#1282 creating objects in a subpatch invalidates pointers to scalars

[Anonymous]

Hi, I'm aware that deleting a scalar from subpatch automatically invalidates all pointers to scalars residing in this subpatch (as an easy way to prevent stale pointers).
Is it desired behaviour that deleting an object also invalidates all pointers? Usually, deleting elements from a linked list shouldn't invalidate iterators to remaining elements. and since the deleted element is an object, there is no way a [pointer] could have pointed to it, so there's no need for preventing stale pointers...

Furthermore, creating an object (e.g [+]) will also invalidate all pointers. I wonder why?

Christof

[Anonymous]

This isn't an answer, but just some context: I guess that usually all
scalars have their own dedicated canvases, and no one usually mixes both
scalars and objects. E.g. when it's necessary to make new scalars or clean
the data, a simple [clear( to the canvas solves it.
As I said this doesn't really answer your question, but maybe points out to
the reason why it went unnoticed - I assume you have a good reason to mix
objects and scalars together.

[jmmmp]