Re: [Cs-dev] fltk on Windows....
| Date | 2014-10-29 14:23 |
| From | Steven Yi |
| Subject | Re: [Cs-dev] fltk on Windows.... |
| Attachments | None None |
| For libraries and headers, I think it's preferable to install them into a location that CMake will find it, rather than setting them yourself in the cache. Normally that means installing the library after building to /usr/local. However, you can add additional paths for CMake to search. If you look at the Cusom.cmake.ex file, you'll see things like: if(WIN32) list(APPEND CMAKE_SYSTEM_INCLUDE_PATH "c:/work/libsndfile-1_0_17") list(APPEND CMAKE_SYSTEM_LIBRARY_PATH "c:/work/libsndfile-1_0_17") list(APPEND CMAKE_SYSTEM_LIBRARY_PATH "c:/Python25/libs") list(APPEND CMAKE_SYSTEM_INCLUDE_PATH "c:/Python25/include") endif() If you copy that to Custom.cmake, then you can add your own paths. So a couple of options here: 1. You can still try doing a "make install". That probably only works if you're within an msys shell. (I'm not certain what the convention is for the standard cmd prompt if you use MinGW's make.) 2. You can create a folder where you'll manually add all of your libs and headers. Then you can add that folder to your paths. This is handy if you want to isolate the dependencies you build from the rest of your system. For the mingw cross-compile scripts for compiling the windows binaries from Linux, that was the approach I took with the dependency builder scripts. 3. You can individually add each folder of each dependency you build to the CMake paths. Hope that helps! steven On Wed Oct 29 2014 at 10:16:26 AM Rory Walsh <rorywalsh@ear.ie> wrote:
Yes, my point is I have no idea what else I need to tell cmake so that |
| Date | 2014-10-29 14:27 |
| From | Michael Gogins |
| Subject | Re: [Cs-dev] fltk on Windows.... |
| Attachments | None None |
The use of Custom.cmake is never explained, therefore I have not been able to use it. An example of Custom.cmake or explanation of how to use it would be much appreciated. As you suggest, I have increasingly relied on installing to /usr/local. But there is not always a working install target or script on Windows, so this is a bit flaky. Regards, Mike ----------------------------------------------------- Michael GoginsIrreducible Productions http://michaelgogins.tumblr.com Michael dot Gogins at gmail dot com On Wed, Oct 29, 2014 at 10:23 AM, Steven Yi <stevenyi@gmail.com> wrote: For libraries and headers, I think it's preferable to install them into a location that CMake will find it, rather than setting them yourself in the cache. Normally that means installing the library after building to /usr/local. However, you can add additional paths for CMake to search. If you look at the Cusom.cmake.ex file, you'll see things like: |