Device types?

The monkey.jungle has lines like the following (where "descentg2" is a device type being compiled):

descentg2.resourcePath = $(descentg2.resourcePath);resources-launcher-60x60
descentmk1.resourcePath = $(descentmk1.resourcePath);resources-launcher-40x40
Is there a way of testing what device is currently being compiled?
I have a work around for some devices that have bugs. I'm currently using the part numbers but that isn't always going to work.
  • One can use strings as flags and put the string files in folders like:

    resources-etrextouch

  • The part number should always work, but understand that some device have multiple part nums

    In the sim you'll see the WW part number but devices can have different part numbers for APAC devices or variants

    As far as while compiling, at least in VSC on win11, you see the compiler command line

    ----

    Is so this so can check the device with a bug/crash?  ERA will do that for you

    Error Name: System Error
    Occurrences: 1
    First Occurrence: 2026-02-03
    Last Occurrence: 2026-02-03
    Devices:
    fēnix® 8 Solar 51mm / tactix® 8 Solar 51mm: 21.18
    App Versions: 0.22
    Languages: rus
    Backtrace:
    BleScanView.saveResult:256
    BleScanBleDelegate.onScanResults:30

  • The part number should always work, but understand that some device have multiple part nums

    That's the thing I want to avoid.

    We should be able to do the following:

    etrextouch:

    =======

    The flag-in-a-string in a device resource folder (like "resources-etrextouch") looks like it works.

  • Is so this so can check the device with a bug/crash?  ERA will do that for you

    No.

    Some devices have things that doesn't work properly. So, I need special code for them. (And some other stuff.)

  • Should you be using "has" may null/range checking?

    Do you have example of any devices where you are having this issue.

    Any chance this is related to the FW/API version of the devices?  In the case of WW and APAC versions of the same device, the FW may be different

  • Should you be using "has" may null/range checking?

    I'm dealing with bugs (things that don't work "as documented"). 

    Any chance this is related to the FW/API version of the devices?  In the case of WW and APAC versions of the same device, the FW may be different

    There's no practical way of determining that. It would be OK to add the work-around if it isn't needed. It ends up being bad if I don't add the work-around if it's needed.

    ==========

    The bug is pretty bad: on the 1030s, the back "button" in widgets doesn't handle the stack popping properly. So, I add one that does. (I reported it years ago. It works fine for device apps.)

    I also need to know the font being used in a layout. (The font used is different for different screen resolutions.)

    There's no way (as far as I know) of getting the font associated with an element. So, I have a string flag that indicates the font being used.

  • With a widget, you really want to ignore onBack and onKey for KEY_ESCAPE and let the system just take the default action.  It's between that way on devices since the early days.

    I do the same for glances that can also run as device apps.

  • The touch-screen Edges have an onscreen back “button”.

    This works as expected for device apps (the app can keep track of states when menu views are popped) This doesn’t work for widgets. So, I add a back item to the menu with a note. Not ideal because people will tend to use the standard back “button”.

    I have two different 1030 models (both have the same problem). I’m assuming (taking a risk) that the problem is only with the 1030s.

  • So what do you do in your app delegates for the back button?  What I'm saying is try commenting that out and let the system take the default action

  • In the added back item, I pop the view and keep track of the parallel change of state of my program.

    Basically, it uses nested dynamic menus.

    Again, the default system action works for the device app but not for the widget. It should work the same, right?

    It works fine on my (new) Fenix 8 but the Fenix 8 doesn’t have the screen back button (which is the thing not working properly).

    The touch-screen Edges don’t have the physical back button.

    Do you get a KEY_ESCAPE with a Menu2? Does the physical back button generate KEY_ESCAPE for Menu2?