In the pre loaded code for labview (We’re using Mecanum w/Arm), there’s an error in and error out connected to each other. I was wondering if that meant I didn’t need to wire anything to an error out, as if say the connected error in and error out was universal.
In LabVIEW, error in and out is how errors are handled. The errors are data, specifically, a cluster of a Boolean, number, and string. If an error happens in one node, it passes that out to its caller.
If the caller if the node wants to check the error, it can access the values and write code to deal with it. Or the caller can string the error out of one node into another node. This makes the second node able to see the input error. Typically, it decides not to do anything except pass the error through. This error propagation is similar to how errors are thrown in other languages, except that the error is thrown forward instead of backwards.
To your original question, you generally do not need to use errors wires between nodes since the WPILib elements actually duplicate the errors into the reference wire as well. And if you get an error and don’t wire it to anything, it will show up on the DS – this is true on the roboRIO, but the simulator will put up a dialog telling you instead.
Greg McKaskle