Re: delwin() and deleting windows with subwindows

[Bill Gray]

On 8/11/22 17:53, G. Branden Robinson wrote:
> At 2022-08-11T12:35:49-0400, Bill Gray wrote:
> >     Delete subwindows recursively,  and that access attempts to read
> >     freed memory.  One shouldn't be doing such things;  once a parent
> >     window is deleted,  you shouldn't try to do anything with the
> >     subwindow.  But I'll bet you that some code does so,  and recursive
> >     deletion would break that code.
>
> Why not implement recursive deletion for now, and see if anything
> squawks?  Any application that chases pointers into freed memory is
> surely busted at a fundamental level, quite apart from violating the
> documented assumptions of the XSI Curses spec.

   I was tempted to keep recursive deletion (the code was all written 
and tested),  but I've chickened out.

   The problem is that with any of the current approaches,  you quite 
definitely _wouldn't_ be chasing pointers into freed memory,  because 
the subwindow wouldn't actually be freed;  you'd either get an error 
code,  or only the parent window would be freed.  The subwindow would 
remain as allocated memory and accesses to it would be safe.  And as 
Thomas points out,  exactly what the spec's "documented assumptions" are 
is a little ambiguous;  I could be persuaded either way on that point.

   Right now,  ncurses,  SVr4 curses,  and PDCursesMod will just give 
you an error code if you delwin() a parent window... which means you may 
fail to free up memory you meant to free up,  but I don't see any other 
downside to it.  If you later call delscreen(),  all windows and 
subwindows associated with that screen will be freed anyway,  with both 
ncurses and PDCursesMod.

-- Bill

> Regards,
> Branden