Bill Gray

Just did a trawl through the code to see how signal handling is currently done. In both PDCurses and PDCursesMod : - `ozdemo` sets up a handler to call `endwin()`...

(Following applies to PDCurses and PDCursesMod). Your first bullet point doesn't apply to WinCon, because it already only reads `PDC_RESTORE_SCREEN` at startup. It does apply to both OS/2 and DOS....

> That shouldn't be done for multitude of reasons, one is that a `refresh()` should get the last state back and you'd want to be able to `endwin()` again and...

@GitMensch - this should be fixed in the PDCursesMod fork now. Give it a try and see what you think. Solution was in two (fairly easy) parts, which could also...

Looks as if `idx` lies outside the actual range of the `kbtab[]` or `ext_kbtab[]` arrays. In other words, it's a key that should be ignored, but isn't. I don't currently...

It _is_ related inasmuch as @zingchen is getting `vk=0xe7`, and when the code looks at `kptab[0xe7]`, it is (at the very least) an invalid access; `kptab[]` only contains 226 =...

At some point, it may be possible to use VT/xterm control sequences for mouse (and keyboard) input. Those sequences return a stream of presses and releases, leaving it up to...

Thank you. I've always assumed that `PDC_MOUSE_MOVED` and `MOUSE_MOVED` were reserved for reporting button-less mouse movements. That's the way it works in ncurses (on the platforms that support such movements...

Of the six static `save_*` variables in `pad.c`, the four "minimum" values are never anything but zero. The maximum column/row are set to the then-current minimums of LINES/COLS and nlines/ncols,...

Hi @GiorgosXou - I am still a little baffled here. I thought maybe those six static variables really were used "properly" at some time in the past, and maybe they...