[EMAIL PROTECTED] writes:
Tk is using its own build system which doesn't seem to rely on
ExtUtils::MakeMaker for anything but trivial things.
No it isn't. It is using MakeMaker for almost everything, I should know
I wrote it.
It does have a prefix script which does configure-like things,
and does
[EMAIL PROTECTED] writes:
I've been trying but have been unable to get ExtUtils::MakeMaker to
correctly build multiple XS modules in one distribution. I.e. instead
of having something akin to what h2xs would create, a single .pm file
and one .xs file I'm trying to build a distribution with
Demerphq [EMAIL PROTECTED] writes:
Thus, one could rename Cwd.dll - .oldCwd.foo, copy the new
file into place, then setup a pending delete in the registry
with a simple command line util like sysinternals.com's
movefile that queues up pending file operations to take
place upon reboot,
Marcel Holan [EMAIL PROTECTED] writes:
Hi MM-Team, hi Michael,
I'm trying to make a nice CPAN-package called Goo (see http://thegoo.org) and
have serious problems using ExtUtils::MakeMaker (MM).
Now the problems may very well result from my inexperience with MM, as I never
needed anything more
Michael G Schwern [EMAIL PROTECTED] writes:
2) PL files are processed *before* pm_to_blib is run.
Anything you create will be considered for copying into blib. BUT this
means you cannot rely on having the uninstalled module in your path.
I think Tk was assuming that, I know something I wrote
Sisyphus [EMAIL PROTECTED] writes:
Two main ways:
A. Compute something in Makefile.PL before calling WriteMakefile
and adjust args passed to that so that vars end up the way you want.
In this case it means adding something to
LIBS =
passed in. (This is what Tk does - it is
Sisyphus [EMAIL PROTECTED] writes:
Hi,
I'm wanting to add libraries to 'EXTRALIBS' and 'LDLOADLIBS' in a
generated makefile. Can this be done from the Makefile.PL that generates
that makefile ?
Yes.
if so, how? :-)
Two main ways:
A. Compute something in Makefile.PL before calling
Michael G Schwern [EMAIL PROTECTED] writes:
In part one I layed out the current state of things and what's broken. In
part two here I'll lay out the new design in the hopes it gets vetted for
sanity.
Now, the obvious solution would be just make the directories targets.
INST_LIB = blib/lib
Bill Moseley [EMAIL PROTECTED] writes:
On Thu, Nov 18, 2004 at 03:15:35PM -0500, Michael G Schwern wrote:
On Thu, Nov 18, 2004 at 10:00:59AM -0800, Bill Moseley wrote:
$VERSION = '6.17'
When I build perl with make perl it seems that the library path is
not used in the binary. That is
Ton Hospel [EMAIL PROTECTED] writes:
Hi,
I recently wanted to have several XS files in the same directory, and
was following the method described in the ExtUtils::MakeMaker FAQ. That
worked fine for two XS files, but from the third on, that started dumping
core. As far as I could determine this
Rafael Garcia-Suarez [EMAIL PROTECTED] writes:
Alexey Tourbin wrote:
Hello,
I would like to propose the following two changes for fixin() function:
- make use of File::Temp in order to make .new and .bak
filenames less predicatble
Why are less predictable names an advantage?
Steve Hay [EMAIL PROTECTED] writes:
Hi Steve, we meet again ;-)
Those message obviously come from the test target in the sub-directory
Makefile's. I would like to suppress them because they look slightly
alarming: It looks as if I haven't written tests for the Foo and Bar
modules, but
Gerrit P. Haase [EMAIL PROTECTED] writes:
What would the fix be? There's this code in ExtUtils::Liblist::Kid which is,
I presume, the problem:
if ($VC and exists $ENV{LIB} and $ENV{LIB}) {
push @libpath, split /;/, $ENV{LIB};
}
which seems to pretty explicitly want to pull
Michael G Schwern [EMAIL PROTECTED] writes:
On Sat, Mar 27, 2004 at 10:25:24PM +0100, Gerrit P. Haase wrote:
LIB is picked up by MakeMaker if it is available in the environment, it causes
failures everytime when LIB contains spaces as it is usually the case in a
default Visual Studio setup.
=?unknown-8bit?B?RGFuaetsIFcu?= Crompton [EMAIL PROTECTED] writes:
Hi,
due to library path ambiguity within Cygwin I've found that
the Makefile PASTHU variable will cause the make to fail with:
cd pTk make -f Makefile all LIB=C:\Program Files\Microsoft SDK\Lib\.;c:\Program
Michael G Schwern [EMAIL PROTECTED] writes:
On Thu, Feb 19, 2004 at 08:35:28AM +, Nick Ing-Simmons wrote:
Michael G Schwern [EMAIL PROTECTED] writes:
One thing to keep in mind is portability. In order for this to be useful
it has to run on pretty much all platforms. Unix, Windows, VMS
Michael G Schwern [EMAIL PROTECTED] writes:
One thing to keep in mind is portability. In order for this to be useful
it has to run on pretty much all platforms. Unix, Windows, VMS, etc...
So I'm trying to keep it as simple as possible.
On Wed, Feb 18, 2004 at 05:29:49PM +, Adrian Howard
Christian Goetze [EMAIL PROTECTED] writes:
On Tue, 13 Jan 2004, Christian Goetze wrote:
I need to do a custom perl build (mainly to support 64bit int), and
make a binary, _relocatable_ RPM out of it, including a variety of CPAN
modules,
You need to do exactly that? It would normally make
Michael G Schwern [EMAIL PROTECTED] writes:
On Thu, Nov 27, 2003 at 03:58:13PM +, Nick Ing-Simmons wrote:
Does Module::Build do nested modules yet?
Dunno. Why is this relevant?
Because Tk:: and Audio:: (my two major XS modules on CPAN) are
both structured that way.
When Module::Build has
Michael G Schwern [EMAIL PROTECTED] writes:
On Wed, Nov 26, 2003 at 07:40:57PM +, Salvador Fandio wrote:
A new version of the C++ support for MakeMaker patch is ready.
The more I think about this the more nervous I get at putting all this new
functionality into MakeMaker in a part that's not
David Dyck [EMAIL PROTECTED] writes:
I didn't see this make it to the ptk list, but I've learned a little more.
I tested Tk-804.025_beta6 with perl 5.8.2 on the same computer and
make test succeded, but it still fails 95% of the tests with bleading
edge perl 5.9.0. I've added the p5p list to the
Rafael Garcia-Suarez [EMAIL PROTECTED] writes:
Michael G Schwern wrote:
The major reason I've been avoiding putting any XS tests into MakeMaker
is being able to reliably determine if the user has a working compiler
and what that compiler is. Perl has it easy, the user obviously has a
C
Salvador Fandio [EMAIL PROTECTED] writes:
Hi,
Now that 5.8.2 has been released, Michael G Schwern has accepted to
consider my patch to add support for C++ modules for inclusion on
MakeMaker but he has requested me to get it reviewed by somebody on
here first so...
A new version of the patch is
Mike Castle [EMAIL PROTECTED] writes:
Tk-804.025_beta5 + ExtUtils-MakeMaker-6.19 = failure
Bother them! - will get a new MM and see what I can do.
This presumably means I have to do another Tk800, and Audio release as well.
From the Changes file:
- Removed now unused dir_target() method.
Michael G Schwern [EMAIL PROTECTED] writes:
On Sun, Nov 09, 2003 at 12:55:21PM +, Nick Ing-Simmons wrote:
Mike Castle [EMAIL PROTECTED] writes:
Tk-804.025_beta5 + ExtUtils-MakeMaker-6.19 = failure
Bother them! - will get a new MM and see what I can do.
This presumably means I have to do
Michael G Schwern [EMAIL PROTECTED] writes:
Here's what I've done.
All the .exists targets are gone.
dir_target() which generated them is gone.
The die if we can't find perl.h check is gone.
I've replaced it all with this:
config :: $(FIRST_MAKEFILE) blibdirs
$(NOECHO) $(NOOP)
blibdirs
Andreas J Koenig [EMAIL PROTECTED] writes:
make does not provide a way to ensure the existence of a directory.
Period. The ritual above achieves that the directory $(INST_AUTODIR)
exists. We wanted to avoid using 'mkdir -p', so we had to create one
directory at a time.
Is it worth adding a mkdir
Michael G Schwern [EMAIL PROTECTED] writes:
On Wed, Aug 13, 2003 at 01:29:52PM +0100, [EMAIL PROTECTED] wrote:
Steve has added some more demos to Tk.
It now fails to build on Win32 as there is a rather long command line.
Recent versions of MakeMaker (ie. 5.8.0 and up) should be smart enough to
Ed Avis [EMAIL PROTECTED] writes:
You could look at what DJGPP does - I think it uses the @ convention,
ie you run
% foo @args
and the file 'args' is read by program foo and taken as arguments. I
don't know how this interacts with make, exactly.
I guess (wildly) that DJGPP's C runtime treats
.
But as POD is paragraph parsed the patch looked sane just the same
- the long line seemed to be part of the text rather than part of the
item tag.
--
Nick Ing-Simmons
http://www.ni-s.u-net.com/
on what options perl was built with, and is
there any way to find out what MakeMaker is thinking?
A user asked me what make program to run on Windows and I realized I
didn't have any idea of the answer, except a vague feeling that it
used to be nmake.
--
Nick Ing-Simmons
http://www.ni-s.u
'command'
--
Nick Ing-Simmons
http://www.ni-s.u-net.com/
.;'
and we just commented out the 2nd line ;-)
--
Nick Ing-Simmons
http://www.ni-s.u-net.com/
with metal tyres.
These new-fangled alloy things with the squidgy rubber stuff just
can't take the flint strewn rutted tracks of the back-woods...
--
Nick Ing-Simmons
http://www.ni-s.u-net.com/
MakeMaker also throws out those arguments as well.
I have resorted to override in Makefile.PL that puts the -Wl,-Bstatic
back in.
Regards,
Jochen
--
Nick Ing-Simmons
http://www.ni-s.u-net.com/
Jochen Wiedmann [EMAIL PROTECTED] writes:
Hi, Nick,
Zitiere Nick Ing-Simmons [EMAIL PROTECTED]:
As far as I know MakeMaker is right to reject -llibbar.a as most
linkers
would as well. e.g. GNU ld you specify this by adding -Bstatic before
-lbar. (And you normally get the -Bstatic
::Data module is more traditional XS
--
Nick Ing-Simmons
http://www.ni-s.u-net.com/
to C_FILES move things from archlib to lib etc.
remove .../*~ files from PM_TO_BLIB etc.)
--
Nick Ing-Simmons
http://www.ni-s.u-net.com/
-args.
But I guess that code should be in good shape by now, and if not
is maintained by people that should know the issues.
--
Nick Ing-Simmons
http://www.ni-s.u-net.com/
invokations
in the Makefiles.
--
Nick Ing-Simmons
http://www.ni-s.u-net.com/
to ask again, hoping that you won't mind this slightly
obsessive style of discussion, whether packlists are actually used for
anything and whether they are needed. And if they *could* one day be
used, are they really necessary for that proposed use?
--
Nick Ing-Simmons
http://www.ni-s.u-net.com/
://modperlbook.org http://apache.org http://ticketmaster.com
--
Nick Ing-Simmons
http://www.ni-s.u-net.com/
--- http://perl.apache.org
mailto:[EMAIL PROTECTED] http://use.perl.org http://apacheweek.com
http://modperlbook.org http://apache.org http://ticketmaster.com
--
Nick Ing-Simmons
http://www.ni-s.u-net.com/
(__TEXT,__text)
/usr/lib/libSystem.dylib(regerror.o) unused definition of _regerror
make[1]: *** [target_static] Error 1
make: *** [apache_httpd] Error 2
==
-Ken
--
Nick Ing-Simmons
http://www.ni-s.u-net.com/
from list by default
far fewer overrides would been needed.
--
Nick Ing-Simmons
http://www.ni-s.u-net.com/
is itself auto generated!) does indeed have
relative paths. But as they end in ...$ they would match just as well
against fill paths.
-Ken
--
Nick Ing-Simmons
http://www.ni-s.u-net.com/
/documents/top10.html
http://www.prospect.org/print/V11/16/ackerman-b.html
http://www.theatlantic.com/issues/2000/12/fallows.htm
--
Nick Ing-Simmons
http://www.ni-s.u-net.com/
47 matches
Mail list logo
