The getBytes is part of the “data utilities” and will convert a bytes representation (e.g. hex string or Uint8Array) into a Uint8Array. It is used on the BytesLike type.

The toBeArray is part of the “maths utilities” and will convert numeric types (e.g. hex string representations of numbers, numbers, bigint) into a big-endian representation of bytes.

The difference you are seeing is when you have leading zeros, such as "0x0012". As a number that is the number 18, which is Uint8Array([ 18 ]), while as a bytes that is Uint8Array([ 0, 18 ]).

For your purposes, you should likely only need getBytes unless you are interested in binary representations of numeric values, which is rare outside writing data encoders or trying to encode payloads.

Does that make sense?