Unsure how to express what I mean, so I’ll try and do my best.
Time and time again, I come up against the same problem: how to trigger nodes and in what order.
I’ll gave my latest example, maybe it makes sense. I have a flow, that takes a bunch of Excell’s, that in turn have a bunch of sheets. It loops through everything, does … things, and because memory and speed limitations, I first export the result in a CSV, then read in the CSV and then convert it to an Excel. Because of the nature of the CSV, I cannot have it on overwrite, so it’s on append. That means I need to delete the CSV on every run, otherwise … problems. Now, the problem. How do I trigger the Delete files/Folders node? As it has no other input than the System connection port, I can’t really add it at the end of the flow. Can’t add it to the start, as I have a List Files/Folders as the first node, for the loop. So, as a workaround, I have them both (I am deleting 2 files) going in a Joinder node, and I linked it to the List files node with a variable. And this is the same workaround I use whenever I have the need to stagger outputs, I trigger them using the variable connection. While it works, I think it would be easier to have these nodes also have an IN port that does nothing other then trigger the node.
In the example below, I used the same trick to trigger the flow beginning:
In this case, I had several unrelated inputs being processed independently, but that needed to end up in the same export. I did not find any other way of doing that, except this useless adding of variable connections.
Hopefully I made enough sense to be understandable.
