Consider the following Makefile:
# start
.PHONY: a b c d all
BLAH := foo
COMMAND = echo $(BLAH)
all: a b c
a b c:
@$(COMMAND)
b: BLAH := bar
c: BLAH := bar
c: COMMAND += snafu $(BLAH)
d: c
d: BLAH := baz
# end
With make-3.79.1, running 'make' yields:
foo
bar
foo snafu bar
This seems to indicate a bug in target specific variable management, as
I would expect the third line to be 'bar snafu bar' because COMMAND is a
recursively expanded variable. Even more interestingly, a 'make d'
yields:
baz snafu bar
which also is not what I expect (which is 'baz snafu baz'), but seems to
be a bit less "incorrect" than the output of target c.
So, is this a real bug, or am I missing something in the docs about target
specific variable limitations? I have a real makefile that doesn't work
because of this, so knowing whether this is a bug or an error I need to
correct would be useful.
Thanks,
Scott
--
Scott Murray
SOMA Networks, Inc.
Toronto, Ontario
e-mail: [EMAIL PROTECTED]
_______________________________________________
Bug-make mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/bug-make
