When I submitted an update to my app yesterday, I got a rejection saying I was using a private API called `addAction:` (apple app ID 643351983). A thorough scrub of my code shows only one place any `addAction:` is defined / called in my (or third party) code: an autogenerated method by Core Data for one of my NSManagedObject subclasses, which has a relationship called "Actions".
Clearly this "private API" search is text-based.
This generated code has been in my app since 2013, largely untouched. App review approved my app, but requested I change the method name in order to prevent future flagging. Unfortunantly this is part of my data model layer and I really don't want to change it. It's not a trivial change as I do a lot of serialization and backing up of data base on the names of my entity properties. I'd be happy to change if I was using a private API, but I'm not, and App Review simply wants me to do all this extra work this to "avoid future confusion."
Something changed with the private API checks around when iOS 10 came out to get this method flagged, and I think the automated process is a little overzelous in what it flags now. Asking developers to take on the extra work, to esentially work around false positives, seems like a bad practice.