"Robert P. J. Day" wrote:
>
> On Tue, 16 Dec 2003, Noel Yap wrote:
>
> > This sounds about right for a recursive make structure.
> >
> > Alternatively, you can go for a non-recursive make (I call it a
> > recursive makefile) structure. See
> > http://aegis.sourceforge.net/auug97.pdf.
>
> as i t
On Tue, 16 Dec 2003, Noel Yap wrote:
> This sounds about right for a recursive make structure.
>
> Alternatively, you can go for a non-recursive make (I call it a
> recursive makefile) structure. See
> http://aegis.sourceforge.net/auug97.pdf.
as i take a closer look at this, just out of curiosi
This sounds about right for a recursive make structure.
Alternatively, you can go for a non-recursive make (I call it a recursive makefile)
structure. See http://aegis.sourceforge.net/auug97.pdf.
In a recursive makefile structure, rather than running make for the subdirectories,
each subdirect
CC'ing [EMAIL PROTECTED]
Evgeny Sologubov wrote:
>
> Hello, Noel,
> Tuesday, December 16, 2003, 5:02:03 PM, You wrote:
> >>[...]
> NY> For example, in the end, I think what you want is to generate
> NY> a dependency file for a project. The usual way to do this is to
> NY> include that dependency
(sorry about all the traffic my previous posting caused -- i didn't realize i
was disturbing the hornet's nest quite that much. i did eventually figure out
that i was looking at the wrong section in the make manual. and now,
onward.)
i want to define a standard set of targets for a multi-le
Hello, Noel,
Tuesday, December 16, 2003, 5:02:03 PM, You wrote:
>>[...]
NY> For example, in the end, I think what you want is to generate
NY> a dependency file for a project. The usual way to do this is to
NY> include that dependency file within the makefile. This inclusion
NY> cannot be done gen
"Paul D. Smith" wrote:
> ny> Anyway, the behaviour I had seen was when I had somithing like the
> ny> following rules:
>
> ny> .SUFFIXES: .pm _wrap.cc
> ny> $(build.DIR)/common/%.pm: $(build.DIR)/common/%_wrap.cc
>
> First note that all your uses of .SUFFIXES here are useless: .SUFFIX
%% Noel Yap <[EMAIL PROTECTED]> writes:
ny> Upon re-reading this portion of the paper, I think I may have
ny> misunderstood exactly what the bug is:
ny> Quick note: why do we have ":" as the command to be run? That's
ny> the Bourne shell's "do nothing" operator. GNU make is actually
ny>
Upon re-reading this portion of the paper, I think I may have misunderstood exactly
what the bug is:
Quick note: why do we have ":" as the command to be run? That's the Bourne shell's "do
nothing" operator. GNU make is actually smart enough to notice that your rule consists
of just the "do noth
%% Noel Yap <[EMAIL PROTECTED]> writes:
ny> Here's what I found in the manual:
ny> We feel that it is cleaner for a rule without commands to always
ny> simply add to the prerequisite list for the target.
ny> IIRC, there's a bug in which this doesn't always happen (I think
ny>
Evgeny Sologubov wrote:
>
> Hello,
>
> I am having a problem with such a rule:
>
> all: prj-Main
>
> prj-%: $(shell cat -s %/Makefile.dep)
>$(MAKE) -C $*
>
> make (a win32 build of v3.80) says:
> /usr/bin/cat: %/Makefile.dep: No such file or directory
> :(
>
> I expected that it shall
"Paul D. Smith" wrote:
> If you read the manual section on "Rules without commands or
> prerequisites" the pros and cons are discussed there.
Here's what I found in the manual:
We feel that it is cleaner for a rule without commands to always
simply add to the prerequisite list for the t
Hello,
I am having a problem with such a rule:
all: prj-Main
prj-%: $(shell cat -s %/Makefile.dep)
$(MAKE) -C $*
make (a win32 build of v3.80) says:
/usr/bin/cat: %/Makefile.dep: No such file or directory
:(
I expected that it shall be expanded to the following:
prj-Main: $(shell cat -s
13 matches
Mail list logo
