Debugger will not "Step In" to function/method

Class with several methods. The debugger will not step into - using debugger interface "Step In' - one of the functions. It will stop there if I add a break point. The function is doing its job and looks no different, to me, from any other I've ever written. No other function in the file behaves like this.

Anyone seen this before? I'm concerned something more may be wrong!

[James Tursa]

Is the function an m-code file? Or is it p-code or compiled code?

[Dave Watson]

It's an m-file

[dpb]

What if you physically rearrange the order of the functions in the file; does the behavior change?

I'd think the only possible way to learn anything here would be to attach the file such that somebody else could test the behavior on another system to see if it is within your specific installation or is somehow actually a characteristic of the specific file/function.

Just ouf of curiosity, is the behavior repeatable if you dbclear all and/or begin a clean session? I've had a few instances where it seemed like perhaps the debugger got confused after a long session and after restarting all seemed normal again. I'm never surprised if there are some warts in such code; it's bound to be extremely complicated to write/maintain.

If nothing seems to make the behavior go away including any patches to the release, it may just be a bug. But, if the Release information you attached is correct, it's almost 10 years since R2015b; it may be time to consider upgrading.

[Dave Watson]

@dpb, thanks for the ideas. Have responded in your text, copied below:

What if you physically rearrange the order of the functions in the file; does the behavior change?

>> Tried that (again) just now with no improvement.

I'd think the only possible way to learn anything here would be to attach the file such that somebody else could test the behavior on another system to see if it is within your specific installation or is somehow actually a characteristic of the specific file/function.

>> That's not practical, unfortunately - file is part of a large code base for an app.

Just ouf of curiosity, is the behavior repeatable if you dbclear all and/or begin a clean session? I've had a few instances where it seemed like perhaps the debugger got confused after a long session and after restarting all seemed normal again. I'm never surprised if there are some warts in such code; it's bound to be extremely complicated to write/maintain.

>> That had no effect. This happens after any and all MATLAB and app code restarts.

If nothing seems to make the behavior go away including any patches to the release, it may just be a bug. But, if the Release information you attached is correct, it's almost 10 years since R2015b; it may be time to consider upgrading.

>> Not practical, really. If there was a good indication that it would help, maybe, but it would mean disruption for a lot of customers. Thanks for the ideas!

[dpb]

If you have kept the maintenance up; you could always submit a bug report and see what TMW says/responds but I'd guess they're not going to be very interested in R2105b at this point unless they can reproduce the issue in a current release.

If you can't pull just the one class module out with a driver routine to let somebody step into the subject function, then I think there's no possibility of anybody being able to do anything remotely.