As in the image, be able to attach a debug node to the input of the mqtt out node. To see what's coming in to that node.
As it stands now, I'd have to insert a dummy node before the mqtt node that would just pass through messages so that I could see what was going through there by attaching to THAT node's output node before it gets to the mqtt node.
It can be a pain if you have, say, five or ten wires to a node and want to debug what it is seeing. Often the easiest way is to insert a do-nothiing node as you can drag all the inputs across to it in one go. It would be nice to be able to do something like that requested here, but I am having difficulty seeing exactly how you would represent it in the editor.
I know how to do that. The point of this request is to implement a method so I don't have to do that.
As in the image(Can you guys see the image in the post?), there are 6 inputs to the mqtt node. To do what you suggest, I would have to make 6 additional connections from all the connected nodes to a debug node. I don't want to have to do that every time this situation arises.
That would require moving sometimes multiple connections from one node to another multiple times as you add and remove the debug nodes to the project. That can add to the chances of something going wrong. A missed, lost. or miswired connection as you move things around.
To be honest, I didn't know about the shift-click thing. Which helps, but makes me wonder about how many other features aren't well known. I would have moved each connection one at a time.
This is a genuine problem. There are lots of features that get added and mentioned in release notes/blog posts. But for new users who haven't followed each release of the project, it can be hard to discover them all.
For 1.0 I plan to recreate a whole series of short videos around the various features of the editor, but for now, you may want to read through the Editor Guide, such as the page on wiring.
An inline debug node would be a great addition. That would increase flow readability a lot, and messages won't have to be cloned. However, where should the disable button be placed?
The topic of proper (inline) debugging and logging - especially in larger projects - is an ongoing issue at work, too.
For now, we are using a custom node with one input and one output. It uses Winston to create proper log files AND has an option to send the log event object to the debug side bar, to provide some kind of inline logging.