IN North America, the ABB/Westinghouse FT test switch is very popular for a variety of reasons, including that it's the same mechanism that was used on the electrotechnical relays of the past. The FT switch has a companion test paddle (separate source) that many relay techs carry several at all times. The FT switch can be used without a paddle, which is also a benefit. For CT circuits, many consider this a superior design, as one can see the shorting mechanism operate.
(ABB FT) Test switches must be wired so that the odd numbered terminals go to the relay and the even numbered to the 'system'. Failure to wire the switch this way defeats many of the benefits of having a switch. Wiring the switch with the even terminals to the relay may be seen as a safety precaution from a design standpoint, but actually makes things much more dangerous for the relay tech.
On output contacts, it's desirable to have both sides of the contact accessible at the test switch. The more positive side should be on the left. Inputs to the relay should also come through a test switch. Having the relay control power through switches also allows access to the battery for relay inputs. Most relay techs use the ABB test paddle when testing the relay, but this also provides isolation for the live blades.
I've seen modern installations with the GE XLA test switch and shake my head. These devices require insertion of 'tuning fork' shorting devices for CT circuits. There is one correct way to make these connections and multiple incorrect ways.