Events from built-in commands yet?

[Anonymous]

If you add a scripted command to the Fusion UI it works

by a scripted event callbacks when the user interacts.

Brian Ekins explained this in an online talk about 2 years ago.

He stated that there weren't callbacks from build in commands.

Has this changed?

I would like to do a scripted sanity check when a user opens

a design. i.e. connect a built-in command to a script event.

[marshaltu]

Hello,

Unfortunately there are no events for built-in commands yet. I would like to recommend you log your requirement to our ideastation so that our product manager can review/prioritize it.

https://forums.autodesk.com/t5/fusion-360-ideastation-request-a/idb-p/125

Thanks,

Marshal

[Anonymous]

Done. Thanks.

[ekinsb]

"Events from built-in commands" could potentially mean a lot of different things to different people.  Can you be a little more specific in what exactly you would like to do?  

[Anonymous]

I want to build an AI looking over my shoulder. Every time I enter something in the UI fusion has a callback the robot can switch on to tell

it a command has been executed, or a value has been entered.

At the moment I only need a robot assisting with user definer parameters. When I edit one, I want it recorded in a csv file of user parameters.

Fusion probably has all this as internal hooks between the UI and the engine. A systematic change to that mechanism to notify

the API would be useful.

[ekinsb]

I don't see us ever supporting the ability to watch and monitor as the user interacts with a command.  What I do expect we'll provide is more events related to changes in the data itself.  For example, when you edit a parameter we would fire an event that a parameter value has changed.  Or file an event when a feature is modified.  Fusion doesn't work the way you described.  It's not watching and reacting to events in the UI, but instead commands define their dialogs, collect input from the user and then call internal functions to do the work.  It's the "work" part that is important because that's what is creating or modifying the model and that's where I see us providing more support in the future.

[Anonymous]

I think we are on the same page.

Most of what I want could be done at the transactional level. I.e. a user has filled

out a form and hits "Ok". At that point the UI talks to the engine, and (if requested)

raise an event in the API.

It might be useful to get a callback per character typed, but that would be pretty specialised.