Thanks Rob for your msg about I need info, on 19 09-19-1994

You could basically just write a small set of function that
access a 4K buffer in memory (an 80x25 buffer with char/attr
pairs) as thow it were a text mode buffer. Then you would just
need to provide a refresh function that would be called from the
WM_PAINT message. It would calculate the cells covered by the
invalidated area, then translate the information in the pseudo
buffer and convert them to very simple GPI calls.

Of course, the functions that you write to manipulate the video
buffer would also need to write to the presentation space as you
made modifications, otherwise output would not occur until the
window was invalidated. A semaphore could be used to control those
times when a write to the buffer and an invalidation event occur
simultaneously.

All of this is pretty academic anyway since you can get the same
thing much easier by the VIO API, but it is kind of fun sometimes
to think about how you do something like that.

A better option would be to just go PM althogether and get it
over with. The VIO interface is dragged along with each version by
greater and greater effort. It will be dropped before too much
longer, although I think now that they have been badgered into
providing it in Workplace OS/2.

___
 X KWQ/2 1.2b X If at first you don't succeed, put it out for beta test (KWQ Beta

--- Maximus/2 2.01wb