/src/libs/libgroff/font.cpp:24:
/usr/src/gnu/dist/groff/src/include/lib.h:182: declaration does not declare 
anything

Perhaps it would work with a define, but I think the right solution is
to just use 'signed' when one wants it...

Cheers,
 Thomas