Cabbage Logo
Back to Cabbage Site

Strange trigger output behavior cabbageGetValue

#1

Hey all,

I’m trying to figure out why exactly I’m seeing this behavior. Basically I am trying to monitor when a widget value changes. I’m unable to use cabbageChanged because my array changes size, has many empty slots and it seems if you pass in an array containing blank/empty strings to cabbageChanged there is a memory leak that occurs, as I notice my memory consumption steadily ticking upwards

So, I’m trying to simply loop through and utilize the optional ktrig output from cabbageGetValue. However, for some reason after a single modification to the widget the kTrig output returned by cabbageGetValue never seems to reset, and the triggered index and the index after return a positive trigger value. So if I modify widget at index 2, after that modification cabbageGetValue is returning a positive value for its trigger output for index 2 and 3.

Here’s my code:

kIndex = 0
kValue = 0
while kIndex < gkVSliderLastIndex do
        kTrig = 0
        kValue, kTrig cabbageGetValue gSVActiveSliders[kIndex]
        if kTrig == 1 then          
            printk2 kIndex
        endif
        kIndex = kIndex + 1
od

I’ve ensured that the proper channel values are being to the cabbageGetValue call

#2

This isn’t going to work because cabbageGetValue has internal state. So it will end up comparing the newest value with the last value from the previous channel. This is why I added the array variation, but you’re right about it not being that useful if you are changing array sizes, although the leak sounds like a bug. How many channels are you looking to monitor?

#3

Ahhh okay, that explains why I’m getting some strange behavior

I’m looking to monitor between 8-50. Max may be somewhere around 150 but currently the most in a single array has been about 50. However I’m monitoring changes for a few different arrays with different types of widgets, but realistically maximum I’m looking at polling 120-150 widgets.

For my reference, could you provide a link to the source for that opcode? I figure I may as well to start familiarizing myself with the source as I foresee myself using Cabbage quite a bit. May as well start opening the door to the possibility of contributing once I wrap my head around it.

One other question I had, if I am polling for a change in value from a widget in one part of the code and then forcibly set the value of a widget in another, would the portion of the code that is awaiting a change get triggered or does manually setting a widgets value not set its internal state in the same way as modifying the widget via UI does?

#4

You can find source to all the opcodes included in Cabbage here:
https://github.com/rorywalsh/cabbage/tree/develop/Source/Opcodes
:+1:

If you’re monitoring a widget for a change, then setting it manually will still trigger the change. So you might end up in an kind of recursive mess.

#5

Ahh okay, so probably need a global state variable of some sort to prevent this