getOxygenSaturationHistory issue

Hi everyone,

here is the code I use:

if(What>19&&What<25){
		var tmp = getIterator(What-20);
    (...)
}



function getIterator(opt){
		var what = [:getHeartRateHistory,:getElevationHistory,:getPressureHistory,:getTemperatureHistory,:getOxygenSaturationHistory];
		if(Toybox has :SensorHistory && SensHist has what[opt]){
			var Num = gHW.toNumber();
			var getMethod = new Lang.Method( SensHist, what[opt]);
			return getMethod.invoke( {:period=> Num} );
		}
		return null;
	}

So it works nice on lot of devices (only 3.2.x)

but not on MARQ, F6/ F6s (non Pro) as soon as I put what = 24 (pulse Ox)

if I put something else than Pulse Ox, no issue.

here is the issue returned:

Error: Out Of Memory Error
Details: Failed invoking <symbol>
Stack:

on memory usage I have: 22.2/92 - peak = 24.8

when Sys.print test (on device with issue) tmp = object, tmp.next() = null

So I don't get at all what happens, can't test on real devices.

jim_m_58 over 6 years ago

There looks to be a bug in the sim where there's no samples in the iterator on some devices, but I think it exposed a bug in your code - handling the case where there are no samples (say right after a device is reset)

Try this:

    	//
    	var it=SH.getOxygenSaturationHistory({:order=>SH.ORDER_NEWEST_FIRST});
    	if(it!=null) {
    		var s=it.next();
    		if(s==null) {Sys.println("s is null");}
    		else {Sys.println("s not null "+s.data);}
    	} else {Sys.println("it is null");}

On a marq in the sim, s is null, while on something like a 945, it's not, and it's got data. If I use period,

var it=SH.getOxygenSaturationHistory({:period=> 10,:order=>SH.ORDER_NEWEST_FIRST});

I see the same.

jim_m_58 over 6 years ago in reply to jim_m_58

Also see https://forums.garmin.com/developer/connect-iq/i/bug-reports/toybox-sensorhistory-getoxygensaturationhistory-strange-samples

There's some other oddness going on.

Shent-Aurkhan (SHN) over 6 years ago in reply to jim_m_58

many thanks,

my code is not complete, after :

var tmp = getIterator(What-20);

I do:

if(tmp!=null){

    var tt = tmp.next();

    if(tt!=null){

        D1=tt.data;

        (...)

    }

}

it is what you suggest right?

you suggest first to test without the period? (I remove it and same result)

Shent-Aurkhan (SHN) over 6 years ago in reply to jim_m_58

I saw it,

I guess I have to find for someone with those devices to test on real

jim_m_58 over 6 years ago in reply to Shent-Aurkhan (SHN)

Actually, I'd probably hold off with doing anything with this unless you have a device to test on.

Right now, I'm looking into what seems to be that the sample are always in oldest first order on a 945.

On a real device, there can be up to 192 samples. One sample is taken every hour, that's today, and the last 7 days.