Re: Makefile Magic. And now also - Autoconf.

[Christian Reiniger <warewolf@mayn.de>]

s369625@student.uq.edu.au wrote:

>>COBJS += $(patsubst %.c,%.o,$(filter %.c,$(SRCS)))
>>CXXOBJS += $(patsubst %.cc,%.o,$(filter %.cc,$(SRCS)))
>>
>>AFAIK this will work nicely as long as we don't rely on implicit rules for
>>building
>Then we add 
>OBJS=$(COBJS) $(CXXOBJS)
>Dynamic, recursive variable expansion GOOD!

Yes, right. That fixes the problem. neat :)

>>>>	rm -f *.o *.a $(TARGETS)
>>>># *.so / *.dll etc (depending on the platform) is also important
>>Hmm, as the error value of rm should be ignored anyway it's propably the
>>best to simply specify all possible suffixes.
>Maybe, it all depends on how much effor we want to put in.  

Perhaps I'm missing something, but I don't think writing
-rm -f *.o *.a *.so *.dll $(TARGETS)
is very tedious ;)

>>>see with them is that they are gnu make dependent.  Perhaps we should switch
>>>to Automake (I certainly don't want to do a sophisticated non-gmake system by

>>But I assume you're right. Writing the makefiles per hand is even worse...

>is OK.  The real issue is whether we want to be dependent on gnu make.
>On non Unix systems it is a moot, and Unix systems normally have gmake,
>but should we rely on that?

Automake generates configure-like scripts, right? So can we rely on having
"sh" on Windows systems?

I think it doesnt matter whether we rely on gmake or sh - it's anyway only
the developers who want to compile PPlay themselves. Joe and Jane user just
get the binary distribution...


Cu
	Christian
--

Shut Windows and Close Gates...