Re: CVS commit: pkgsrc/wm/afterstep

[Patrick Welche <prlw1%cam.ac.uk@localhost>]

On Thu, Jun 27, 2013 at 01:23:48PM +0200, Joerg Sonnenberger wrote:
> On Thu, Jun 27, 2013 at 11:39:56AM +0100, Patrick Welche wrote:
> > On Wed, Jun 26, 2013 at 09:17:45PM +0000, Joerg Sonnenberger wrote:
> > > Module Name:      pkgsrc
> > > Committed By:     joerg
> > > Date:             Wed Jun 26 21:17:45 UTC 2013
> > > 
> > > Modified Files:
> > >   pkgsrc/wm/afterstep: Makefile distinfo
> > > Added Files:
> > >   pkgsrc/wm/afterstep/patches: patch-libAfterBase_parse.c
> > >       patch-libAfterBase_parse.h patch-src_Animate_Makefile.in
> > > 
> > > Log Message:
> > > Fix backtrace linking on NetBSD. Fix inline use.
> > 
> > I had this in my local tree which is slightly different - does it turn
> > out the same in practice?
> > 
> > # backtrace() and backtrace_symbols were added in NetBSD-6.99.8
> > .if !empty(MACHINE_PLATFORM:MNetBSD-*) && 
> > empty(MACHINE_PLATFORM:MNetBSD-[0-5].*) && 
> > empty(MACHINE_PLATFORM:MNetBSD-6.99.[0-7].*)
> > CONFIGURE_ENV+= ac_cv_func_backtrace_symbols=yes
> > CPPFLAGS+=      -DHAVE_BACKTRACE_SYMBOLS
> > LDFLAGS+=       -lexecinfo
> > .endif
> 
> I prefer to check for features, in this case the presence of the header
> file.

Agreed - I was wondering more about the define / configure test variable...

P