conversion of rT.variables in mg/dL

[mountrcg]

Needs to be done in parallel with nightscout/trio-oref#30

This concerns mainly ISF as transmitted from oref to suggestion / enacted. So far I only found one instance where ISF is used.

So far ISF is always transfered from oref in the current user glucose unit, so if mmol/L is used it comes in mmol/L. Therefore currently no conversion is applied in Trio to ISF originating from oref. Drawback is also that a change of user glucose unit will not be reflected in ISF unless a new loop is calculated. Refer also to nightscout/trio-oref#29

Also the watch app needs to adjusted as ISF may also be shown on watch.

[mountrcg]

I need a little help on the watch part, still throws an error in the conversion function

Trio/FreeAPS/Sources/Services/WatchManager/WatchManager.swift

Line 211 in a731a13

from: (units == .mmolL ? isfValue.asMmolL : Decimal(isfValue)) as NSNumber

[mountrcg]

successfully tested!
As Trio does not show ISF anywhere on UI, only watch can be used to review the changes.

Old sitiuation - ISF is only shown converted after a new loop

original state ISF	after unit change, no new loop	after new loop

After change

ISF is immediatly converted after changing units on phone. As it can be seen the popup still shows units from last loop, watch already has ISF converted.

original state ISF	after unit change, no new loop

[mountrcg]

it can also be reviewed with enacted.js in Developer option.

[dnzxy]

Interim fix. Will need changing with Trio 1.0.0, but for 0.2.0 okay.

[bjornoleh]

I tested this now in Xcode / iPhone simulator, and it works great.

• ISF instantly changes units on the watch when changing the units setting
• At loops, the ISF value in suggested and enacted are updated, and displayed with the chosen units both in the enacted box in Trio, and in the Nightscout OpenAPS pill.

We will be diverging a little from openaps/oref0 by doing this, which is probably fine.

@mountrcg, did you consider PRing the oref changes to upstream oref0?

[dnzxy]

From my conversation with Robert, this actually is not something that diverges from oref0, but is a change that he introduced himself about 2 years ago, and we are making it so that ISF is treated like all other settings.

[bjornoleh]

From my conversation with Robert, this actually is not something that diverges from oref0, but is a change that he introduced himself about 2 years ago, and we are making it so that ISF is treated like all other settings.

Ok, thanks for the info, yes this rings a bell. But when checking relevant lines in determine-basal, it seems his changes then were not 1:1 with the current PR:

The PR from 2021 is here:
openaps/oref0@da70837