Hi Paul :)
* Paul D. Smith <[EMAIL PROTECTED]> dixit:
> This is a long-standing deficiency in the parsing of MAKEFLAGS.
> It's something we hope to fix.
OK, then, thanks a lot for your answer :)
Just one more point: is there any way to force the -R or -r flags
then? Not that it is
Hi all :)
In the GNU make info file, section 'Options/Recursion', you
mention that you can use the MAKEFLAGS variable to set additional
options to be in effect in that Makefile. Ok, I need to run a
Makefile with '-R' in effect, so I set MAKEFLAGS to 'R' (I've tested
with '-R' too), but in
Hi Paul :)
> Why doesn't it
> work? It doesn't work because GNU make actually caches the contents of
> directories as it reads them, for performance reasons.
I supposed that some caching was the cause, but for me was
reasonable that $wildcard will not expand if the new file wasn't
cached
Hi Johan :)
I'm writing to the bug mailinglist so anybody can read the
message, ok?, but really is for you ;)
> this seems to be by design, to overcome the fact that
>
> OBJECTS = *.o
>
> doesn't expand the list, whereas
>
> OBJECTS = $(wildcard *.o)
>
> does.
Yes, I know, it
Hi all :))
Don't know if this is a bug, but anyway this doesn't work and it
should (IMHO). The problem is the function $(wildcard) not working
properly when the argument is a generated file. Let me explain it
with an example, better. We have this tiny Makefile:
all:
@touch testfil
