Hi Julian, indeed it does work for simple examples. I had an off-list discussion regarding this thread with Brett Lymm. He asked me to prepare a minimum example which shows the issue. I was not able to produce it, i.e. the minimum example did work. In the real application where it fails it is not possible to track down the issue to the line number which causes it. That means if I simplify the application more and more it works at some point. But there is not a special source code line which makes it fail, there are many lines which's removal makes it work or not.
Initially I forgot to say that it works with newwin(), but I use subwin(). After Brett's hints it does now also work when I don't use start_color(), the full application only fails in color mode now. Brett assumes a refresh() order issue, but this is not very likely. I have two non-overlapping subwins in stdscr which cover stdscr completely (1st: line 0 ... LINES-3, 2nd: LINES-2 ... LINES-1, both COLS wide). I don't use stdscr (except for getch()), every wprint, wearse, wrefresh, wmove etc. got to the subwins. I want to do further tests with new hints I got from Brett. If it doesn't work I'll try again to prepare a minimum example from scratch. Thank you! --Carsten (The real application which proves that it does not work under some circumstances (could be a application bug of course) is on https://github.com/n-t-roff/vddiff, but this is of course not suited to debug the issue.)
