I'm trying to understand how things work in regards to concurrent programming and calling setNeedsDisplay. I basically have Three objects.
Main View - container with different UIView objects, the main one being a UIScrollView
Small Map View - a small UIView that draws a miniature version of one of the other UIView items on screem
Processor - a delegate of the Main View that calculates what's on screen and calls the Main View back with what's in view.
So a simple use case of what's going on is the user touches the ScrollView and then the processor updates what's in view of the scrollView (like calculating coordinates, center point, etc) It does this using blocks and does it asynchronously. This then posts a notification to the MainView object.
When the MainView receives the notification, it just calls
[smallMap setNeedsDisplay]; // example 1
I put some logs around this call, and I do see it gets called right away. However, the drawRect: of this function does not get called right away. It gets called after 2 seconds or so.
I remember reading that setNeedsDisplay
just marks the view for redraw to happen on the next event of the run loop.
But if I add this code instead:
// example 2
dispatch_async(dispatch_get_main_queue(), ^{
[smallMap setNeedsDisplay];
});
My view gets redrawn right away.
I guess I'm confused as to why I have to ask for the main event loop to call setNeedsDisplay to immediately redraw something. Like in example 1, by me calling setNeedsDisplay, is that done in the background or something and that's why it doesn't get redrawn right away? I'm trying to understand the difference in what's going on behind the scenes so I know what to look for the in future. Like should I have all my calls that need to be immediately redrawn in something similar to the example 2 block? Or is it because I'm processing my data asynchronously that I need to then ask for the main queue? Thanks!