What's wrong with this workaround --> Disable wires/connections between nodes - #8 by gregorius
1 Like
Thatâs what I get for scanning the thread first thing in the morning. Very good workaround! Thanks
The OP implied there was a way to create some sort of macro to do this with a keyboard shortcut. Making macros isnât possible, right? He was just referring to using 2 shortcuts in a row, right?
Just discovered this shortcut that isn't normally assigned:
Assign that with some keyboard shortcut. And then assign this to a keyboard shortcut
And then it should be quick to disable wires.
3 Likes
Well, it will be quick to disable the wire. But you would still have to DEPLOY to get it applied.
Add a shortcut for the deploy action, so that's three shortcuts.
It wasn't meant as a put down. I was only saying that it is still a step which needs to be done.
Did I sound like I took as such? Sorry - I didn't read it at all like that. It was dark and stormy and early humourless morning.
I, for my part, map deploy to Cmd-S so it's not actually any work to deploy ...
Since this basically works with the workaround now, I was wondering whether this feature could be added quite easily. From an implementation perspective, the GUI would provide a âdisable wireâ function, and in the background there would still be a junction node but hidden, meaning it would no longer be visible to the user - done. A bit âuglyâ in terms of implementation, but maybe worth considering?
No.
Either properly so that it's fully shown in frontend, i.e. that an junction node was inserted and it has been disabled or ..... - oh, in fact, I can't think of an alternative 
So, I for one, am against such a solution but hey, IDC.
"Properly" would probably mean that wires can have properties (now that's a cool sentence 
). But I don't think that's the case so far. The junction node also seems to be a workaround to avoid exactly this problem.
By IDC, do you mean "I donât care"? About what?
Another constellation that the frontend or NR has to work out whether to "hide" the junction node:
It's the hiding of the node that isn't considered in the suggestion. It's not that simple and as soon as you go into the details, you discover this is non-trivial consideration.
The junction node servers another purposes, it is being misused here. It's purpose is to bundle multiple connections from many nodes to one node or vice versa. Visually it looks nicer. For the execution of the flow, the junction node can be eliminated, it plays no role.
It's a typo, I meant to write IDGFB
It's indeed possible that hiding the JN node might create more problems than it solves. If still considered, I would start with the simplest case (N ---> JN ---> N), and probably stop there, until wires are allowed to have properties.
IDGFBE
Confucius once said may the boomerang bend and come back as a banana.
Have a go yourself, edit the editor code and see if you can implement this. That's the best approach, to come with a working solution and not an open problem.
If you don't code JS, it's the best time to start: plenty of AI helpers to guide you along the path of becoming a JS Jedi.
Damn these dark and stormy mornings....
You should consider changing your LLM model 
I might, if my family lets me.
1 Like
Folks often confuse me with an LLM but then again an LLM would say that.
Why not just disable the node the wire goes to, or comes from ? Saves adding a junction then disabling it.
If it's for example a node with multiple outputs/inputs, and I want to disable a single connection. Disabling the node would block everything, no?
Once I have some time to go back to the flow I was working on, I can provide a proper example, where it would make sense to disable a wire, in my opinion.
Has node-red-contrib-flowgate (node) - Node-RED been consideredâŚ
1 Like
That would work as well. I just want things to be tidy (without extra "nodes" and logic).
Yes - though in my experience it is fairly rare to have a node with multiple outputs feed another with multiple inputs - usually (for me) at least one of them just has a single connection - which is the end I would disable, but yes no doubt multiple to multiple do occur.
