In many cases, the t_next argument to Message() isn't used. In C++, there's a constructor that only takes timestamp and data, but in Python you have to explicitly specify None for t_next if you don't want to use it. It would be nice if you could omit it, using Message(t_cur, data=data). For that, both t_next and data need to have a default value (data because of Python syntax requirements).
If we default both to None, then Message(t_cur) is also valid, and sends a nil value. That sounds reasonable enough, and may even be a nice syntax for sending an empty trigger message. We should be consistent and allow the same syntax in C++ then, with an extra Message(double timestamp) constructor.
In many cases, the
t_next
argument toMessage()
isn't used. In C++, there's a constructor that only takestimestamp
anddata
, but in Python you have to explicitly specifyNone
for t_next if you don't want to use it. It would be nice if you could omit it, usingMessage(t_cur, data=data)
. For that, botht_next
anddata
need to have a default value (data because of Python syntax requirements).If we default both to
None
, thenMessage(t_cur)
is also valid, and sends a nil value. That sounds reasonable enough, and may even be a nice syntax for sending an empty trigger message. We should be consistent and allow the same syntax in C++ then, with an extraMessage(double timestamp)
constructor.