[surface release];
surface = nil;
+
+ [self setNeedsDisplay:YES];
}
#endif
[self initWithFrame: r];
[self setAutoresizingMask: NSViewWidthSizable | NSViewHeightSizable];
+#ifdef NS_DRAW_TO_BUFFER
+ /* These settings mean AppKit will retain the contents of the frame
+ on resize. Unfortunately it also means the frame will not be
+ automatically marked for display, but we can do that ourselves in
+ viewDidResize. */
+ [self setLayerContentsRedrawPolicy:
+ NSViewLayerContentsRedrawOnSetNeedsDisplay];
+ [self setLayerContentsPlacement:NSViewLayerContentsPlacementTopLeft];
+#endif
+
FRAME_NS_VIEW (f) = self;
emacsframe = f;
#ifdef NS_IMPL_COCOA
}
+#ifdef NS_IMPL_COCOA
+/* If the frame has been garbaged but the toolkit wants to draw, for
+ example when resizing the frame, we end up with a blank screen.
+ Sometimes this results in an unpleasant flicker, so try to
+ redisplay before drawing. */
+- (void)viewWillDraw
+{
+ if (FRAME_GARBAGED_P (emacsframe)
+ && !redisplaying_p)
+ {
+ /* If there is IO going on when redisplay is run here Emacs
+ crashes. I think it's because this code will always be run
+ within the run loop and for whatever reason processing input
+ is dangerous. This technique was stolen wholesale from
+ nsmenu.m and seems to work. */
+ bool owfi = waiting_for_input;
+ waiting_for_input = 0;
+ block_input ();
+
+ redisplay ();
+
+ unblock_input ();
+ waiting_for_input = owfi;
+ }
+}
+#endif
+
+
#ifdef NS_DRAW_TO_BUFFER
- (BOOL)wantsUpdateLayer
{
{
NSTRACE ("[EmacsView updateLayer]");
+ /* We run redisplay on frames that are garbaged, but marked for
+ display, before updateLayer is called so if the frame is still
+ garbaged that means the last redisplay must have refused to
+ update the frame. */
+ if (FRAME_GARBAGED_P (emacsframe))
+ return;
+
/* This can fail to update the screen if the same surface is
provided twice in a row, even if its contents have changed.
There's a private method, -[CALayer setContentsChanged], that we