> DOS itself doesn't use/support any mouse, it is up to an application to > interpret the responses given through the INT 33h API, which is > implemented either by the BIOS or a "driver" (TSR_ like CTMouse. While > probably not completely impossible to add another TSR that would > intercept INT33h and instead feeds simulated keyboard events into INT > 9h. I am not sure that something like this exist, at least neither > myself nor any of my friends/clients ever had a need for such a tool...
Not quite so helpful but the old XT I had many years ago (an Amstrad PC1640) had the mouse connected to the keyboard rather than a serial port as was usual for the time, and I believe it worked by sending unused key codes in response to mouse actions. The mouse driver would pick up these key codes and handle the int 33h stuff. However because of this, the machine could be configured at the hardware level to send different key codes for the mouse buttons. So running a DOS application to change the machine's CMOS settings was all it took to reassign the mouse buttons to send any keystroke you wanted, and the setting would persist across power cycles without needing any TSR running. I wonder whether the author of the original application being discussed here had a similar machine, and assumed all PCs could have their mouse buttons configured similarly. Cheers, Adam. _______________________________________________ Freedos-user mailing list Freedos-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/freedos-user
