Alex Makarenko wrote:
Using \${CTEST_CONFIGURATION_TYPE} works. Is there some official document
where I could look up these variables in the future?
There is some documentation in the CMake book. There is also a
user-maintained list here:
http://www.cmake.org/Wiki/CMake_Useful_Variables
Christian Verbeek wrote:
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
My directory structure is something like this
root/
|-proj1/
| |-share
|
|-proj2/
|-share
My top level CMakeLists.txt is in proj1 using proj2 with SUBDIRS(
proj2 ) to
Orion Poplawski wrote:
Why is FindPythonInterp called that rather than FindPython? Would be
more in line with others (FindPerl, FindRuby), etc.
This is historical. No one has contributed a real FindPython that uses
disutils:
http://www.cmake.org/Bug/bug.php?op=showbugid=2257
-Brad
Jorge Rodriguez wrote:
I am having some problems building a 64 bit version of my application
using the NMake generator on 64 bit Windows with Visual Studio 8.0.
I am using the Microsoft Platform SDK command prompts that can be opened
through the start menu. I can use the Visual Studio 8 2005
Sagnes, Frederic wrote:
Hello,
I am building a project for Windows and Unix that relies on static multi
threaded runtime library on Windows (static STLport needs it). I'm using
Visual Studio .Net 2003 7.1
How can I switch the default libraries (/MD and MDd switches) to the static
ones
Jan Woetzel wrote:
Hi,
has anybody a geneal purpose FindBoost.cmake?
I started with the one from Andrew Maclean, see
http://public.kitware.com/pipermail/cmake/2005-October/007324.html
There is a bug entry where Andrew and I are developing one. It will be
added to CMake soon.
William A. Hoffman wrote:
At 04:41 PM 7/13/2006, Alan W. Irwin wrote:
How do I get access to the actual link flags used for the link command
generated by cmake (which presumably includes all the appropriate -L and -l
flags to access lib[xyz] and libexternal[12] as well as platform-dependent
Alexander Neundorf wrote:
Original-Nachricht
Von: Brad King [EMAIL PROTECTED]
Alexander Neundorf wrote:
Hi,
it would be nice if it would be possible to have for each object file
also a target which only preprocesses the file and one which
preprocesses and compiles
Erk wrote:
I'm switching from autotools to cmake.
I did found everything I want but the
@configure_input@ thing.
For emulating my custom configure-generated file
I use the CMake CONFIGURE_FILE which is
just fine.
Nevertheless I found interesting the predefined
@configure_input@
Louis Kruger wrote:
Hi,
I tried to build CMake 2.0.6 - 2.4.2 without success. ( I need it for
VTK which I need for ProcessView.)
1. It seem to fail on creating directories,
2. Seem to fail on copying of a file
3. Seem to fail on setting permmissions.
I also wished that CMake
Francesco Montorsi wrote:
Hi all,
I'm involved in a project (about a package manager) which needs to
be able to build source packages which are using a cmake-based build
system.
Since I'm quite new to cmake, I'm asking you some help about the
following question:
is it possible to
Brandon J. Van Every wrote:
Brandon J. Van Every wrote:
Can a CMake script invoked with
EXECUTE_PROCESS(
COMMAND ${CMAKE_COMMAND} -P myscript.cmake
RESULT_VARIABLE myresult
)
return a value? I don't really want to use stdout. I need that for
status messages. I could regex the stdout
Alan W. Irwin wrote:
I believe you need more information than -L and -l options; you also
need to
know the special link options that are required (for rpath, shared versus
static, bundle or whatever) on the particular platform that is being used
for the build.
Bill mentioned above the
Jan Woetzel wrote:
Jan Woetzel wrote:
(1) I think XEmacs is using an xterm but doing his own extar coloring
to mark e.g. errors in red. Is that the problem?
I think the problem should be fixed in CMake to be even more
conservative with coloring, e.g. with a runtime check on env. var
[EMAIL PROTECTED] wrote:
It appears that Digital Unix's libc doesn't have snprintf/vsnprintf,
which are used in quite a few places - mostly in source under the
Utilities and CTest directories.
I think this is the first time anyone has tried building CMake on
Digital Unix. Officially supported
Christian Ehrlicher wrote:
Von: Brad King [EMAIL PROTECTED]
Christian Ehrlicher wrote:
snip
What kind of files are these ?
Ideally you should add them to the list of sources for the target. If
they
don't have a source-file suffix, they won't be compiled. Or you could
use
Brandon J. Van Every wrote:
Using CMake 2.4.2. Here's my mighty CMakeLists.txt:
ADD_CUSTOM_TARGET(test
COMMAND ${CMAKE_COMMAND} -E echo It's difficult to handle apostrophes.
)
I used CCMake under Cygwin to generate the Makefile. It dies thus:
Brandon J. Van [EMAIL PROTECTED] ~/echo
Filipe Sousa wrote:
Philipp Meinen wrote:
Hello
I recently used cmake for a project.
So I added a line like this to my CMakeLists.txt:
PROJECT(ProjectName CXX)
Later on, when i tested the build-environment
i saw error lines like this:
cd /some/directory cr libFOO.a
Brandon J. Van Every wrote:
I explained the situation pretty clearly, I thought.
E:\msys\1.0\bin\makefile does *NOT* run under a Windows command prompt.
It's not supposed to. As I said before, there's a comment at the top of
the binary that says it's not supposed to. Just to reiterate
Brandon J. Van Every wrote:
Brad King wrote:
Brandon J. Van Every wrote:
I explained the situation pretty clearly, I thought.
E:\msys\1.0\bin\makefile does *NOT* run under a Windows command prompt.
It's not supposed to. As I said before, there's a comment at the top of
the binary
Brandon J. Van Every wrote:
Ok, the CMake 2.4.2 documentation for the projectName version of
TRY_COMPILE is only 3 short sentences. IIUYC, you are saying I'm going
to have to write a CMakeLists.txt wrapper, and a subdirectory, for each
and every tool like makeinfo, if I want to prove that
Flávio P. Duarte wrote:
Hi,
I am trying to test a shared library using cmake. In my top level
directory, I have the source directory (src), which contains the source
files to build the library, and a test directory (test), which contains
the source files to build the binaries that will be
Louis Kruger wrote:
One of the persons indicated that more info is required ito the problems
that I have. I am busy compiling/capturing problems, but the results are
kind of big. How should I handle that in terms of posting?
Just cut-and-paste the portion of your log at which the first error
Alan W. Irwin wrote:
On 2006-07-24 11:14-0400 Brad King wrote:
Summary: Just set INSTALL_RPATH on the module target to be the rpath
needed in the install tree. Then things will just work.
I followed that advice for INSTALL_RPATH with libplplotd, and it worked,
but
unlike the build tree
Karl Merkley wrote:
I have a simple Fortran project that I am testing with cmake.
PROJECT(multi_patch Fortran)
SET( SRCS
aAdjKeep.f
Main_mp.f
)
ADD_EXECUTABLE(multi_patch ${SRCS})
However, the first file is a Fortran 95 module and when I try to build I get
the following
Giulio Eulisse wrote:
Ciao,
I'm trying to use cmake to build a project with many libraries, with
lots of dependencies. For this reason I would like to do something that
allows the following.
Given 3 libs A B C each one with its own dir and with
A depends on B
B depends on C
I want to
Anton Deguet wrote:
Hello,
I am trying to setup a couple of nightly builds on a single Windows box.
Since the build requires DLLs (to test our __declspec and Python
wrappers), I have been trying to use the MSVC devenv command line
with /useenv to set the DLL and Python search paths per
[EMAIL PROTECTED] wrote:
I need to compile a code generator, run the code generator, then compile the
code that the code generator produces. Do I need to complie the generator,
then use the EXECUTE_PROCESS command to run it, then have the generated code
compiled, or should I use the
Flávio P. Duarte wrote:
my project is an application with 4 libs. The application is located in
the root directory and each lib corresponds to one subdir in the root
dir. Since I am using Qt, I have a bunch of header files that are
generated at compile time and, to make things worst, one lib
Alan W. Irwin wrote:
One of our PLplot developers with access to Mac OS X ran into the following
error message for the shared libraries for that platform:
ld: common symbols not allowed with MH_DYLIB output format with the -
multi_module option
I have no idea what constraints there are
Dean Inglis wrote:
Hi,
I'm trying to run CMake 2.4.2 from
a command prompt to get an NMake Makefiles build config
(current cvs VTK) using the free MS VS8 compiler. Cmake
keeps coughing up a
CMake error: your RC compiler: CMAKE_RC_COMPILER_FULPATH-NOTFOUND
etc. etc.
If I edit my
Marc-André Laverdière wrote:
Hello CMakers,
I'm really a n00b...
So, I am building each subdirectory into a static library and I want to
put all those libraries into a nice big dynamic library. The big dynamic
library is the objective here.
I'm doing it as such:
ADD_LIBRARY(XYZ SHARED
Henty Waker wrote:
Hi there
I have some C code that I cross compiled for a small mobile point-of-
sale device running uClinux on an arm processor.
I currently use plain Makefiles to build a number of static libs that
are linked to a single exe. The Makefiles work fine, but they're messy.
Tim Teulings wrote:
On the other hand PKGCONFIG already has options to directly return
apropiate values for these two variables, so why are they not used in
the PKGCONFIG macro? I would expect that the result of the PKGCONFIG
macro is already optimal prepared for easy call of CMake macros. It
speedy wrote:
VC++ 6 generator -
[snip]
// add libraries to executables and dlls (but never include
// a library in a library, bad recursion)
// NEVER LINK STATIC LIBRARIES TO OTHER STATIC LIBRARIES
if ((target.GetType() != cmTarget::SHARED_LIBRARY
Crni Gorac wrote:
Everything in my project, let's call it foo, in pure ANSI C, except
that I'm using flex and bison for parsing input, then I need getopt()
function for processing program command line options and finally, I'm
using MPI library. Source code files are distributed between
Alan W. Irwin wrote:
I am in the process of emulating the autotools STDC_HEADERS macro which
does
the following:
# Specifically, this macro checks for stdlib.h', stdarg.h',
# string.h', and float.h'; if the system has those, it probably
# has the rest of the ANSI C header files. This
Brandon J. Van Every wrote:
William A. Hoffman wrote:
The CMAKE_BUILD_TYPE is by default empty on all UNIX/posix platforms.
Sort of goes back to when there was no build type for unix. I suppose
the chicken project could force it to not be empty. if empty, then do
a cache force to what you
Richard Fuchs wrote:
I'm getting this error
CMake Error: Error required internal CMake variable not set, cmake may
be not be built correctly.
Missing variable is:
CMAKE_Java_LINK_EXECUTABLE
where/how does this variable get set?
What is your CMakeLists.txt file code that causes this
Brian Macy wrote:
I have a need to determine what directory object files are placed in
when built using CMake. In 2.0, I just picked them up in the same
directory as the source. In 2.4, they are placed in a completely new
location (src_dir/CMakeFiles/project.dir/). Is there an CMake
Richard Fuchs wrote:
So why when I have a simple CMakeLists.txt file like below am I getting
this error? Why is it trying to put the .class files in some funky
directory instead of just where the .java files are? Is there a way to
set the OBJECT_DIR that's in the CMakeJavaInformation.cmake
Matt Rogers wrote:
As suggested, I've copied most of CMake's code into the KDevelop 4 codebase
for use in the cmake integration. However, I'd really love to have an
external library of some sort that I can link against rather than having to
keep up with CMake's code. This leads to a few
Steve Johns wrote:
Recently, in an MS VC++ 7.1 (.NET 2003) build prepared using CMake, I've
had a problem with the application finding a config file that it needs
to read at startup. Even though I placed the file in the app's startup
dir (i.e. 'blahblah/Debug', when debugging ) the app was
Jorge Rodriguez wrote:
Brad King wrote:
This property is not actually stored in the project files, so CMake
cannot set the value. It is stored in some file inside the %APPDATA%
directory for Visual Studio.
Actually I believe it's stored in the projectname.suo file. It's a
hidden file
Richard Fuchs wrote:
Spoke to soon, it seems to be a problem with where javac is invoked.
With the current makefiles that are created, the javac -d option has a
relative path from where the source files are located, but javac is
being invoked from the root project dir. I was able to modify
William A. Hoffman wrote:
At 02:21 AM 8/3/2006, Jan Woetzel wrote:
Christian Ehrlicher wrote:
SET(myproject_SRCS
myproject.cpp
myproject.jisp
)
ADD_LIBRARY(myproject SHARED ${_myproject})
myproject.jisp.rule is created and executed correct, but myproject.jisp doesn't
have
Crni Gorac wrote:
Back to my OpenMP based project, I was able to build following cmake
test regarding is given compiler supporting OpenMP or not. Basic idea
is like in corresponding macro from autoconf macros archive: try to
check is one of OpenMP functions (omp_set_num_threads() here)
Brandon J. Van Every wrote:
Cygwin expected nomenclature is:
cygwhatever-xxx.dll - dynamic link lib where -xxx is a version number
libwhatever.dll.a - import lib
libwhatever.a - static lib
MinGW expected nomenclature is:
libwhatever.dll - dynamic link lib
libwhatever.dll.a - import lib
Richard Fuchs wrote:
Could I just use EXECUTE_PROCESS to call make (with a hand created
Makefile) or even the java compiler itself to get around the problems
I'm having with cmake and java not liking each other very much?
You can use ADD_CUSTOM_COMMAND and ADD_CUSTOM_TARGET to do whatever you
[EMAIL PROTECTED] wrote:
FIND_LIBRARY( MY_LIB NAMES my MY PATHS c:\\pkg\\c\\cmake\\my\\o
c:/pkg/c/cmake/my/o )
You don't need the backslash version. CMake always uses forward slashes
in its own scripts.
#link in the support libray
TARGET_LINK_LIBRARIES( which my )
You have to use the
Alexander Neundorf wrote:
Attached is a first try of a patch against 2.4 branch.
It introduces two new variables CMAKE_CREATE_PREPROCESS_RULES and
CMAKE_CREATE_ASSEMBLE_RULES. I think especially the name
CMAKE_CREATE_ASSEMBLE_RULES is not good (since actually everything except
assembling
Nuno Lopes wrote:
Thanks for your help. But in theory the cflags can include anything they
like, (e.g. -DHAVE_XPTO -I/tmp/xpto -Wl,-O), so probably
INCLUDE_DIRECTORIES isn't the right one to choose, am I right?
Full-fledged integration of CMake and pkgconfig has not been
implemented. The
Dataflow wrote:
This was working with 2.0 and feels natural, but no longer works with 2.4.
It boils down to the ordering of LINK_DIRECTORIES.
-
add_executable(app
src1.cpp src2.cpp ...
)
link_directories(dir1;dir2;dir3)
target_link_libraries(app lib1 lib2 lib3)
Alan W. Irwin wrote:
I am running into a dependency issue with CMake 2.4.3 on Linux. I find
if I run cmake in an initially empty build directory followed by make, then
all
is well, i.e., everything builds without problems. But if I change nothing
in the source tree and run make again, the
Gheorghe Postelnicu wrote:
Below is the result of rpm -q and rpm -ql respectively.
[snip]
/usr/lib64/swig1.3/swig.swg
Okay, the FindSWIG.cmake script does not account for the possibility
that a non-library file is installed in lib64. Please submit a bug
report here:
http://www.cmake.org/Bug
Mike Jackson wrote:
I am trying to create a Top level Cmakelists.txt file to build my
project. My project is laid out kinda like vtk in that we have
subfolders with base libraries to be built in a specific order.
I have the first few lines in this top level cmake file to be:
SUBDIRS (
Thomas Hunger wrote:
Hi,
When I do this to SET_TARGET_PROPERTIES:
SET_TARGET_PROPERTIES(extree PROPERTIES
COMPILE_FLAGS \n)
make (not cmake) will complain:
CMakeFiles/XXX/build.make:61: *** missing separator. Stop.
This is not out of thin air. I made a
James Bigler wrote:
If I have a rule that generates some intermediate files, how do I add
those files to the list that get removed when I do a make clean?
I'm assuming you're talking about intermediate files when building a
custom command. If they are listed as the OUTPUT of the custom command
frederic heem wrote:
Problems began after upgrading cmake today on a fedora core 5.
I put before the upgrade some home made FindXXX.cmake in /usr/share/CMake.
The
problem is that the fedora and suse packager has changed /usr/share/CMake
to /usr/share/cmake, for consistency I suppose. The
Xavier Larrode wrote:
Hi all
i'm using cmake to compile lots of different libs, with ADD_SUBDIRECTORY...
And i've got a problem with a sub CMakeLists and linking my new lib
with other libs.
my CMakeLists :
[...]
SET(INSTALL_LIB_HOME ${GVTROOTDIR}/lib)
Xavier Larrode wrote:
Is it something (or something plan) to generate pkgconfig extension file
(.pc) with cmake, on the INSTALL process maybe?
CMake currently does not have enough information from CMakeLists.txt
files to do this automatically. There is nothing preventing you from
configuring
Scott Amort wrote:
I have come across some strange behaviour with testing and my CMake
files. My project is structured like this:
main project
-- src
-- test
-- subproject1
src
test
[snip]
The idea is that the main project builds all the subprojects first, then
itself; or, a
Stephen Adler wrote:
It looks to me like the
double quotes is canceling out glob'ness of *~.
[snip]
FILE(REMOVE
core
core*
*~
There is no globbing here. The clean code is a CMake script, not a
shell script. The make clean step is meant to clean files specifically
created by the build or
Michael Bell wrote:
Ok, it certainly makes sense that running cmake globally would
simplify things, which is a good thing. It does seem to cause a
problem for me that maybe you can help me work around:
Say I have a several subdirectories, each with C++ code and perl
wrappers to that code.
Brad King wrote:
A completely different approach would be to make sure that
myscript.pl.in has execute permission in the source tree.
...and a much simpler way is to not actually make them executable at all
but instead use FIND_PACKAGE(Perl) to get the perl executable and then
run the script
[EMAIL PROTECTED] wrote:
I try to write a CMakeList.txt file which create a DLL project under Visual
Studio .NET 2003 by using ADD_LIBRARY.
How do I make my project to generate both the .dll and the .lib ?
ADD_LIBRARY(mylib SHARED src1.cxx src2.cxx ...)
Make sure the source files use
Eduard Bloch wrote:
CONFIGURE_FILE is used (see below) to adopt an autoconf-style config
file, however the result is not cleaned up when the clean target is run.
It is not that bad as long as the generated file is not shipped and so
the generation is triggered on users machine, but when
Anders Sundman wrote:
Boost will produce the library files: libboost_date_time-gcc-mt-s.a or
boost_date_time-gcc.lib, or add some other nifty suffix. The
decorations specifying toolset, run-time, etc. that the library was
compiled with.
CMake has no formal knowledge of such decorations. This
[EMAIL PROTECTED] wrote:
Is there a way to get rid of the additional quot
Given
ADD_DEFINITIONS(-DSN_TARGET_PS3)
Then cmake autogenerates:
PreprocessorDefinitions=,quot;SN_TARGET_PS3quot;,
I just want:
PreprocessorDefinitions=SN_TARGET_PS3
Is there a template that I can
Tristan Carel wrote:
The module `UseSwig' provided in the version 2.4.3 provides a macro
named `SWIG_ADD_MODULE' to define a C/C++ library wrapped for one of
the targets supported by Swig. This macro uses the command
`ADD_CUSTOM_COMMAND' to define the rule as below:
wrapper.cxx: foo.i
swig
Yann Renard wrote:
I'm trying to use cmake under windows in order to generate VC8
makefiles. I have read
http://cmake.org/Wiki/CMake_Generator_Specific_Information chapter about
windows and written down a sript that looks like this :
set path=d:\yrenard\platform-sdk\bin;%path%
set
Doug Henry wrote:
No, since it is a dependency of a source file that is included in the
executable I would expect it to be handled (it was in previous
releases). It seems that cmake does know about it, because it gives an
error with the correct path to the file that needs to be generated. I
Yann Renard wrote:
Brad King wrote:
You need to run CMake from the same command prompt in which the
environment is set. From that same command prompt try:
mkdir myproj-build
cd myproj-build
cmake c:/path/to/myproj -GNMake Makefiles
Make sure you have a clean source tree in c:/path
Eric Noulard wrote:
Let's say I want to cross-compile my Windows application/library
on my Linux box using a cross-gcc (like mingw)?
Is it possible to indicate to CMake that we will use
a cross compiler?
In order to avoid (for example) to get a mylib.so file
if we want a mylib.dll.
You
Arjen Markus wrote:
Hello,
I have trouble getting CMake to recognise Compaq Visual Fortran
as a valid Fortran compiler under Windows.
Philippe Poilbarbe sent me his configuration module for CVF, but
he also noted that there is another problem - CVF does not like
the default options used
Marco Canini wrote:
Hi,
that's correct.
In this case a config file should be installed only if it doesn't exist
in order to preserve what the user wrote.
But I guess cmake doesn't support it yet.
This is what the INSTALL(CODE) or INSTALL(SCRIPT) commands are meant to
do: create custom
Brandon J. Van Every wrote:
Yann Renard wrote:
Cheers list,
when using cmake for building my libraries, I add additional commands
in order to copy all the corresponding header tree to a target
directory. On clean stage, I would like to remove this tree so removed
the .h files one by one
Joerg Mayer wrote:
cmake itself is 4-clause BSD license (including the advertising clause)
which is incompatible with GPLv2.
The CMake license should be free enough to allow you to copy and modify
the code and redistribute it under any license you choose as long as the
original notice appears.
Xavier Larrode wrote:
Hi all,
i have a regex like this :
SET(GTKMM_LIBDIR_EXTRACT_REGEX [-][l]([a-zA-Z0-9/._-]*))
And gtmm is using the lib sigc++-2.0
So i need to add to my regex the character + .
i tried with \+ ; \\+ '+' but nothing work ...
Make sure the - is still the last character
Luigi Calori wrote:
After several test and trials, I ve the suspect that if one use
ADD_CUSTOM_COMMANDS without DEPENDS, then the generator -Visual Studio
7 .NET 2003-, is not
running the command at build but requires a rebuild.
Am I completely wrong or is a known issue?
I happened to fix
Radu Mihai wrote:
Hello, first time on the list
I am working on a project that links with some external libs that we
build ourselves (OpenSceneGraph).
To accommodate the use of release and debug versions the current layout
looks like this:
osg/lib.rel/ .
osg/lib.dbg/ .
The
Toni Timonen wrote:
When cross-compiling windows dlls with mingw compiler using linux, the
dlls are not installed to the correct location (and the import library
is not installed at all). It seems that some things about the platform
are fixed at runtime.
The patch attached fixes this issue
Stephan Tolksdorf wrote:
If a subdirectory is included with EXCLUDE_FROM_ALL, targets in the
subdirectory are listed in the generated Visual Studio solution, but
targets recursively included in subdirectories of the subdirectory are
not. Is this a bug or a feature? (I'm using CMake from CVS
[EMAIL PROTECTED] wrote:
ADD_EXECUTABLE(foo foo.cxx)
TARGET_LINK_LIBRARIES(foo bar.lib)
and I got a -lbar on the link line from CMake 2.4.3 with the MSYS
Makefiles and MinGW Makefiles generators.
This example works for me, too. Maybe there's some issue on setting the
LIbrary via
Ryan Phillips wrote:
I am having a problem when I include INCLUDE_DIRECTORIES(${SOMEVAR})
within my project. The generated CFLAGS (under linux) include my
CMAKE_SOURCE_DIR appended with my ${SOMEVAR}, ie:
SOMEVAR=/usr/include/openssl/
upon compilation:
Brandon J. Van Every wrote:
Brad King wrote:
This would be a nice feature but I don't know how to implement it. Try
generating a project and then manually editing the project files in such
a way that the IDE selects a different configuration by default. Don't
forget to remove the .suo
Joerg Mayer wrote:
I'd like to follow development of cmake a bit more closely. In order to
do that, many other open source projects have mailing lists that follow
the cvs commit and the bug tracking system.
Is something like that available and just not mentioned on the mailing
lists page or,
Warren Turkal wrote:
Is there a way to override the mechanism for finding dependencies?
No.
I am trying to work around the crappy support for preprocessed Fortran
dependency generation. See bug #2361 [1].
Unfortunately we do not have time to solve this problem ourselves right
now. Your time
William A. Hoffman wrote:
At 03:20 PM 9/8/2006, Warren Turkal wrote:
On Friday 08 September 2006 13:11, you wrote:
#ifdef FOO
use abc
#else
use xyz
#endif
That's exactly the type of thing causing the problems.
Brad, how hard do you think it would be to run the fortran stuff
through cpp
Alexander Neundorf wrote:
Yes, this would be a good idea.
Actually, just login in to the Wiki and create a new wiki page with the
same contents.
It still needs to be canonized by Kitware, so that users aren't looking
at the wrong page.
As a test I ported the A simple example to the
Arjen Markus wrote:
add_custom_target(${CMAKE_CURRENT_BINARY_DIR}/${file} ALL
[snip]
When I run this in ..\build, using
cmake -G NMake makefiles ..\src
[snip]
CMake Error: Cannot open file for write:
Brad King wrote:
Peter Visser wrote:
I just discovered cmake, I find cmake easy to setup and it works
fantastic.
However, a small problem occurs. I'm using: cmake 2.4-patch3 under
MSYS-1.0.10.
The problem occurs when using EXEC_PROGRAM() or EXECUTE_PROCESS().
Instead of sh.exe from MSYS
Darby J Van Uitert wrote:
I am linking against a library that I have as a static and shared
library with the same name (foo.a and foo.so). When I build my project
with BUILD_SHARED_LIBS on, I want it to look for foo.so and when I build
static libs, I want it to link against foo.a. But it
Brad King wrote:
Darby J Van Uitert wrote:
I am linking against a library that I have as a static and shared
library with the same name (foo.a and foo.so). When I build my project
with BUILD_SHARED_LIBS on, I want it to look for foo.so and when I build
static libs, I want it to link against
Alan W. Irwin wrote:
So far all but one of those who responded to the discussion are in support
of the idea, but it is also fair to say that only a handful of subscribers
to this list have responded yet.
I am particularly interested in the opinion of Bill Hoffman, Brad King, and
other
Alexander Neundorf wrote:
I'd like to exclude some files from being scanned for dependencies, i.e. all
files which have ecos/install/include/ as part of their path.
I found CMAKE_C_INCLUDE_SCAN_REGEX, but I don't think I can use it for
excluding directories, or is there a way to do this ?
Kedzierski, Artur CIV NAVSURFWARCENDIV CORONA wrote:
Some packages provide modules/programs that
provide information where they were installed. For
example, FooLibrary may provide
FooLibrary.m4 for auto-tools,
FooLibrary.pc for pkg-confing,
or
Foo-config for anything else.
As
Kedzierski, Artur CIV NAVSURFWARCENDIV CORONA wrote:
I am sorry to bomb the mailing list with all my questions.
However, CMake is pretty exciting and my group is very likely to
migrate to it.
I found a link that mentions that CPACK can generate an RPM:
Philip Lowman wrote:
1.) [minor] After adding all of my header files (.h) and template files
(.t) to the ADD_LIBRARY command I noticed that the header files were
being placed in a separate folder called Header Files in the VC
Projects. Cool. Unfortunately, however, the template files were
Mike Melanson wrote:
I am working on a fairly large software project that I autotool'd some
time ago. The build system has been working reasonably well. However, I
have hit a possible limitation that I don't know how to solve with
autotools. So I wanted to know if CMake can solve this problem:
801 - 900 of 15210 matches
Mail list logo
