🚀 go-pugleaf

	I also just compiled elm-2.4PL24 on ODT 3.0, using gcc-2.6.0,
	and found just none of the problems Stuart describes.

	It all went very smoothly.

In article <stuart.780456403@coral.cs.jcu.edu.au>,
Stuart Kemp <stuart@coral.cs.jcu.edu.au> wrote:
>Just compiled elm-2.4 PL24 on an Alpha, OSF 3.0, using gcc-2.6.0
>
>Problems:
>
>1) POSIX_SIGNAL is (correctly) defined in the config file, but this
>   causes problems elsewhere, coz there is a conditional
>	#define signal posix_signal
>   in the hdrs/defs.h
>   Later on, when the system's <sys/signal.h> gets included, the result
>   of the above #define is conflicting prototypes for the posix_signal
>   function. I commented it out for this compilation.
>2) Configure did not find TZNAME_USE_TZNAME automatically
>3) Configure did not find `struct utimbuf' (in /usr/include/utime.h)
>
>gcc-related things:
>
>I realise that the elm code is not supposed to be strictly ANSI C, but it
>would be nice if it were! There were loads of warnings from gcc; the full
>output can be obtained via anon ftp to coral.cs.jcu.edu.au in the file
>elm-24PL24-OSF3.0-gcc.warnings.gz (30K).
>
>BUGS:
>- newmbox.c lines 541 and 812: too many arguments for format
>  (The '1' should not be there.)
>
>Most warnings are about:
>
>- Implicit function declarations
>- Implict return-types in function declarations, with then default
>  to "int", which causes warning about "return without value".
>- Unused variables
>- %-specifier does not match argument in the printf-family of functions
>- `/*' within comment
>  (These have been removed from the above anon-ftp file).
>  These warnings are are in config.h with lines like
>	/*#undef        I_SYSUTIME    /**/

-- 
Jean-Pierre Radley   jpr@jpr.com     XC custodian    Sysop, CompuServe SCOForum