Dave Troy

No the only solution I was able to find was to run under x86 Java and Rosetta2, which I'm trying to avoid if possible. Please post here if you do...

Hm, OK. :) Digging further into your code, I guess what I was missing here was the `modificationComparatorBlock` option. It seems `isEqual:` is insufficient to detect Core Data property changes....

Thanks for the input. The best guess I have right now is that `isEqual:` just doesn't detect the property changes. So, it's deciding it's the same object (which it is),...

Hm, yeah. That's sort of what I figured initially. If that's the case, I can only guess there is something wrong with the plumbing in how `NSFetchedResultsController` is supplying into...

I just did a quick look to see if `willUpdateDataModel` is being called more than once. So far that does not look to be the case, so it seems more...

OK, attached is a sample project that exactly illustrates the issue I am having. I've included tests against `NSFetchedResultsController` which all work fine, and tests against `TLIndexPathController`, which fail. You...

Ah, ok, thanks. I will look into this. Yeah, I do need to sort on `messageCount`. So, to me the current behavior is a bug and it seems this fix...

Yeah. I think that would be necessary for anyone wanting to use TLIPT with a sort order specified, and properly report such zero-distance moves. There doesn't seem to be any...

Thanks. I am using my hack in the meantime and it seems to be pretty solid.

One other note, in case it is relevant. I am seeing issues where when a predicate is set on the `TLIndexPathController` fetchRequest, sometimes `.insertedItems` is not populated with an entity...