Thanks, Paul!
Hmm ... I see.
How about a way to call "system" without catching stdout, let stdout be
copied
as is to the child process?
The return value could be the status of the command, so let's call it status
$(status command)
Evaluates the command in sub-shell and returns the exit status.
%% mst <[EMAIL PROTECTED]> writes:
m> It turns out that there's no easy way, now, to determine inside the
m> makefile whether the output is to tty or now.
m> the simple $(shell perl -e 'print 1 if (-t STDOUT)')
m> does not work, since anything inside a shell is sending output
m> to she
%% Soren A <[EMAIL PROTECTED]> writes:
>> I don't understand this. If bcopy() isn't there and memmove() is
>> (which wasn't being checked but now will be) then the code in make.h
>> as-is works.
>>
>> Doesn't it?
sa> It really doesn't. [...]
Please note my parenthetical statement
"Paul D. Smith" <[EMAIL PROTECTED]> wrote in
[EMAIL PROTECTED]:">news:[EMAIL PROTECTED]:
> %% Soren A writes:
>
> sa> Yes, and because of that, build on Win32-ish was failing; but
> also sa> you need to have configure check for bcopy(), because
> bcopy() sa> isn't there on Win32- Gnu-ish
%% Soren A <[EMAIL PROTECTED]> writes:
sa> Yes, and because of that, build on Win32-ish was failing; but also
sa> you need to have configure check for bcopy(), because bcopy()
sa> isn't there on Win32- Gnu-ish platforms (I have been told it is
sa> part of GlibC, but Cygwin, etc are based
*I* Soren A <[EMAIL PROTECTED]> wrote in
[EMAIL PROTECTED]:">news:[EMAIL PROTECTED]:
> A part of needs to be rewritten because it appears that at
> some point an ad-hoc (?) addition was made to try to deal with lack of
> bcopy(), but was done not properly, by getting into configure and
> making
"Paul D. Smith <[EMAIL PROTECTED]>" <[EMAIL PROTECTED]> wrote in news:E17oVxZ-
[EMAIL PROTECTED]:
> From: "Paul D. Smith <[EMAIL PROTECTED]>" <[EMAIL PROTECTED]>
> Newsgroups: gmane.comp.gnu.make.devel
> Subject: CVS psmith make: Whoops; configure wasn't looking for memmove.
>
> CVSROOT: /cv
This bug has been fixed in CVS. The fix will be available in the next
release of GNU make.
--
---
Paul D. Smith <[EMAIL PROTECTED]> Find some GNU make tips at:
http://www.gnu.org http://m
got the following internal assertion in make:
/tmp>cat make.bug
var = 1
export var
%:
$(MAKE) -f test.bug
x: var += a
/tmp>cat test.bug
all:
echo $(var)
/tmp>make -f make.bug x
make -f test.bug
make: expand.c:489: allocated_variable_append: Assertion
`current_variable_set_lis
--- Begin Message ---
Hello!
It turns out that there's no easy way, now, to determine inside the makefile
whether the output is to tty or now.
the simple $(shell perl -e 'print 1 if (-t STDOUT)')
does not work, since anything inside a shell is sending output
to shell and not to tty!
I ended up
Hello!
The following test demonstates a problem in Gnu make 3.79.1
It is a combination of recursive make, overriding variables
for targets and += operator.
does someone have fix/work-around for this?
thanks,
MST
/tmp>cat make.bug
var = 1
export var
%:
$(MAKE) -f test.bug
x: var += a
/tm
%% Koen Claessen <[EMAIL PROTECTED]> writes:
kc> I get the not very informative error "dependency `' doesn't
kc> exist" when I am using gmake on the attached makefile. I
kc> could not find anything about it in the manual on gnu.org.
kc> Maybe it is a bug?
Sorry, but you have not provide
Dear gmake developers,
I get the not very informative error "dependency `' doesn't
exist" when I am using gmake on the attached makefile. I
could not find anything about it in the manual on gnu.org.
Maybe it is a bug?
Thanks,
/Koen.
--
Koen Claessen
http://www.cs.chalmers.se/~koen
Chalmers Uni
13 matches
Mail list logo
