On Fri, Mar 23, 2007 at 12:28:20PM +0100, Enrico Forestieri wrote: > On Fri, Mar 23, 2007 at 08:51:57AM +0100, Peter Kümmel wrote: > > Andre Poenitz wrote: > > > On Fri, Mar 23, 2007 at 12:06:27AM +0100, Peter Kümmel wrote: > > >> Merging qt4 on linux does not work, because xlib.h defines some symbols > > > > > > There is no xlib.h on my system. > > > > > >> and gcc couldn't resolve symbols, which looks like a gcc bug. > > > > > > I know you won't believe me, but gcc bugs are less rare than > > > you seem to believe. > > > > I also think it's not really a bug but a it's > > very mysterious compiler error: > > #include <Qt> > > which defines the enum Qt::CursorShape > > but then gcc complains about using CursorShape > > viewFullSourceCB->setCursor(QCursor(static_cast<Qt::CursorShape>(0))); > > > > Maybe Qt::CursorShape could not be used as type? I also had this > > behavior when adding the enum for the windows-maximized flag: > > gcc doesn't like the enum as type only using int was the > > solution. > > This must be a clash occurring with the X.h include file which defines > CursorShape as follows: > > #define CursorShape 0 /* largest size that can be displayed > */ > > Maybe this can be hacked using some preprocessor magic, too: > > #define CursorShape XCursorShape > <X includes here> > #undef CursorShape > > BTW, I think that Xlib.h must be present on Andre's system...
Urm. yes. But not xlib.h... Obviously also my own parser is case sensitive. Xlib.h certainly would have triggered something... Andre'
