SDK 6.2.0

This seems weird. Big jump to SDK 6? And a dramatic reduction in size?

NOTE: In Eclipse I had to go into PREFERENCES and manually point to the new SDK. The SDK Manager did not update that from SDK 4.2

Top Replies

mcinner1 over 2 years ago in reply to yp +2

1) SDK 6.2.0 no longer allows IDs with start underscore - e.g. _myName, which I do with all Resources/Setting IDs. So there is a lot of work waiting because I have to change the IDs.
2) if I therefore do…
_psx_ over 2 years ago +1

AlphaMonkeyC

What will the version number look like in 2024 q2 when there will be no system7?

6.12.0?
mcinner1 over 2 years ago in reply to yp +1

Yes - I, too, rolled back to previous 4.version and then all Edge x40 devices crash on simulator (on every dc. Command like dc.fillcircle….)

I‘ve just testet a small datafield with Edge 1040 and…

All Replies

0 gasteropod over 2 years ago in reply to andyj713

That looks like a type error. I have type checking turned off since the apps pre-date types by a long time. Maybe something changed related to types? You could turn off type checking to test/as a workaround.

It's not clear to me if you are saying 6.2.1 specifically did this, or if you'd see the same thing with 6.2.0 too? Cannot see anything in the 6.2.1 changes which would account of this kind of issue being introduced there if it did not exist in 6.2.0. That would suggest a mistake if it is the case.

0 andyj713 over 2 years ago in reply to gasteropod

I didn't try 6.2.0, but it's probably a type thing. I added some lines of code to declare a variable using getApp() and I'm using that instead of the class path and the compiler likes that better. I suppose this falls into the category of things that we used to be able to do because the compiler was supposed to catch it but didn't but now it is?

0 jim_m_58 over 2 years ago in reply to andyj713

Sounds to me like something typechecking is flagging something that isn't really a problem in the code You may want to look at what you added for type checking

0 andyj713 over 2 years ago in reply to jim_m_58

It was this prior to 6.2:

			try {
				tryProp = AppBase.getProperty(propName);
			} catch (err instanceof Lang.Exception) {
				getPropErr = err.getErrorMessage();
			}

Now it's this:

			var app = Application.getApp();
			try {
				tryProp = app.getProperty(propName);
			} catch (err instanceof Lang.Exception) {
				getPropErr = err.getErrorMessage();
			}

The compiler was complaining about a symbol, not type, but it seems like it can resolve symbols better when they are part of a object variable vs. a class reference? Maybe it's an obscure enforcement kind of thing?