I think a more typical, high performance operation for copying arrays would be using mem copy operations. This is how Go handles a lot of slice operations under

I think a more typical, high performance operation for copying arrays would be using mem copy operations. This is how Go handles a lot of slice operations under the covers. If it's a C style array you could always just keep it one larger than the actual size; perhaps place a null value at the end and swap it out as needed. Just build an abstraction over it.