Our legacy product, Lavastorm Analytics Engine, had a Bypass node which inspects each input at compile time in descending order, and connecting the first valid input as its single output. This node makes compile-time decisions about input availability. In other words, the output of a bypass is the first, or topmost, input that can be satisfied. To be satisfied means that, when a graph is started, the first or topmost input that is connected to an output and has all predecessor nodes enabled, will have its data passed to the output of the node.
Note: Infogix plans to introduce the Bypass node in Data360 Analyze in the future but in the meantime this workaround can be used.
There are 5 examples of the Bypass workaround. The number on the workaround indicates the number of input pins and the Bypass Workaround data flow demonstrates what they do.
In Data360 Analyze, here is what the bypass workaround looks like with 2 inputs:
And this is what it looks like in BRE:
In Data360 Analyze, here is what the bypass workaround looks like with 4 inputs:
Here is what it looks like in BRE: