> -----Original Message-----
> From: nobu.nokada / softhome.net [mailto:nobu.nokada / softhome.net] 
> Sent: Tuesday, January 18, 2005 6:44 AM
> To: ruby-core / ruby-lang.org
> Subject: Re: What do when when T_DATA is redefined?
> 
> 
> Hi,
> 
> At Tue, 18 Jan 2005 02:27:39 +0900,
> Berger, Daniel wrote in [ruby-core:04261]:
> > I'm having a T_DATA conflict with a Solaris extension that 
> I'm trying 
> > to write.  The problem stems from the nfs/nfs.h, which is in turn 
> > doing an include on tiuser.h which in turn does an include on 
> > sys/tiuser.h.  It's this last file, sys/tiuser.h, that does 
> a "#define 
> > T_DATA 0x004" and is causing the problem.
> > 
> > I have no choice but to include nfs/nfs.h, otherwise I can't see 
> > struct mntinfo, which I need for this particular extension that I'm 
> > writing. Moving the '#include "ruby.h"' to the bottom of the 
> > #include's is not an ideal solution either, since that means I'm 
> > redefining a value that nfs.h may be using.
> 
> Can't you divide the source to separated two files, and 
> perhaps a header?
> 
> -- 
> Nobu Nakada

I'm afraid I don't understand.  Can you be more specific?

Dan