Oleksandr Gavenko schrieb am 26.07.2012 um 22:26 (+0300):
> In such situation I use cache technique and real shell scripts, like:
>
> old := 01-xxx 02-yyy 002-zzz
>
> -include Makefile.vars
>
> .PHONY: Makefile.vars
> Makefile.vars:
> { \
> printf 'new :='; \
> for i in $(old); do printf " "$${i#*-}.exe; done; \
> echo; \
> } >$@
But why do you prefix the -include with a minus? As far as I can see,
make will fail the first time around with this cache technique when
there is no Makefile. Would the user not want to see the error message?
Simply because this is a somewhat funky technique.
Michael
_______________________________________________
Help-make mailing list
Help-make@gnu.org
https://lists.gnu.org/mailman/listinfo/help-make
