Pin visibility

[mhusinsky]

The info about optional pins is not very helpful. Possible values in the configure inspector are optional and hidden (apart from visible)

I cannot read the difference between those two and the explanation hardly makes it clearer.
Also: when is it a good idea to use a hidden pin, which is not visible and cannot be activated in the inspector?

[joreg]

looks like you stumbled on the "wrong" documentation there. the docs for the definition side of pin visibility are here: https://thegraybook.vvvv.org/reference/language/operations.html#visibility-for-inputs-and-outputs hope this explains better.

what you found is the docs for the application side, where there are indeed only two options: show or not
obviously screenshots would add to the clarity in both cases.

re hidden: @azeno @gregsn do we have an example of hidden inputs in the real world?

[mhusinsky]

ah, I did not see that (the page I linked was the first meaningful search result).
I added a link to the paragraph so it is easier to find.

[azeno]

Hidden pins are currently used for the "Node Context" pins. They are treated in a special way by the system, as it injects the ID of the node itself. There are cases where you want to get access to your own "Node Context", so one needs to create that pin manually. However you probably still wanna hide it from the user -> use the Hidden flag.

Upcoming developments in the inspector branch will enable another scenario for them: the ability to store a model on application side, but not showing it to the user. As an example think of Kairos. It's Timeline node currently shows a white Model input where it stores its timeline. That pin shouldn't be visible at all. It currently only is visible because we didn't create pins on application side if the definition said it's hidden. So we couldn't store anything there. With the inspector branch this restriction will be lifted, all pins will have a counter part on application side and therefor storing values will be possible.