Thanks Eike,
that helps, but I can't understand why it is necessary?
Within the trace message output there was no semicolon, see below.
My first try was this:
message(EXTRA_LIBS=${EXTRA_LIBS}) # =>
EXTRA_LIBS=/usr/lib/libpthread.dylib/usr/lib/libcrypto.dylib # OK but no
spaces
message(EXTR
What is the different between cmake.app and cmake-gui?
When i try to create a new project with cmake-gui started from shell
it always fail.
But the GUI version from /Application works fine.
claus-kleins-macbook-pro:ftplib clausklein$ cmake-gui -G Ninja /Users/
clausklein/Workspace/cpp/ftp
We use a complex cmake project with a lot of build order dependencies.
If something is misspelled, I would expect that cmake find this error.
Example:
cmake_minimum_required(VERSION 2.8.5)
project(ftplibcpp CXX)
add_executable(ftpList ftpList.cpp)
target_link_libraries(ftpList ftpcpp ${LIBS})
LIBS})
add_executable(ftpGet ftpGet.cpp)
target_link_libraries(ftpGet ftpcpp ${LIBS})
add_executable(ftpPut ftpPut.cpp)
target_link_libraries(ftpPut ftpcpp ${LIBS})
# install the bin
install(TARGETS ftpPut ftpGet ftpList DESTINATION bin)
On 14.07.2012, at 12:22, Eric Noulard wrote:
2012/7/
77 (INCLUDE)
# CMakeLists.txt:3 (project)
# adjust the default behaviour of the FIND_XXX() commands:
# search NEVER for programs in the build host directories!
set(CMAKE_FIND_ROOT_PATH_MODE_PROGRAM NEVER)
# search headers and libraries in the target environment only.
set(CMAKE_FIND_ROOT_PATH_MODE_L
claus-kleins-macbook-pro:build clausklein$
On 12.07.2012, at 22:23, Claus Klein wrote:
The ranlib and ar from build host is used.
Any idea what I should try or do would help?
--
Powered by www.kitware.com
Visit other Kitware open-source projects at
http://www.kitware.com/opensource/opensource.html
I tried with the current nightly cmake binary to cross compile with
mingw tools on Darwin.
But it fails and I have no right idea what goes wrong.
First, the rc compile was not found:
+ cmake -G 'Unix Makefiles' -DCMAKE_C_COMPILER=i386-mingw32-gcc -
DCMAKE_CXX_COMPILER=i386-mingw32-g++ -DCMAK
version, but not sure.
This future needs a runtime test, IMO
Claus
On 03.06.2012, at 22:55, Peter Kümmel wrote:
On 03.06.2012 01:02, Claus Klein wrote:
Hi Peter,
after a quick test with our code, I found again the problem with unix
ar utile for static libs. it does not support a linker
I found the problem.
Now Eclipse and Ninja works too. .-))
NinjaAddToEclipseGenerator.patch
Description: Binary data
//Regards
Claus
On 03.06.2012, at 12:15, Claus Klein wrote:
Hi Peter,
I have tried it myself, but with different luck.
The Kdevelop works, but EclipseGenerator crashes
02.06.2012, at 18:58, Claus Klein wrote:
Hi Peter,
thanks for the commit. I will do my tests with this code again.
But I have 2 more questions:
It is possible to use Nina with Kdevelop and Eclipse generators too?
And, too give more people a chance to test with ninja, is it
possible to enable it
Hi Peter,
after a quick test with our code, I found again the problem with unix
ar utile for static libs. it does not support a linker response file!
And is is not clear to me who sets this cache values:
CMAKE_CXX_RESPONSE_FILE_LINK_FLAG:STRING=-Wl,@
CMAKE_CXX_USE_RESPONSE_FILE_FOR_OBJECTS:BO
Hi Peter,
thanks for the commit. I will do my tests with this code again.
But I have 2 more questions:
It is possible to use Nina with Kdevelop and Eclipse generators too?
And, too give more people a chance to test with ninja, is it possible
to enable it at nightly builds for Windows and MAS-
Hi Bill,
I have tried to prepare a patch. It generates links rules like that:
# Rule for linking CXX executable.
rule CXX_EXECUTABLE_LINKER
command = $PRE_LINK && /opt/local/libexec/ccache/g++ $FLAGS -Wl,-
search_paths_first -Wl,-headerpad_max_install_names $LINK_FLAGS -Wl,@
$out.rsp -o $
Hi Bill
Yes send me the patches please
Claus
On 29.05.2012, at 23:12, Bill Hoffman wrote:
> On 5/29/2012 4:00 PM, Claus Klein wrote:
>> Hi All,
>>
>> I am working with Nightly cmake build with ninja enabled on windows.
>> While testing a real complex project
Hi All,
I am working with Nightly cmake build with ninja enabled on windows.
While testing a real complex project build, code generation and
compiling works, but linking fails.
The command line with a view pre and post build rules including a huge
count of libs and linkerflags is to long.
Please tack a lock at my builds tooAll of this errors are NOT caused by ninja!http://open.cdash.org/buildSummary.php?buildid=2315144But with ninja I get the same result.Some of the errors are caused by the fact that I use the current gcc47 compiler (build with macports), which does not understand t
Ninja builds.
You can use the same source tree if you wish.
Richard
On May 27, 2012, at 11:19 AM, Claus Klein wrote:
Hi Richard,
I have set it to i386.
I have Xcode 3.1 installed year ago to bootstrap macports.
I never use it!
This problems for example are caused that my gcc47 compiler is
ard
On May 27, 2012, at 7:21 AM, Claus Klein wrote:
I have uploaded my test result:
http://open.cdash.org/viewTest.php?onlyfailed&buildid=2312336
Claus
--
Powered by www.kitware.com
Visit other Kitware open-source projects at
http://www.kitware.com/opensource/opensource.html
Please keep
I have uploaded my test result:
http://open.cdash.org/viewTest.php?onlyfailed&buildid=2312336
Claus
On 27.05.2012, at 12:10, Claus Klein wrote:
Hi all,
while testing the ninja generator I created a reference binary tree
with gmake an gcc47 installed with macports.
I created a new e
y it without using ninja, but rather Unix makefiles
and everything else the same. It will be interesting to see what
fails under those conditions. Again, submitting to the dashboard is
a better way to document the particulars of the failure.
Richard
On May 27, 2012, at 5:26 AM, Claus Klein wrote:
ay 27, 2012, at 5:26 AM, Claus Klein wrote:
Only a view tests fails, but most of the problems are NOT caused by
ninja.
It seems, that while testing, cmake on Darwin use options only
valid for old native apple gcc?
But I use gcc47 installed with macports!
--
Powered by www.kitware.com
Hi all,
I tried to build cmake with ninja generator enabled and build cmake
itself with ninja.
Nice, it works ;-))
Only a view tests fails, but most of the problems are NOT caused by
ninja.
It seems, that while testing, cmake on Darwin use options only valid
for old native apple gcc?
Hi all,
while testing the ninja generator I created a reference binary tree
with gmake an gcc47 installed with macports.
I created a new empty dir with
mkdir "/Users/clausklein/Downloads/My Cmake Test"
and created the makefiles with a build scripts.
It seems that the spaces in CWD makes tro
er generators, you do not need to add the -
DMAKE_SUPPORTS_SPACES
Richard
On May 22, 2012, at 5:18 PM, Claus Klein
wrote:
claus-kleins-macbook-pro:CompileCommandOutput clausklein$ "/usr/
local/CMake 2.8-8.app/Contents/bin/cmake" -G Ninja -
DMAKE_SUPPORTS_SPACES=1
--
Powered by w
it works on my macbook:
"/usr/local/CMake 2.8-8.app/Contents/bin/cmake" --version
cmake version 2.8.8.20120520-g4742e
claus-kleins-macbook-pro:CompileCommandOutput clausklein$ "/usr/local/
CMake 2.8-8.app/Contents/bin/cmake" -G Ninja -DMAKE_SUPPORTS_SPACES=1
-- The CXX compiler identification
: UNCOMMENT THIS LINE after
first successfully build!
)
add_executable(test ${file_list})
On 18.05.2012, at 23:48, Bill Hoffman wrote:On 5/18/2012 5:37 PM, Claus Klein wrote:Hi,I downloaded the newest version 2.8 of cmake for MACOS and dit not foundthe new generator for ninja.So I downloaded the sources and b
Hi,
I downloaded the newest version 2.8 of cmake for MACOS and dit not
found the new generator for ninja.
So I downloaded the sources and build it myself.
It seems to work fine and just 2 times faster than with make.
Why is it disabled?
With regards
Claus
--
Powered by www.kitware.com
Hi Brad,
I have tested this version today, but the problem still exist!
Best regards,
Claus
On 23.08.2010, at 14:13, Brad King wrote:
On 08/21/2010 03:31 AM, Claus Klein wrote:
can you please check if this may related also to this BUG:
http://cmake.org/Bug/view.php?id=11153
It happens on
!
That is really strange.
Claus
On 19.08.2010, at 22:59, Claus Klein wrote:
Hi,
I have a project with CORBA idl files, which will be compiled with a
tao idl compiler to source and header files via a:
add_custom_command(OUTPUT output1 [output2 ...] ...)
I defined a source group with
Hi,
can you please check if this may related also to this BUG:
http://cmake.org/Bug/view.php?id=11153
It happens on Windows only at a root dir.
Thanks,
Claus
On 16.08.2010, at 15:31, Brad King wrote:
On 08/13/2010 06:03 PM, J Decker wrote:
Yes, this patch fixes the problem thank you.
Tha
2010 20:34, Claus Klein skrev:
Today, I discovered than in the special case of a substituted
PROJECT_SOURCE_DIR,
all build files for subdir packages are created in PROJECT_SOURCE_DIR
and not at the PROJECT_BINARY_DIR.
I use windows cmd shell
subst w: z:/some/deep/path/to/source/dir
Than I create a new
Hi Alex,
No, I dit not write a bug report yet.
The 10994 seems an other problem, I have the CMakeLists.txt at the
root dir, together with my build subdir.
So my binary build tree is located at inside the source tree at
project root.
And this is the root of the substituded drive.
Claus
On
Hi,
I have a project with CORBA idl files, which will be compiled with a
tao idl compiler to source and header files via a:
add_custom_command(OUTPUT output1 [output2 ...] ...)
I defined a source group with:
source_group(idlfiles REGULAR_EXPRESSION *.idl)
The generated sour
Today, I discovered than in the special case of a substituted
PROJECT_SOURCE_DIR,
all build files for subdir packages are created in PROJECT_SOURCE_DIR
and not at the PROJECT_BINARY_DIR.
I use windows cmd shell
subst w: z:/some/deep/path/to/source/dir
Than I create a new build dir:
I have a cpp project with some lib targets and executables, which
depends on the libs and the header files.
I want to link the libs from my build/src tree and not the installed
versions.
With this snip below from my CMakeFile.txt it works on my MacBook and
cygwin after a make clean,
but wit
Hi,
I was missing a right cmake module to find CppUnit lib and header files.
Perhaps this is useful for the next cmake release?
It works with pkgconfig if available, but too without, tested on Win32
and MAC.
regards,
Claus
FindCppUnit.cmake
Description: Binary data
_
Alexander Neundorf schrieb:
...
So indeed a relative CMAKE_INSTALL_PREFIX is interpreted relative to the build
dir.
Do you think this should be handled differently ?
Then c:/usr would have to be translated at some point to "/c:/usr" if cross
compiling on some UNIX.
Would this make sense ?
Al
On 08.01.2010, at 02:21, Alexander Neundorf wrote:
On Wednesday 06 January 2010, Claus Klein wrote:
Yes, I did it again and again with rm -rf build.
Same result!
As I said: "The problem occurs always when I run cmake again, no
matter of Windows or Windows-gcc is used."
I don&#
/Modules/FindLibSmi.cmake".
+ exit
claus-kleins-macbook-pro:libsmi-latejuly09-cmake clausklein$
I hope you see that something should changed with make install.
//regards
Claus
On 08.01.2010, at 22:26, Claus Klein wrote:
Hi David,
Yes, this works fine when I change the CMAKE_INSTALL
?
Is there a template for this generated cmake_install.cmake?
Claus
On 08.01.2010, at 17:49, David Cole wrote:
On Fri, Jan 8, 2010 at 12:37 AM, Claus Klein
wrote:
On 08.01.2010, at 02:25, Alexander Neundorf wrote:
On Thursday 07 January 2010, Claus Klein wrote:
Hi
i want to install a
On 08.01.2010, at 02:25, Alexander Neundorf wrote:
On Thursday 07 January 2010, Claus Klein wrote:
Hi
i want to install a cross compiled (build host is a MAC OS X) project
to a temporary state dir to get an archive to distribute the
binaries.
The target is win32 (mingw), compiled to be
Hi Paul,
yes, I do it in this way and it works good.
Only if the target system is Windows, there may be a Problem if prefix
is something like "C:/programs" ...
But in most embedded unix like systems, there is no problem.
Claus
On 07.01.2010, at 21:30, Paul Chavent wrote:
Hi cmake mai
Hi
i want to install a cross compiled (build host is a MAC OS X) project
to a temporary state dir to get an archive to distribute the binaries.
The target is win32 (mingw), compiled to be installed at c:/usr as
CMAKE_INSTALL_PREFIX.
This prefix is used while compile the binaries, so I can't
Yes, I did it again and again with rm -rf build.
Same result!
As I said: "The problem occurs always when I run cmake again, no
matter of Windows or Windows-gcc is used."
Claus
On 06.01.2010, at 20:23, Alexander Neundorf wrote:
On Wednesday 06 January 2010, Claus Klein wrote
CMAKE_SYSTEM_NAME:UNINITIALIZED=Windows-gcc
claus-kleins-macbook-pro:build clausklein$
Perhaps this his the reason?
Claus
On 04.01.2010, at 19:22, Alexander Neundorf wrote:
On Sunday 03 January 2010, Claus Klein wrote:
I have problems to crosscompile on MAC-OS for MinGW.
How kann I prevent this strange
scompiling for MinGW
//regards
Claus
On 04.01.2010, at 19:22, Alexander Neundorf wrote:
On Sunday 03 January 2010, Claus Klein wrote:
I have problems to crosscompile on MAC-OS for MinGW.
How kann I prevent this strange compiler flag on a Darwin Plattform?
Tha
/Bug/view_all_bug_page.php
Mike
On Jan 3, 2010, at 8:22 AM, Claus Klein wrote:
Thanks Bart,
I have xcode installed, but not every compiler on my MAC compiles
for MAC-OS.
I have mingw32 installed from macports, and the target is Win32, so
this option test
"-mmacosx-version-min=
if you select the 10.5 sdk
this is from memory, so just give it a brief check.
On Jan 2, 2010, at 18:30, Claus Klein wrote:
I have problems to crosscompile on MAC-OS for MinGW.
How kann I prevent this strange compiler flag on a Darwin Plattform?
Thanks
Claus
---
I have problems to crosscompile on MAC-OS for MinGW.
How kann I prevent this strange compiler flag on a Darwin Plattform?
Thanks
Claus
-
cmake -DCMAKE_C_COMPILER=i386-mingw32-gcc -DCMAKE_CXX_COMPILER=i386-
mingw32-g++ --debug-trycompile --debug-output -
according to
Notes to authors of FindXXX.cmake files
We would like all FindXXX.cmake files to produce consistent variable
names.
Please use the following consistent variable names for general use.
XXX_FOUND :
Set to false, or undefined, if we haven't found, or don't want to use
XXX.
XXX_
www.bluequartz.net
Principal Software Engineer Dayton, Ohio
On Dec 21, 2009, at 3:59 PM, John Drescher wrote:
On Mon, Dec 21, 2009 at 3:50 PM, Claus Klein
wrote:
Hi all,
I tried to use:
set(CMAKE_USE_RELATIVE_PATHS ON)
in my CMakeLists.txt.
But I noticed
Hi Aashish,
yes the set command is well explained. Thanks.
But for some CMAKE variables this not true!
I guess that GUI vars have always set with FORCE?
Claus
Aashish Chaudhary schrieb:
Is this useful?
http://www.cmake.org/cmake/help/cmake-2-8-docs.html#command:set
_
John Drescher schrieb:
On Mon, Dec 21, 2009 at 3:50 PM, Claus Klein wrote:
Hi all,
I tried to use:
set(CMAKE_USE_RELATIVE_PATHS ON)
in my CMakeLists.txt.
But I noticed, that some cmake variables can't be changed in that way.
In the CMakeCache.txt, it is still OFF?
I have no
Hi all,
I tried to use:
set(CMAKE_USE_RELATIVE_PATHS ON)
in my CMakeLists.txt.
But I noticed, that some cmake variables can't be changed in that way.
In the CMakeCache.txt, it is still OFF?
What goes wrong?
Which is the right way do set CMAKE_BUILD_TYPE and CMAKE_USE_RELATIVE_PATHS?
//regar
On Tuesday 23 January 2007 18:10, kitts wrote:
> On Tuesday 23 January 2007 00:03 IST, kitts wrote:
> > 2) On windows, we would like to use VS as the IDE and KDevelop on linux.
> > Both are capable of crosscompiling (use non-native compilers). I think
> > this is only possible by setting CMAKE_C_CO
On Saturday 20 January 2007 15:04, Alexander Neundorf wrote:
> Hi,
>
> Von: Peter Soetens <[EMAIL PROTECTED]>
>
> ...
> > > I miss some features needed for cross-compiling, linking and
> > > installing to
> > > a stage dir. The PKGCONFIG(package includedir libdir linkflags
> > > cflags),
> > > fo
On Monday 15 January 2007 11:28, Jorrit Schaap wrote:
> Hi,
>
> I'm using cmake on linux with gcc.
> The dependencies feature of cmake is realy great, but sometimes I would like
> to switch it of. A complete build of all our libs takes a couple of ours.
>
> Whenever I add for example a comment t
On Friday 12 January 2007 08:43, Alexander Neundorf wrote:
>
> Von: Claus Klein <[EMAIL PROTECTED]>
>
> > Hi,
> >
> > to be more flexible, here is a patch for UsePkgConfig.cmake
> >
> > # PKGCONFIG("dbus-1 >= 1.0.0" ...) is now als
On Friday 12 January 2007 11:39, Axel Roebel wrote:
> On Friday 12 January 2007 10:17, Claus Klein wrote:
> > On Friday 12 January 2007 08:43, Alexander Neundorf wrote:
> > > Von: Claus Klein <[EMAIL PROTECTED]>
> > >
> > > > Hi,
> > &
On Friday 12 January 2007 08:43, Alexander Neundorf wrote:
>
> Von: Claus Klein <[EMAIL PROTECTED]>
>
> > Hi,
> >
> > to be more flexible, here is a patch for UsePkgConfig.cmake
> >
> > # PKGCONFIG("dbus-1 >= 1.0.0" ...) is now als
On Wednesday 10 January 2007 22:58, Claus Klein wrote:
> Hi,
>
> I think it would be more usefull if the Module section in the cmake(1) man
> page would be sorted.
>
> ck
>
Hello again,
a question to the coders:
Are patches welcome and if so, should it be a patch agains
Hi,
to be more flexible, here is a patch for UsePkgConfig.cmake
# PKGCONFIG("dbus-1 >= 1.0.0" ...) is now also possible.
# Optional, for stage building, the
# PKGCONFIG_OPTIONS and PKG_CONFIG_PATH environment variable are used if set!
# see man pkg-config for more info; ck
e.g. PKG_CONFIG
Hi,
I think it would be more usefull if the Module section in the cmake(1) man page
would be sorted.
ck
--- cmake-2.4.5/Source/kwsys/Directory.cxx.org 2006-12-04 17:04:05.0 +0100
+++ cmake-2.4.5/Source/kwsys/Directory.cxx 2007-01-10 22:47:44.0 +0100
@@ -139,6 +139,10 @@
this
On Wednesday 10 January 2007 20:22, Bill Hoffman wrote:
> Claus Klein wrote:
> >
> > thanks for quick anser.
> >
> > If it works with:
> > cmake -DSIZEOF_SIZE_T=4 ..
> > it is ok for me.
> >
> > But my real intention is:
> >
> > It
On Wednesday 10 January 2007 17:00, Bill Hoffman wrote:
> Claus Klein wrote:
> > Hallo,
> >
> > the following Modules use TRY_RUN, which must fail while crosscompile:
> >
> > CheckTypeSize.cmake:TRY_RUN(${VARIABLE} HAVE_${VARIABLE}
> > FindThreads.cm
Hallo,
the following Modules use TRY_RUN, which must fail while crosscompile:
[EMAIL PROTECTED]:~/downloads/cmake-2.4.5/Modules> grep TRY_RUN *.cmake
# OK CheckCSourceRuns.cmake:TRY_RUN(${VAR} ${VAR}_COMPILED
# OK CheckCXXSourceRuns.cmake:TRY_RUN(${VAR} ${VAR}_COMPILED
CheckTypeSize.cmake
ckage again.
As a last point, it wout be nice if cmake generates makefiles according to:
http://www.gnu.org/prep/standards/standards.html#Makefile-Conventions
I think, most people expect this as common standard, or not?
For me, cmake is a good tool, but if more project change to it without ca
67 matches
Mail list logo
