Somebody else noticed strange behaviors of your desktop app under xojo2024.2?

Did someone else noticed weird behavior of your desktop app running under xojo 2024.r2?

I did develop an app and do upgrade every time also to the newest xojo desktop. It runs fine all the time till I upgraded xojo to 2024.r2

Buttons did become unresponsive, selections could not be made anymore.

However, when I fall back to 2024.r1 everything is working fine again.

Windows, Mac, Linux, All?

If you can create a sample project that reproduce the problem create a public Issue. If not you may want to privately share your project with Xojo.

I am working with Windows 11 and it happened on:
my laptop: Intel i5, 10th Gen, 8 core, mem 6Gb
My desktop: intel i5, ? Gen, 4 core, 16Gb

Xojo desktop for Windows 2024.r2

Does it also happen with one of the more complex sample projects like “EddiesElectronic”? Any more details regarding your app? Does it “hang” immediately after start?

1 Like

I’m also worried about this condition. I’ve read that even Xojo is unstable in unknown conditions, so how to reproduce is unreliable. At least on Windows, my main platform of interest.

I think I’ll keep at 2024r1.1 for a while.

1 Like

Can you make us a build from 2024r2 so that we can have a look if you can’t reproduce the behaviour?

1 Like

As Alberto said, can you cut a small sample screen from your app showing the wrong button behavior that works ok in previous versions but fails in R2?

1 Like

I’m not currently seeing any issues on Mac, is it Windows only?

This kind of thing seems Windows only:

1 Like

I noticed when debugging an application In windows, once the app opens and is over the xojo IDE, most of the buttons in my application are un-responsive on the app until I minimize xojo.

Looks like a match.

1 Like

ok. here is what I encountered and why I stick on xojo_2024-1.r1

Loading the desktop app in Windows xojo-1r2 goes without problem.

Then I start the program and directly after start this happens:

When I press continue, nothing happens. After about 5 times pressing continue I suddenly do get a runtime error I never had before:

When I press ok, the IDE closes the app and comes back in write/debug mode.

When I load the app in Xojo_2024-1r1 it goes well:

When I run the app no errors appear (never did before) and the app runs like a champ:

I just did reprocess this, so I guess one can reprocess this also with another desktop app.

Concerning the complexity, what @Thomas_Kaltschmidt mentioned, I think this app is complex enough.

The first time I did not directly notice the behavior, because I quit the app, but later on I noticed that the IDE pauzed and that was the reason why my buttons did not respond.

I do only have the question now, why 1r2 is running into this runtime error, which I cannot make chocolate of it, and all previous versions never did run into this error.

In the meanwhile I did fallback to xojo_2024-1r1

So you get an unhandled exception.
Are you handling those in App.UnhandledException’s?

If not:

  • Add the EventHandler UnhandledException to App
  • Add the following line of code in the Event Handler: break (*)
    • It’ll then break in the Debugger while debug-running
    • Look at the stack trace - where do you get the InvalidArgumentException from?

My guess… maybe some Window Width/Height value, or some ScrollBar Property.

(*) or something along these lines to show the stack trace (just for debugging purpose, don’t use like this for production):

Function UnhandledException(error As RuntimeException) Handles UnhandledException as Boolean
  MessageBox String.FromArray(error.Stack, EndOfLine)
End Function

What does ‘start the program’ mean? Are you clicking ‘Run’ to debug and then you get the errors/dialogs? (just want to be sure)

Can you share in private your app to Xojo so they can take a look?
Maybe they know what ‘Minimum or Maximum values are over the limit.’ and what changed between 24r1.1 and 24r2 and can help even without your app?

Dump the stack into a message, look for where this new exception is coming from.

According to the message in the screenshot, it seems a coded value is outside the valid range, and the new compiler version now crashes because of.

Everybody knows that, but he needs to locate WHERE and WHAT is being affected. Maybe the stack trace may help us to locate.

I’m personally very curious about what “component” raised it.

But as I guess Rob is from NL, and it is almost 21h00 there, it may take some time to see an answer.

yes, sorry…run to start debug

That’s not a crash. It is an unhandled exception. It is always good practice to add an exception handler to the applications class. It acts as a last line of defence for exceptions you haven’t already taken care off.

Add an event to App, choose UnhandledException. It will be called if you cause an exception and don’t deal with it elsewhere. You will be passed an object into the event that tells you what went wrong (more or less).

See : UnhandledException

Also if you enable ‘Break on exceptions’ on the debug menu it should stop on the line of code that is causing the problem.

1 Like

Hi @Jürg_Otter, I followed your instructions and putting ‘break’ in the function UnhandledException only breaks the run and shows nothing.

When I put “MessageBox String.FromArray(error.Stack, EndOfLine)” in the UnhandledException function I do get below, which also tells me not more:

Any other sugestions?