On Friday 12 May 2006 18:21, Alexander Neundorf wrote: > On Friday 12 May 2006 18:03, David Faure wrote: > > On Friday 12 May 2006 17:46, Alexander Neundorf wrote: > > > Different usage: > > > > > > set(mySrcs foo.cpp widget.cpp mainwindow.cpp) > > > kde4_automoc(${mySrcs}) <- detect automatically where to run moc > > > kde4_add_executable(kfoo ${mySrcs} ) > > > > > > as opposed to: > > > > > > set(mySrcs foo.cpp widget.cpp mainwindow.cpp) > > > qt4_wrap_cpp(mySrcs foo.h widget.h) <- list the files which need to be > > > moc'ed, no autodetection, moc_foo.cxx will be created and added to the > > > list mySrcs > > > > I don't understand the "cpp" in "qt4_wrap_cpp" then, if that function is > ...must be for historical reasons. :( Already historical reasons in the new build system.... How old is qt4_wrap_cpp? Can it be changed? See also below, not only the name seems wrong...
> > about moc'ing header files ;) > > about moc'ing any files. Doesn't seem to work though. I get a moc_kovariabletest.cxx [strange extension when qt+kde standardized on .cpp] which doesn't include any other file, so it doesn't compile. The way moc is called seems to be for the case where the .cpp includes the .moc (no #includes in the moc file), but the build system compiles moc_foo.cxx separately so it can't be #included (and it has the wrong name for that anyway). > > Anyway, in my case foo.moc would be preferred over moc_foo.cpp, when > > included by the .cpp file. > > With the qt4_wrap_cpp() the generated files wouldn't be included, but > compiled > as separate files. I think for unit tests this should be ok. it's ok but not perfect ;) -- David Faure, [EMAIL PROTECTED], sponsored by Trolltech to work on KDE, Konqueror (http://www.konqueror.org), and KOffice (http://www.koffice.org). _______________________________________________ Kde-buildsystem mailing list Kde-buildsystem@kde.org https://mail.kde.org/mailman/listinfo/kde-buildsystem