NSRect vs. CoreGraphics

I have a general question concerning drawing pretty simple shapes in an NSView. Is there a "rule" when to use the NSRect class struct (drawrect) versus using the CoreGraphics framework? What I understood so far I could use either one for my osx application. Can I?

First off, there is no NSRect class (it's a struct), you're referring to the NSView method drawRect:. Generally speaking, the Core Graphics framework is "lower-level" than the AppKit drawing methods, which are usually just "wrappers" around Core Graphics. It's also plain C instead of Objective-C, so AppKit is typically a bit easier to use, but you can mix and match freely between the two.

For example:

- (void)drawRect:(NSRect)rect
{
    //This uses AppKit:
    [[NSColor redColor] set];
    [[NSBezierPath bezierPathWithOvalInRect:[self bounds]] fill];

    //This does the same, using Core Graphics:
    CGContextRef ctx = [[NSGraphicsContext currentContext] graphicsPort];
    CGContextSetRGBFillColor(ctx, 1.0, 0.0, 0.0, 1.0);
    CGContextFillEllipseInRect(ctx, NSRectToCGRect([self bounds]));
}