Hi Seth,
I had the same problem some time ago. In my case, the environment
variable LIBRARY_PATH contained the boost directory, which instructs the
linker to use it as standard library directory.
CMake asks the compiler for implicit linking directories. I am not
completly sure, but the directories should be documented in some
CMakeFiles/*Output* file.
I hope, that helps a little.
Andreas
On 23.09.2013 10:01, S'orlok Reaves wrote:
Good afternoon all,
This is my first post to this list, so please redirect me if I'm in the wrong
place. So far cmake has been great, but I've run into an issue that I am having
trouble understanding.
I'm searching for the Boost libs on a server that has two different versions
installed. One is in my home directory; the other is a system-wide install
(which is older).
In my CMakeLists.txt:
set(Boost_NO_SYSTEM_PATHS ON)
set(BOOST_ROOT "/home/me/opt")
find_package(Boost COMPONENTS system REQUIRED)
MESSAGE("${Boost_LIBRARIES}")
#later:
target_link_libraries (My_Program ${Boost_LIBRARIES})
This prints out the following, so I know the correct library is indeed found:
/home/me/opt/lib/libboost_system.so
Now, I "make" the project, and everything compiles, but fails to link,
specifying several missing symbols in boost::system. The linker command includes the
following:
#g++ ....etc., then:
-o ../My_Program -rdynamic /usr/local/lib64/libsoci_core.so -lboost_system
-Wl,-rpath,/usr/local/lib64
If I change it to the following, it compiles, links, and runs fine:
-o ../My_Program -rdynamic /usr/local/lib64/libsoci_core.so
/home/me/opt/lib/libboost_system.so -Wl,-rpath,/usr/local/lib64
Clearly the "-lboost_system" flag is catching the old libraries. From reading the cmake
docs, I was under the impression that the full path (/home/me/opt/lib/libboost_system.so) would
only be truncated to an "-lboost_system" if it was on the implicit path (e.g., /usr/lib).
In that case, I have two questions:
1) Why is boost_system not being included via its complete path? Is this a bug,
or am I doing something wrong?
2) Why is soci_core being included by the full path. Is it because "/usr/local"
is not on the implicit path?
The second question is just out of curiosity; it is the first question that is
causing me a lot of grief right now. Any suggestions?
The server I'm compiling on is running Debian Squeeze (6.0.7), with a
locally-installed version of cmake (2.8.11.2).
Thanks,
Seth
--
Powered by www.kitware.com
Please keep messages on-topic and check the CMake FAQ at:
http://www.cmake.org/Wiki/CMake_FAQ
Kitware offers various services to support the CMake community. For more
information on each offering, please visit:
CMake Support: http://cmake.org/cmake/help/support.html
CMake Consulting: http://cmake.org/cmake/help/consulting.html
CMake Training Courses: http://cmake.org/cmake/help/training.html
Visit other Kitware open-source projects at
http://www.kitware.com/opensource/opensource.html
Follow this link to subscribe/unsubscribe:
http://www.cmake.org/mailman/listinfo/cmake
--
Powered by www.kitware.com
Please keep messages on-topic and check the CMake FAQ at:
http://www.cmake.org/Wiki/CMake_FAQ
Kitware offers various services to support the CMake community. For more
information on each offering, please visit:
CMake Support: http://cmake.org/cmake/help/support.html
CMake Consulting: http://cmake.org/cmake/help/consulting.html
CMake Training Courses: http://cmake.org/cmake/help/training.html
Visit other Kitware open-source projects at
http://www.kitware.com/opensource/opensource.html
Follow this link to subscribe/unsubscribe:
http://www.cmake.org/mailman/listinfo/cmake
