Ping! Paul, could you please answer the question I asked in this thread (and Mark repeats a variant of it below)?
TIA > From: "Mark Galeck (CW)" <[email protected]> > Date: Tue, 15 Dec 2009 10:25:08 -0800 > Accept-Language: en-US > acceptlanguage: en-US > > OK, I understand now Eli's reasons of assigning "failure" and "success". > Thank you. > > So now your question to Paul is this, let me rephrase myself: > > SHELL=cmd.exe > > vpath %.s foobar0 foobar1 > > %.o: %.s > echo $< > > foobar0/foobar.s: > > (and as always, foobar.s only in foobar1, not in foobar0). > > > Here, the current behavior of make is > > C:\tmp>make foobar.o > echo foobar0/foobar.s > > How can this possibly be "correct"?? GNU make manual only states that for > the rule with no commands or prerequisites, make "imagines" the target has > been updated, even if it does not exist. People are using this property, to > prevent failure if the file does not exist, for example, for > auto-dependencies in the well-known website > > http://make.paulandlesley.org/autodep.html > > This is not going to work, if make still found foobar0/foobar.s after the > user deleting it. > > Mark > > _______________________________________________ Help-make mailing list [email protected] http://lists.gnu.org/mailman/listinfo/help-make
