drawMapRect context + Grand Central Dispatch - ios

Question

I have this function:
- (void)drawMapRect:(MKMapRect)mapRect zoomScale:(MKZoomScale)zoomScale inContext:(CGContextRef)context
{
TileOverlay *tileOverlay = (TileOverlay *)self.overlay;
NSArray *tilesInRect = [tileOverlay tilesInMapRect:mapRect zoomScale:zoomScale];
CGContextSetAlpha(context, tileAlpha);
for (ImageTile *tile in tilesInRect)
{
__block UIImage * image;
CGRect rect = [self rectForMapRect:tile.frame];
NSString *path = [[NSString alloc] initWithFormat:#".../%#.png", tile.imagePath];
NSLog(#"Loading tile from URL %#", path);
image =[UIImage imageWithData: [NSData dataWithContentsOfURL:[NSURL URLWithString: path]]];
CGContextSaveGState(context);
CGContextTranslateCTM(context, CGRectGetMinX(rect), CGRectGetMinY(rect));
CGContextScaleCTM(context, 1/zoomScale, 1/zoomScale);
CGContextTranslateCTM(context, 0, image.size.height);
CGContextScaleCTM(context, 1, -1);
CGContextDrawImage(context, CGRectMake(0, 0, image.size.width, image.size.height), [image CGImage]);
CGContextRestoreGState(context);
}
}
As you know dataWithContentsOfURL blocking thread until in will be done. I want to add image loading block to the GCD section.
I tried to do it like this:
- (void)drawMapRect:(MKMapRect)mapRect zoomScale:(MKZoomScale)zoomScale inContext:(CGContextRef)context
{
TileOverlay *tileOverlay = (TileOverlay *)self.overlay;
NSArray *tilesInRect = [tileOverlay tilesInMapRect:mapRect zoomScale:zoomScale];
CGContextSetAlpha(context, tileAlpha);
for (ImageTile *tile in tilesInRect)
{
__block UIImage * image;
CGRect rect = [self rectForMapRect:tile.frame];
dispatch_async(dispatch_get_global_queue(DISPATCH_QUEUE_PRIORITY_LOW,0), ^{
NSString *path = [[NSString alloc] initWithFormat:#".../%#.png", tile.imagePath];
NSLog(#"Loading tile from URL %#", path);
image =[UIImage imageWithData: [NSData dataWithContentsOfURL:[NSURL URLWithString: path]]];
CGContextSaveGState(context);
CGContextTranslateCTM(context, CGRectGetMinX(rect), CGRectGetMinY(rect));
CGContextScaleCTM(context, 1/zoomScale, 1/zoomScale);
CGContextTranslateCTM(context, 0, image.size.height);
CGContextScaleCTM(context, 1, -1);
CGContextDrawImage(context, CGRectMake(0, 0, image.size.width, image.size.height), [image CGImage]);
CGContextRestoreGState(context);
});
}
}
But I got context error. Please help me with this stuff.
How to use context operations in GCD block?

Answer 1

My first note is that MKOverlayView is depreciated. You should consider switching to MKOverlayRenderer.
There is no situation where you should use GCD from within a -draw__ method. That includes MKOverlayView -drawMapRect:zoomScale:inContext: as well UIView -drawRect:. Instead, you should be using an NSOperationQueue, in conjunction with -canDrawMapRect:zoomScale:zoomScale and setNeedsDisplayInMapRect:.
Here's some sudo-code for it:
- (BOOL)canDrawMapRect:(MKMapRect)mapRect zoomScale:(MKZoomScale)zoomScale
{
BOOL hasAtLeastOneTile = NO;
TileOverlay *tileOverlay = (TileOverlay *)self.overlay;
NSArray *tilesInRect = [tileOverlay tilesInMapRect:mapRect zoomScale:zoomScale];
for (ImageTile *tile in tilesInRect) {
if ([tile isAvailable]) {
hasAtLeastOneTile = hasAtLeastOneTile || YES;
} else {
// Add operation to NSOperationQueue to fetch tile
__weak MKOverlayView *weakOverlay = self; // Weak ref to prevent retain cycles
NSOperation *op = [NSBlockOperation blockOperationWithBlock: ^{
//TODO: Load Tile
[weakOverlay setNeedsDisplayInMapRect:mapRect];
}];
[self.operationQueue addOperation:op];
}
}
return hasAtLeastOneTile;
}
Then in your -drawMapRect:zoomScale:inContext: you draw what tiles you have available and skip the ones that are not.