Rich, Take a look at Frame.c. The mouse code works much better now. It does have a problem with not moving far enough, but it at least moves in the right directions. Jonathan
Rich,
Take a look at Frame.c. The mouse code works much better now. It does have a problem with not moving far enough, but it at least moves in the right directions.
Take a look at TrackZoom() in mathtool.c (in the win directory). We really should be setting the position of the zoom box (or whatever) to the position of the cursor, instead of adjusting the position of the zoom box as we move the cursor. I think WinFract only moves the zoom box when the left mouse button is pressed while the cursor is inside the zoom box. Jonathan
Rich, Do you recall what is driving the size of KEYBUFMAX? It is defined in two places (frame.h and wintext.h) as 80. The way the mouse is currently set up, we look for a delta of 8 pixels, and if we have it, we then move the cursor 1 pixel. This accounts for the slow movement when using the mouse. When I change the code to move the cursor the correct number of pixels (8 or greater keystrokes), then the buffer overflows (the assertion in frame_add_key_press() is false, which dumps us out). Jonathan
participants (1)
-
Jonathan Osuch