Re: Bottom ripped-off line content lost on screen resize

[Thomas Dickey]

On Fri, 28 Dec 2007, Katarina Machalkova wrote:

> On Monday 24 December 2007 17:07:21 Thomas Dickey wrote:
> > On Thu, Dec 20, 2007 at 10:59:58AM +0100, Katarina Machalkova wrote:
> > > Any hint what's missing here (some screen redraw call, or,...)?
> >
> > It looks like a bug (in ncurses).  I've modified test/ncurses.c to look
> > into this detail, can see that the first refresh _after_ getting a
> > character after the KEY_RESIZE is painting the soft-keys in the expected
> > place.
>
> If it's really a bug, do you need any further input (a testcase etc.) from me?

no - I have enough for a first cut from the modified test/ncurses.c

> Any clue in which release it is going to be fixed?

I'll probably work on that tomorrow - since it's a bug.
(Otherwise I'd be chipping away at the threading code).

> Btw, shortly I had posted my original request for help, I added some logging
> and sleep()s into my app and found a workaround (as usually ;-) ).
> I realised that the window associated with the bottom ripped-off line is
> always (even after screen resize) painted at the original position. So in
> screen resize handler, I have to explicitly move it to the downmost position
> (using mvwin() routine) and call doupdate() after that.

yes - but since it's a window maintained by ncurses, ncurses ought to
move it.  Other windows whose position you've set would have to be
moved by your application - ncurses can only make a reasonably good
guess how to handle those.

--
Thomas E. Dickey
http://invisible-island.net
ftp://invisible-island.net