Hi, (Since I'm not used to one this list, so this reply might not on the thread)
I'm a linux input method developer (fcitx) myself. The only problem is, libX11 provides xim implementation, while xcb doesn't, so actually xim for qt5 is a little bit hard to implement. But still there is some reason to not worry about that. XIM is a so problematic protocol and no one really touch it for years, in that sense nearly all input method provides its own protocol, and most of them already done the qt4 support if they have a custom protocol. This situatation will be only likely to change until wayland is coming. The reason is, since im module is so-widely-used (TM), it make not much sense to improve a more than 10 years ago protocol, and it would be hard to make it widely adoptted by both application and input method side. And the current situation is not bad. I've already done qt5 support for fcitx before qt5 release actually. https://github.com/fcitx/fcitx-qt5/ While uim have done the same: https://github.com/uim/uim/tree/master/qt5 And for some other widely used IMF, like hime, and gcin, it's not hard for them to port qt4 support to qt5. In my memory there is not much alive one that need to be counted in. BTW, I don't think put specific IMF support in Qt is correct way, since it would be easy to suffer from upstream's change. Regards,
signature.asc
Description: This is a digitally signed message part.
_______________________________________________ Development mailing list [email protected] http://lists.qt-project.org/mailman/listinfo/development
