finkel@TAURUS.BITNET (10/06/88)
I'm writing a program that uses console.device, and I also want to read mouse
input . I wanted to use console.device to read mouse clicks and position,
but I could get only the left button press, and not the location. Carolyn
Sheppner ( did I spell correctly ) advised me to use IDCMP for the mouse
input, and that's what I did.
Now, my program uses the RKM I/O routines on the console.device chapter.
In the main loop, I do something like that:
Status=Wait(1L<<(IDCMP_SIGNAL_NUMBER)||1L<<(IO_SIGNAL_NUMBER));
if (Status&(IDCMP_MASK))
{
}
if (Status&(IO_MASK))
{
/* process characters here, use ConGetChat() as in RKM,
i.e., use the GetMsg and WaitIO ( or was it WaitPort ? ),
since we use asynchronous I/O.
*/
}
The problem is, that I get IO signals from the IO port when there are no
characters to read from the console.device, thus my program waits again
in the ConGetChar() routine. The net effect is that keyboard input to
my program works fine, but the mouse input works well only after I press
the <return> key.
I am sure that the problem is that I'm getting false signals ( thanks to
SDB! ), I just don't know why.
If I will not solve this, I'll return to my original method, ( i.e. read
mouse events using console.device and get mouse position from
Window->MouseX and Window->MouseY, although I don't like it ).
Any help?
-------------------------------------------------------------------------------
Udi Finkelstein | Bitnet: finkel@taurus.bitnet or finkel@math.tau.ac.il
Tel Aviv University | Internet: finkel%taurus.bitnet@cunyvm.cuny.edu
Israel | UUCP: ...!psuvax1!taurus.bitnet!finkel
-------------------------------------------------------------------------------