Annoying error message...

[stam]

Hi all,
as of 9.6.6, i've had an issue with the debugger. I used to get moderately helpful error messages with a button to jump to the offending script (you know the one i mean...)

As of 9.6.6, this appears for a split second, then disappears because the following dialogue appears and closes the debug message:

I'm just running this in the IDE/browse tool, no remote debugger is in use (i do have the pro licence but this never used to happen in the IDE).
This is driving me nuts because i cannot debug a super annoying error because i cannot see the error message.

Any suggestions?

[dunbarx]

Hi.

Can you run the very same setup in 9.6.1 or 9.5? Does it happen there/

Craig

[richmond62]

Can you run the very same setup in 9.6.1 or 9.5? Does it happen there/

As of 9.6.6

I would take stam's posting to say that it doesn't.

[dunbarx]

I would take stam's posting to say that it doesn't.

So would I. But before filing a bug report I thought it prudent to ask.

Craig

[stam]

Good point guys; this certainly didn't happen with earlier versions but as i have IDEs for 4 languages eating up huge chunks of storage on my meagre laptop i usually get rid of older versions as soon as the newer version appears OK - so all deleted now.
Will re-download and see what happens...

[jacque]

The error says it can't give more info because the stack is password protected. Is it? If so, try removing it for now.

[stam]

None of my stacks are protected. I may be wrong, but I read this as the remoteDebugger stack being password protected…

Not sure how and even if I should remove that!

[jacque]

None of my stacks are protected. I may be wrong, but I read this as the remoteDebugger stack being password protected…

Not sure how and even if I should remove that!

I think you're right, and even if you wanted to you probably couldn't remove the password. This weirdness should probably be a bug report.

[stam]

I checked earlier versions. This affects versions 9.6.4 onwards for me - then again i've only had access to remoteDebugger features with the newer 'standard' versions (i.e. 9.6.4 +). On earlier versions i only have an indy license which doesn't have access to the remoteDebugger, and of course this error does not occur.

I've submitted a bug report here, for what it's worth: https://quality.livecode.com/show_bug.cgi?id=23543

[jacque]

I've never seen this particular problem but for now you could try adding an errorDialog handler. It would at least let you know which line in the script caused the problem. There's info and a sample handler in the dictionary.

[FourthWorld]

I've never seen this particular problem but for now you could try adding an errorDialog handler. It would at least let you know which line in the script caused the problem. There's info and a sample handler in the dictionary.

That raises another issue: should the error info divulge any internal details about protected scripts?

[jacque]

It may not be the debugger script. I'm getting spurious error messages when trying to build for Android that have nothing to do with the real problem. At any rate, all the error would tell you would be what script it was in and which line. If it's the remote debugger script then something odd is really wrong and if it's protected no one could look anyway. If it turns out to be stam's script then he'll know where to look.

With this problem and the one I've been having, I'm starting to think the error messages are screwed up. Maybe the numbers are mismatched.

[stam]

I've never seen this particular problem but for now you could try adding an errorDialog handler. It would at least let you know which line in the script caused the problem. There's info and a sample handler in the dictionary.

That raises another issue: should the error info divulge any internal details about protected scripts?

Hi Richard,
No, the error is not in revRemoteDebugger, it’s in my own code.
After many hours of scratching my head I finally located this and fixed it. And besides in none of the cases I’ve had should the remoteDebugger ever have been involved - I’ve never used the remote debugger!!!!

But for some reason the message about the remoteDebugger now comes up with any number of errors - as far as I can see this seems to be the usually less helpful “can’t find background” error in both cases of major errors I’ve had (I say less helpful because the normal message errors never specify which object cannot be found, but at least the “normal” debug window would have a button to take you to the offending script).

[FourthWorld]

No, the error is not in revRemoteDebugger, it’s in my own code.
...
But for some reason the message about the remoteDebugger now comes up with any number of errors...

I don't understand how the two statements can coexist. I'll step out to find stronger coffee...

[stam]

let me put it this way - when i'm supposed to get the debugger error window, i now get a dialogue that tells me revRemoteDebugger is password protected, and unfortunately this dialogue closes the debug error window. This is on running projects in the IDE, where the remoteDebugger should not be invoked.

This seems have happened recently - at first I didn't even realise this was a bug. If you look at a different bug report i made here: https://quality.livecode.com/show_bug.cgi?id=23521#c5, you can see that when asked to reproduce this, i stated that this was now happening (thinking this may be normal course of events), but really when i submitted the bug in Jan 2022, this wasn't happening and i had described a different error.

Sometime between mid-Jan and now, something has happened to my system that seems to want to redirect the error tracing to remoteDebugger...