How important is the "count" column in the application stats? Does it really matter how many methods are called, or is it more about total busy %? If an application separates out functionality into many methods, does that actually slow the hub down more so than an application that does the same thing but in a single big function?
That is the number of time the app (or driver) "runs," or wakes, not the number of methods called per se -- though given that these all happen via callbacks, each would naturally be at least one method call. But I don‘t think it counts anything beyond that. Are you finding evidence to the contrary?
I had sone discussions around this with a developer here. It is about how many methods are called, but it is more complicated then that to. Depending on how the method is called you can get elevated numbers for time as well. So probably good to look at both but as long as overall time isnt to bad how many times methods are called just means how busy a app/device is.
I would just remember these are to help point in the right direction and not exactly perfect numbers.
Is there a good, efficient way to log over time which methods/functions are initiating an instance of the app to run? I have some webhooks in an app in particular and need to see how often those are being used, without having to stare at the log for hours on end. Best I can think of is to log to a file but seems like there should be a better way.
I am not an expert of Hubitat apps... but I did have this thought - perhaps you could just create Application State variables for each method, and simply increment that value each time the method is called. You could then look at the app's settings to see the value of these state variables. I would only do this during development and debug, but would remove this afterwards.
