There is another link that might be interesting here:
http://interix-wgcc.sourceforge.net/
Not to mention that Microsoft's "Services For Unix"
(aka "Interix", or "Subsystem for Unix Applications"
in Vista) provides its own wrapper around Bill's
compiler.
Regards
Harri
Hi Benoit,
Benoit SIGOURE wrote:
[1] http://www.tsunanet.net/~tsuna/cccl
[2] http://www.tsunanet.net/~tsuna/make-3.81-cygwin.patch
http://www.tsunanet.net/~tsuna/make-3.81-cygwin_MAKE_expansion.patch
Can't be bothered to find the related posts on GNU Make W32 ML
archive, sorry.
It
Hi Ralf,
Ralf Wildenhues wrote:
Hello Harald,
* Harald Dunkel wrote on Mon, Apr 30, 2007 at 01:04:25PM CEST:
I have to install several manpages in separate
subdirectories. But it seems that something like
toolmandir = ${prefix}/tooldir/man
toolman_MANS = hello.1
does not
Hi folks,
I have to install several manpages in separate
subdirectories. But it seems that something like
toolmandir = ${prefix}/tooldir/man
toolman_MANS = hello.1
does not work. Did I miss something in the documentation
here? (I am using Automake 1.9.6.)
Regards
Harri
Hi folks,
Does anybody know a patch for depcomp to support the
-showIncludes flag of Microsoft's recent compiler?
Regards
Harri
?
Regards
Harri
===
Stepan Kasal wrote:
> Hello,
>
> On Fri, Oct 21, 2005 at 10:58:44PM +0200, Harald Dunkel wrote:
>> Stepan Kasal wrote:
>>> `IFS=:; x=':$(VPATH)'; \
>>> for d in $$x
Hi folks,
As suggested earlier on this list I am using a wrapper
around cl.exe to make it look like a Unix compiler.
Works pretty well, except that the *.Po files are
empty after the build. Depcomp is called with depmode=none
:-(.
Probably Cygwin's makedepend wouldn't work for cl.exe,
but looking
Hi Ralf,
Ralf Wildenhues wrote:
> Hi Harald,
>
> * Harald Dunkel wrote on Wed, Apr 26, 2006 at 10:44:55AM CEST:
>> it seems that DEFAULT_INCLUDES is set deeply within
>> automake.pl to
>>
>> DEFAULT_INCLUDES = -I. -I$(srcdir)
>
> Only if not using th
Hi folks,
it seems that DEFAULT_INCLUDES is set deeply within
automake.pl to
DEFAULT_INCLUDES = -I. -I$(srcdir)
What is the reasoning here? Doesn't this undermine the
difference between
#include "myheader.h"
and
#include
?
Regards
Harri
Hi folks,
Ralf Wildenhues wrote:
> Hi Daniel,
>
> * Daniel Kraft wrote on Tue, Dec 06, 2005 at 05:12:38PM CET:
>
>>Is this something possible with automake? Or do I need to write special
>>makefile-constructs for that (e.g., a small library of every directory's
>>sources) my own?
>
>
> Yeah, y
Hi Peter,
Peter Ekberg wrote:
>
> Here's my current diff from libtool cvs head. I have been buried
> in work lately so I have not tested it in a while. Use at your
> own risk and all that...
>
Many thanx.
Harri
Hi Peter,
Ralf Wildenhues wrote:
>
> You may search the archives of the libtool-patches mailing list for the
> dozens of mails Peter Ekberg has written to support all of this nicely,
> including static and shared libraries, the libtool part of "-c -o"
> support, C++ source file names, flags for l
Hi Ralf,
Ralf Wildenhues wrote:
>
> Erm. Before continuing on this road any further: have you tried using
> cccl? It's a wrapper script for MSVC designed to do exactly what looks
> like quite a bit of work inside the autotools. I'm not saying it can't
> be done, or should not be done, but it h
PS: Sorry, I pasted the wrong Makefile.am into the EMail, see below.
Harald Dunkel wrote:
> Hi Ralf,
>
> Ralf Wildenhues wrote:
>
>>Harald, please consider *providing more and useful information* if you
>>want help. I for one can't read your mind, and almost every
Nor does it set CXX accordingly.
in the Makefile.
The patch I had sent was an attempt to workaround the problem,
but it is not sufficient. Any help would be highly appreciated.
Regards
Harri
--- Begin Message ---
Hello,
On Fri, Jul 01, 2005 at 10:14:28AM +0200, Harald Dunkel wrote:
> It se
Ralf Wildenhues wrote:
> Hi Harald,
>
>
> Please consider this thread:
> http://lists.gnu.org/archive/html/libtool/2005-10/msg00069.html
>
How is this thread related to copying a missing script? Please
note that C support is broken, too, as soon as there is some
C++ code in your Makefile.am.
Hi folks,
Harald Dunkel wrote:
> Hi folks,
>
> What is the criteria for copying the compile script into
> the source directory tree? I have some *.cc code, it is
> mentioned in my Makefile.am file, configure detects that
> the compile script must be used, too, but Automake do
Hi Ralf,
Ralf Corsepius wrote:
> On Tue, 2005-11-08 at 08:15 +0100, Harald Dunkel wrote:
>
>>Hi folks,
>>
>>I would like to build 32bit and 64bit libraries within the
>>same Makefile.am. In the install directory tree the libs
>>should get the same name, but t
Hi folks,
I would like to build 32bit and 64bit libraries within the
same Makefile.am. In the install directory tree the libs
should get the same name, but the 64bit library is supposed
to be installed in ${exec_prefix}/lib64, of course.
Obviously automake can't distinguish libraries in the
buil
Hi folks,
What is the criteria for copying the compile script into
the source directory tree? I have some *.cc code, it is
mentioned in my Makefile.am file, configure detects that
the compile script must be used, too, but Automake doesn't
provide it.
???
Any help would be highly appreciated.
H
Hi folks,
Stepan Kasal wrote:
>
> I committed the patch attached to this mail.
>
Could you please apply the attached patch as well? It is
necessary on Windows to detect '\' as a directory path
separator.
Many thanx
Harri
--- compile~ 2005-10-24 12:03:48.0 +0200
+++ compile 2005-10-26
Stepan Kasal wrote:
>
> I looked at /usr/share/automake-1.9/compile and I cannot understand
> how this could happen.
> I think that the parametr to compile should look like
> some/path/main.c
> which becomes cfile, and then cofile is assigned as
>
> cofile=`echo "$cfile" | sed -e 's|^.*/||'
Hi,
Stepan Kasal wrote:
> Hello,
>
>
> Suggested code for today: ;-)
>
> `IFS=:; v=$(VPATH); \
> for d in '' $$v; do \
> test -n "$$d" && d=$$d/; \
> test -f $${d}%SOURCE% && echo $$d && break; \
> done`%SOURCE%
>
Works for me, as it seems.
Reg
Hi Stepan,
Stepan Kasal wrote:
>
> I looked at /usr/share/automake-1.9/compile and I cannot understand
> how this could happen.
> I think that the parametr to compile should look like
> some/path/main.c
> which becomes cfile, and then cofile is assigned as
>
> cofile=`echo "$cfile" | sed -
Stepan Kasal wrote:
>
> Your proposal looks good, except I think it might be better to use single
> quotes around $(VPATH).
> And the quoting around $d, which I introduced, is redundant.
>
> That yields:
>
> `IFS=:; x=':$(VPATH)'; \
> for d in $$x; do \
> test -n $$d &&
Hi Stepan,
Stepan Kasal wrote:
>
> So I suppose something like this
>
> `IFS=:; for d in '' '$(srcdir)' $(VPATH); do \
> test -n "$$d" && d=$$d/; \
> test -f "$$d"%SOURCE% && echo "$$d" && break; \
>done`%SOURCE%
>
> Could you test whether this works fo
Hi folks,
Ralf Wildenhues wrote:
>
> What exactly does not work for you?
I tried to use
VPATH = $(srcdir):$(top_srcdir)/some/common/dir
in Makefile.am.
> In what circumstances?
I am not sure what you mean by "circumstances". I am trying to
build a C++ library (no libtool) using bot
Hi folks,
Would it be possible to rely upon $(VPATH) in depend2.am?
Instead of
`test -f '%SOURCE%' || echo '$(srcdir)/'`%SOURCE%
something like this could work, too (not verified yet):
`test -f '%SOURCE%' || (VPATH=$(VPATH); test "$$VPATH" ||
VPATH=$(srcdir); for i in \`IFS=:;
Hi folks,
AFAICS libraries are installed using ${INSTALL_DATA}.
Wouldn't it be more consequent to use ${INSTALL_PROGRAMS}
here?
Both libraries and programs is something that could be
stripped, i.e. you could use 'install -s'. But defining
INSTALL_DATA to use 'install -s' doesn't work, of course.
Simon Richter wrote:
> Hi,
>
> Harald Dunkel wrote:
>
>> I would like to create the directories for some
>> $SUBDIRS as needed at runtime, e.g.
>
>
> erm, why?
>
>> Unfortunately Automake complains about the missing
>> directories in the source
Hi folks,
I would like to create the directories for some
$SUBDIRS as needed at runtime, e.g.
:
SUBDIRS = mydir
mydir/Makefile: FORCE
${INSTALL} -d mydir
generate $@
:
Unfortunately Automake complains about the missing
directories
Sander Niemeijer wrote:
>
> On woensdag, sep 28, 2005, at 17:04 Europe/Amsterdam, Harald Dunkel wrote:
>
>>> autoconf sets CFLAGS/CXXFLAGS to "reasonable defaults", that's all. If
>>> these defaults cause problems on your platforms, you have to override
Ralf Corsepius wrote:
> On Tue, 2005-09-27 at 12:53 +0200, Harald Dunkel wrote:
>
>>Ralf Corsepius wrote:
>>
>>>Nope. You don't seem to have understood how things are working:
>>>
>>>AM_CFLAGS/AM_CXXFLAGS are supposed to take flags
Ralf Corsepius wrote:
>
> Nope. You don't seem to have understood how things are working:
>
> AM_CFLAGS/AM_CXXFLAGS are supposed to take flags having been specified
> by a package's developer.
>
Understood.
> CFLAGS/CXXFLAGS are suppoed to take flags having been specified by a
> package's inst
Brian wrote:
> I have a need to force three files to not be optimized. I've followed the
> instructions in the manual for setting them up in their own library, and
> then using LIBADD to combine it with the original library.
>
> If I use AM_CXXFLAGS, the -O0 is superceded by a -O2. The same occurs
Stepan Kasal wrote:
>
> ad 1) The "install-am -> all-am" dependency causes everuthing to be checked.
>
> ad 2) I see the fillowing dependencies in a Makefile.in:
>
> install-exec-am: install-binPROGRAMS install-libLTLIBRARIES
> install-binPROGRAMS: $(bin_PROGRAMS)
> install-libLTLIBRARIES: $(lib
Hi folks,
It seems that the *_DATA objects are verified twice
during make install. In the Makefile you find
install-am: all-am
@$(MAKE) $(AM_MAKEFLAGS) install-exec-am install-data-am
*_DATA objects are verified on 2 paths:
1) install-am -> all-am -> $(DATA) -> *_DATA
2) install-am -> i
Hi Ralf,
Ralf Wildenhues wrote:
>
> How about using either
> - a convenience archive, or
> - one single large Makefile.am with subdir-objects
> (see "info Automake Alternative")
>
A convenience library doesn't work, because I have pretty many
different CPP defines, probably set either in conf
Bob Proulx wrote:
> Harald Dunkel wrote:
>
>>Question about make depend:
>>
>>If I set
>>
>> SRCDIR = ../src
>> noinst_PROGRAMS = hello
>> hello_SOURCES = ${SRCDIR}/hello.c
>
>
> Shouldn't you be using normal VPATH? T
Hi folks,
Question about make depend:
If I set
SRCDIR = ../src
noinst_PROGRAMS = hello
hello_SOURCES = ${SRCDIR}/hello.c
in my Makefile.am, then make complains
Makefile:242: ../src/.deps/hello.Po: No such file or directory
make[1]: *** No rule to make target `../src/.de
Hi folks,
Please see subject. Of course I would agree that this
dependency is usually a good thing, but sometimes it might
be helpfull to do a 'make install' for another prefix e.g.
in your stow directory without verifying all the dependencies
again.
It would be easy to split the old install targ
Hi folks,
Would it be possible to strip libraries at install
time? Or do I have to set CFLAGS/CXXFLAGS?
Regards
Harri
Ralf Wildenhues wrote:
>>
>>This would assume that I am allowed to change the foreign
>>Makefile.am files,
>
>
> Yes. What prevents this?
>
The version control system, maybe? Code coming
from another business units that I have to use
as is? Or simply because it is read-only?
>
>>and it assu
Hi Ralf,
Ralf Wildenhues wrote:
> Hi Harald,
>
> * Harald Dunkel wrote on Tue, Aug 16, 2005 at 12:39:46PM CEST:
>
>>I would need some way to add recursive targets to all generated
>>Makefiles, without asking every owner of every Makefile.am
>>to append some includ
Hi folks,
I would need some way to add recursive targets to all generated
Makefiles, without asking every owner of every Makefile.am
to append some include statement or some keyword to be replaced
by AC_SUBST_FILE.
What would be your suggestion to solve this?
Regards
Harri
Hi folks,
Would it be possible to tell Automake to print some
warning message for this bad Makefile.am:
bin_PROGRAMS += PreciousUtils_checker
PreciousUtils_checker_SOURCES =
precious/Utils/unittest/src/SLListTest.cc \
precious/Utils/unittest/src/DLListTest.cc \
precious/Ut
Hi folks,
my Makefile.am deep inside a subdirectory contained this:
:
SUBDIRS = unittests \
../generic_datainterface \
../generic_resource_type \
../generic_memorymodel \
../hub_backend \
../genesys_su
Ed Hartnett wrote:
> Howdy all!
>
> If I have some tests, and want to run them with command line options,
> how do I do it?
>
> For example, I have to have a shell script to run tst_parallel below,
> because it needs to be called with the poe command, and some env vars
> set. In other cases, I ne
Bruno Haible wrote:
> Harald Dunkel wrote:
>
>>It would be pretty cool if Automake could append this
>>automagically when generating the Makefile.in files.
>
>
> How could it know to which Makefile.ins it should append it? There are
> typically several Makefile.a
Tom Howard wrote:
> Hi Bruno,
>
>
> I used ax_add_mk_macro in ax_dist_rpm.m4 so that in my projects I can run
>
> make dist-rpm
>
> to create an rpm distribution. Using these macros, all I have to add to
> the Makefile.am is @INCLUDE_MK@ at the end.
>
It would be pretty cool if Automake coul
In the meantime I have patched automake, e.g.
diff -ur automake-1.9.5/lib/am/library.am automake-1.9.5.new/lib/am/library.am
--- automake-1.9.5/lib/am/library.am2003-06-02 09:08:40.0 +0200
+++ automake-1.9.5.new/lib/am/library.am 2005-05-22 20:22:51.0 +0200
@@ -15,6 +15,7 @@
Hi folks,
The targets in BUILT_SOURCES are unconditionally built for
'make all' and 'make install' and 'make check'. Very clumsy.
I would like to generate some code for an optional convenience
library exactly when needed.
Would there be a way to introduce a derived variable
libmylib_a_PREDEPEND,
Stan Guillory wrote:
>
> 2) Is there a way to force automake to set the depmode to makedepend?
>
What happens if you run
make CCDEPMODE=depmode=makedepend CXXDEPMODE=depmode=makedepend
? If it works as expected, then you might add a test for the
compiler version to your configure.ac, a
Stepan Kasal wrote:
> Hello,
>
> On Tue, Apr 19, 2005 at 02:50:03PM +0200, Harald Dunkel wrote:
>
>>Why are the configure substitutions removed?
>
>
> because they are not known at the time when automake runs.
>
Understood.
Would it be possible to compute
Hi folks,
I have a question about how the *_DEPENDENCIES are computed
from *_LDADD:
If I write a Makefile.am
bin_PROGRAMS = myprog
MYLIB = liba.a
myprog_LDADD = \
$(MYLIB) \
libb.a \
-Lsomepath \
-lother
the
Warren Young wrote:
> Harald Dunkel wrote:
>
>>
>> If I add something like
>>
>> lib_LIBRARIES = /some/static/old/lib2install.a
>
>
> ...then you're probably doing something wrong. That should probably go
> in LDADD.
>
No, LDADD does not help
Hi folks,
If I add something like
lib_LIBRARIES = /some/static/old/lib2install.a
to my Makefile.am, then the generated Makefile includes
/some/static/old/lib2install.a in the list of files to
remove for the clean target.
This is certainly not correct. Only targets that are
built by this Makefil
Ralf Wildenhues wrote:
>
> Stab in the dark:
> Idea 1)
> Put only everything to be installed both times in
> bin_PROGRAMS
> etc, the other stuff in
> noinst_PROGRAMS
> etc. Use install-data-local and install-exec-local for installing the
> extra stuff within `make install', but only if some v
Kevin Teich wrote:
> Hello,
>
> I would like to add a target to our source tree that only installs
> certain files such 'make install' will install everything normally, but
> 'make release' will only install certain files.
>
I would be interested in something like this, too.
A possible implemen
Stepan Kasal wrote:
Hello,
On Tue, Mar 08, 2005 at 11:56:56PM +0200, Paul Pogonyshev wrote:
foo.c foo.h : $(srcdir)/foo.list $(PARSE_LIST)
$(PARSE_LIST) $(srcdir)/foo.list foo.h foo.c\
|| (rm -f foo.c foo.h ; exit 1)
This rule can break with parallel make.
You can solve the
60 matches
Mail list logo