Re: Recommended way to distinguish between pads and windows?

[Thomas Dickey]

On Sun, 6 Sep 2009, Mark Dickinson wrote:

> Thomas Dickey-2 wrote:
> > On Sun, 6 Sep 2009, Mark Dickinson wrote:
> > > but with ncurses 5.7, the internals of WINDOW are hidden, so the
>
> > hmm - that's actually a configure option (not "always" hidden, but can
> > be made so).  Which package/distribution is turning this on?
>
> Aha.  I forgot to include system information:  apologies.  This is
> the 'official' Snow Leopard version of ncurses, from a clean install of
> OS X 10.6 on a MacBook Pro.  The /usr/include/ncurses.h
> file includes the lines:
>
> #ifndef NCURSES_OPAQUE
> #define NCURSES_OPAQUE 1
> #endif
>
> and now that you mention it, I see that the macports version of
> ncurses has a '#define NCURSES_OPAQUE 0' here instead.  I wonder
> why Apple decided to do this.

I don't know - I get occasional feedback from people who are maintaining
ports, but don't recall that any of those are Apple developers.

> > I didn't think of adding an is_pad(), but for other features, see
> > the ncurses "opaque" manpage, e.g.,
> >
> >         bool is_cleared(const WINDOW *win);
> > ...
> >         WINDOW * wgetparent(const WINDOW *win);
> >
> > Not exactly what you want, but a pad wouldn't have a parent window.
> > (subwin's and derwin's would have one, but not plain newwin's).
>
> Thank you.  That's very helpful.

no problem.  I'll add an is_pad() and is_subwin() for later use.
(the other flags really seem too low-level to be visible)

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