Tue 2003-01-28 17:00, Bob Jones (1:343/41) wrote to Andrew Clarke:

 > Well, I have the startup.mk file issue resolved to the point where 
 > DMake doesn't complain about not having it...  Or having errors in 
 > it....  It took some editing over the default versions that came in the 
 > Max source distribution.  Unfortunately, when I get down to making an 
 > actual target in Max's main make file, I don't get anything other than 
 > brief running of dmake and end of program with no messages (error or 
 > otherwise).  It does complain if I don't make a proper target....  So, 
 > I believe I'm not setting some environmental parameters Scott used to 
 > control the processing.  I am strongly leaning towards simplyfing the 
 > make setup he used.  I can understand how things evolve over time, but 
 > the OS/2 command scripts weren't his last development environment.....  

Write your own makefiles from information extrapolated from the ones
supplied.  You can still use dmake if you want to.  :-)

 > Since open watcom claims to plan to port the code to Linux, *if* that 
 > happens, it might easy getting Max to compile there.... 

I don't know about that.  It's one thing for them to port to compiler,
another thing entirely to port their C libraries.  I don't really see them
doing that in a hurry, particularly since the main reason why anyone would
want to use Watcom C in Linux is for their optimiser (whereas the Watcom C
libraries weren't terribly efficient), although I hear GNU C 3.x has better
optimisation than previous version, so that might negate people using
Watcom altogether.  Borland's Linux C++ compiler (part of Kylix 3) just
uses the standard #include files from /usr/include/, and links with glibc.

-- mail{at}ozzmosis.com

--- Msged/NT 6.1.1