> I have a number of files that need variable substitution, so I do this in > a Makefile.am:
> foo: foo.in > @rm -f foo > @sed \ > -e 's,@@perlmoduledir@@,$(libexecdir)/perl,' \ > -e 's,@@swishbindir@@,$(bindir),' \ > -e 's,@@perlbinary@@,$(PERL),' \ > $(srcdir)/foo.in > foo > But I'm doing that in more and more Makefile.am files in my project > tree -- and sometimes I need to do it to a file where there isn't a > Makefile in that directory. > I'm feeling like I'm using the same bit of sed code in too many places. > Is there a better way to manage this? I'd like to have the sed code > (above) defined in one place and then maybe list all the files that need > to be operated on. Use "include"?
