Re: Patch for autoconf/145

[Akim Demaille]

Pavel Roskin <address@hidden> writes:

> > I have one desire though: avoid any kind of magic.  Why do we have to
> > do all this, I have still not understood.  If the user wants to shoot
> > in its foot, then Autoconf should not prevent it.
> 
> Good point, but it doesn't invalidate the patch.

I know.

> > Now, Automake issues that VPATH.  Why?  Why don't we have the same
> > approach as for any other variable: AC_SUBST?  Yep, OK, there is the
> > problem of *completely* getting rid of VPATH.  Then how about the
> > approach used with SET_MAKE?  Or with AMDEP?
> >
> > @address@hidden = @VPATH
> >
> > or
> >
> > @SET_VPATH@
> 
> That's a good idea. However, some old makefiles (or makefiles generated by
> Automake we introduce it) rely on Autoconf removing VPATH.

That's only a problem of having them Sync.

> I believe that Automake can switch to @SET_VPATH@, but I'm in doubt what
> package should AC_DEFINE it - Autoconf or Automake or both?

Who cares.  AC_SUBSTing twice is fine.

The only question is *what* is supposed to be in there.  Sorry for
being so naive, but I really never paid attention to VPATH.

~/src/ace % grep VPATH **/Makefile                               nostromo 20:20
autoconf-2.49d/=build/doc/Makefile:VPATH = ../../doc
autoconf-2.49d/=build/m4/Makefile:VPATH = ../../m4
autoconf-2.49d/=build/Makefile:VPATH = ..
autoconf-2.49d/=build/Makefile:# This target untars the dist file and tries a 
VPATH configuration.  Then
autoconf-2.49d/=build/man/Makefile:VPATH = ../../man
autoconf-2.49d/=build/tests/Makefile:VPATH = ../../tests
autoconf-2.49d/=build/tests/Makefile:# thanks God for VPATH.  Hm...
Makefile:# This target untars the dist file and tries a VPATH configuration.  
Then
tests/Makefile:# thanks God for VPATH.  Hm...


Hm, I see.  Do we want something any different from simply $(srcdir)?


> I don't like that magic either. I believe there are historic reasons -

I believe this too, and you know my opinion about maintaining history
alive ;)