Actionscript to Obj-C / Cocoa: How to represent BitmapData? - objective-c

Question

I'm trying to port the as3delaunay library to Obj-C (for fun and to learn some more Obj-C). It's going pretty well, but I don't really understand how to convert uses of BitmapData to Cocoa. Here are several of the relevant parts of the aforementioned library:
In Edge.as:
internal function makeDelaunayLineBmp():BitmapData
{
var p0:Point = leftSite.coord;
var p1:Point = rightSite.coord;
GRAPHICS.clear();
// clear() resets line style back to undefined!
GRAPHICS.lineStyle(0, 0, 1.0, false, LineScaleMode.NONE, CapsStyle.NONE);
GRAPHICS.moveTo(p0.x, p0.y);
GRAPHICS.lineTo(p1.x, p1.y);
var w:int = int(Math.ceil(Math.max(p0.x, p1.x)));
if (w < 1)
{
w = 1;
}
var h:int = int(Math.ceil(Math.max(p0.y, p1.y)));
if (h < 1)
{
h = 1;
}
var bmp:BitmapData = new BitmapData(w, h, true, 0);
bmp.draw(LINESPRITE);
return bmp;
}
That is called in the following function, in selectNonIntersectingEdges.as:
internal function selectNonIntersectingEdges(keepOutMask:BitmapData, edgesToTest:Vector.<Edge>):Vector.<Edge>
{
if (keepOutMask == null)
{
return edgesToTest;
}
var zeroPoint:Point = new Point();
return edgesToTest.filter(myTest);
function myTest(edge:Edge, index:int, vector:Vector.<Edge>):Boolean
{
var delaunayLineBmp:BitmapData = edge.makeDelaunayLineBmp();
var notIntersecting:Boolean = !(keepOutMask.hitTest(zeroPoint, 1, delaunayLineBmp, zeroPoint, 1));
delaunayLineBmp.dispose();
return notIntersecting;
}
}
It also appears as a parameter in SiteList.as:
public function nearestSitePoint(proximityMap:BitmapData, x:Number, y:Number):Point
{
var index:uint = proximityMap.getPixel(x, y);
if (index > _sites.length - 1)
{
return null;
}
return _sites[index].coord;
}
What would be a good way to represent this behavior and/or the use of BitmapData in Obj-C / Cocoa?

Answer 1

The Core Graphics library has very powerful graphics capabilities -- even though it's all C based. You can create an 8-bit RGBA (red, green, blue, alpha) bitmap context like this:
size_t const bytesPerRow = 4 * width;
CGColorSpaceRef colorspace = CGColorSpaceCreateDeviceRGB();
CGBitmapInfo bitmapInfo = kCGImageAlphaPremultipliedLast;
CGContextRef ctx = CGBitmapContextCreate(NULL, width, height, 8, bytesPerRow, colorspace, bitmapInfo);
CFRelease(colorspace);
This corresponds to what the AS3 BitmapData has.
ctx now has a reference to the bitmap context which you eventually have to release with CGContextRelease(ctx).
You can manipulate the context, i.e. draw into it with the various CGContext* functions. If you eventually need to save it as an image (e.g. JPEG data), use CGBitmapContextCreateImage().

Answer 2

It looks like you are needing a CGBitmapContext to use in place of the BitmapData type.
Things to look into:
Quartz 2D Programming Guide read the "Graphics Context" section, especially the "Creating a Bitmap Graphics Context" part.
CGBitmapContext reference