[ Thread Index |
Date Index
| More lists.liballeg.org/allegro-developers Archives
]
On Monday 20 December 2004 02:38, Jon Rafkind wrote:
> set_gfx_mode so the user has to explicitly enable window resizing. In
> this case, the user should be aware that they have to use acquire_screen
> and release_screen around the usage of screen.
So, the screen *needs* to be acquired before drawing onto it if the screen
is allowed to be resized? In other words, does acquiring the screen
disallow the resizing temporarily?
I suppose this is ok, as the resizing is a new feature and requiring the
screen be locked for it is ok - even if there are situations where you
don't want the screen locked. What would happen if the graphics driver
were to reverse the lock, as the Windows DirectX driver does for hardware
acceleration? I know this patch only addresses X11, but it's something to
think about anyway if we want to have the same behavior in Windows.
Oh, last but not least, what happens to videobitmaps and subbitmaps of
screen when the window is resized? It would be quite bad if they became
invalid without any way of informing the user of this, especially if this
were to happen during a drawing opertion (though in that case locking the
screen should prevent this from happening).
I don't suppose it would be possible to do this, but ideally, I would
disallow the resizing unless the user-defined callback says it's ok to
resize at this moment.
Evert