On Monday 27 July 2015 17:59:24 Marc Mutz wrote: > On Monday 27 July 2015 15:12:03 Curtis Mitch wrote: > > Hmmm, I don't think that helps with > > https://codereview.qt-project.org/#/c/122268/, for example. > > > > Oh well. > > We could have syncqt.pl check for Q_DECLARE_TYPEINFO(T, X) for any T it > generates a header for, but that doesn't work for several reasons: > > 1. It doesn't catch nested types > 2. It doesn't work when the typeinfo is declared without the macro, e.g. > with QTypeInfoMerger. > 3. It doesn't catch private types (both in _p.h and .cpp)
I have a pending patch to syncqt that makes it scan _p.h. > 4. It doesn't catch enums (Q_RELOCATABLE_TYPE will be set for enums, but in > private API, we still want to manually set Q_PRIMITIVE_TYPE). > > Of course, one could use some GCC or Clang output to get a list of all > > I doubt there's a better way currently than manually marking all types. > That's why it's important to insisit on it in reviews. -- Thiago Macieira - thiago.macieira (AT) intel.com Software Architect - Intel Open Source Technology Center _______________________________________________ Development mailing list [email protected] http://lists.qt-project.org/mailman/listinfo/development
