Proposed resolution:

This wording is relative to N5014.

1. Modify [hive.operations] as indicated:

   [Drafting note: If issue LWG 4320 is decided to be acted upon and the proposed solution accepted, the proposed complexity wording becomes:

   -11- Complexity: If `empty()` is `false`, exactly `size() - 1` applications of the corresponding predicate, otherwise no applications of the predicate. For each element erased as a result of this function call, at worst 𝒪(log n) in the capacity of the block containing the element. Additionally, if any active blocks become empty of elements as a result of the function call, at worst linear in the number of element blocks.

   ]

   template<class BinaryPredicate = equal_to<T>>
     size_type unique(BinaryPredicate binary_pred = BinaryPredicate());
   

   -7- Preconditions: […]

   -8- Effects: […]

   -9- Returns: […]

   -10- Throws: […]

   -11- Complexity: If `empty()` is `false`, exactly `size() - 1` applications of the corresponding predicate, otherwise no applications of the predicate. Additionally, if any active blocks become empty of elements as a result of the function call, at worst linear in the number of element blocks.

   -12- Remarks: […]

Currently we specify for `erase` [hive.modifiers] p16:

Complexity: Linear in the number of elements erased. Additionally, if any active blocks become empty of elements as a result of the function call, at worst linear in the number of element blocks.

This is to allow for potential `hive` implementations as vectors of pointers to blocks, as opposed to linked-lists of blocks (reference implementation approach). In that scenario removing a block would involve a vector erasure and subsequent pointer-to-block relocations. swap-and-pop of block pointers would not work as this would re-arrange the sequence during erasure. Anyway, we have neglected to apply this same complexity to `hive::unique`. It would be rare that this would occur for `unique`, as it would involve (1) a block A with only one non-erased element left in it and (2) a block B preceding/following block A in the sequence, whose last/first element is equal to the element in block A. But it is possible, so the same consideration in terms of time complexity applies.

This is a separate but related issue to LWG 4320, but it would affect the outcome wording of that issue for `unique`.