ripoffline() behavior with resizable windows

[Bill Gray]

   I've been investigating how ripped-off lines work (or don't work) in 
PDCurses and PDCursesMod,  and comparing with how ncurses does it.  I'm 
at a point where I could probably use some advice.

   In ncurses and PDCurses,  ripped-off lines (at top or bottom) are 
drawn once,  using the callback function supplied to ripoffline().  If 
the user resizes the screen,  ncurses does "the right thing" and moves 
the lines ripped off at bottom so that they stay at the bottom.  You can 
expand/contract the screen vertically and never have a problem.

   Neither library ever calls the callback function again,  though.  So 
if the line was,  say,  80 columns at startup,  it won't get 
extended/contracted as the screen expands/contracts.  Here's the test 
program I've used :

https://www.projectpluto.com/temp/ripoff.c

   Compile and run as,  for example,

./ripoff b b t

   ("rip off two lines at the bottom,  then one at the top"),  resize 
horizontally,  and you'll see what I mean.  In particular,  if you 
shrink the window horizontally and re-expand,  you lose the text in the 
columns you shrank over.

   The SLKs (soft-label keys) _do_ get redrawn correctly under such 
resizing,  despite SLK lines being (sort of) a subset of ripped-off 
lines in ncurses.  (They are entirely separate beasts in PDCurses or 
PDCursesMod... something I intend to fix in the latter.)

   It seems to me as if the callback function should be called (at 
least) in situations where COLS changes : in initscr( ) and whenever the 
screen is resized horizontally.  Does this make sense to you?

-- Bill