2007/7/4, Juergen Lock <address@hidden>:
Ah, thanx, I didnt see that one! :)
I have a qeuestion tho: does this patch do deadkey handling on the host?
Yes.
IIRC I needed that because some VNC clients I tested didn't send dead
keys
to the server. They just waited for a full character to be available in
its buffer to send the KeyEvent message, as they assume the server will
know how to represent it. That is, if I press ´ and o in some VNC
clients, only an ó (0x00f3) message is sent after pressing the last
key, not the 0xfe51 / 0x006f combination. The server (qemu) has to know
how to build the character for its guest.
Because I didn't have to do anything like that for the de patch, the
guest could do that itself just fine. (tested with a linux livecd
guest with (under X) and without deadkeys (console).)