How To Add A Command

How To Add A Command

Adding a command usually touches both the application layer and, if it changes editor state, the domain layer.

Decision Tree

flowchart TD
    Start["New command"] --> Pure{"Does it change pure editor state?"}
    Pure -- Yes --> Domain["Add or reuse domain transition"]
    Pure -- No --> App["Add application/workbench behavior"]
    Domain --> Action["Map command to AppAction"]
    App --> Action
    Action --> Registry["Register in CommandRegistry defaults"]
    Registry --> Config["Optional: expose through config/keymaps"]
    Config --> Tests["Add tests in owning layer"]

Typical Steps

1. Add or reuse an AppAction in rim-application/src/action.rs.
2. If the command changes pure editor state, add the transition in rim-domain.
3. Handle the action in rim-application/src/action_handler/.
4. Register the command in rim-application/src/command.rs.
5. Add or update tests in the owning crate.

Where Not To Put It

• not in rim-app unless it is runtime-only
• not in an adapter unless it is purely host-specific input mapping

Config Exposure

If the command should be user-bindable:

• give it a stable command id in the registry
• ensure exported defaults include it
• confirm keymap/config load still works through rim-application::config