fix: FMI varname parse bug #3934

[EduardoMartinezLopez]

This PR Fixes #3934

Checklist

• Appropriate tests were added
• Any code changes were done in a way that does not break public API
• All documentation related to code changes were updated
• The new code follows the
  contributor guidelines, in particular the SciML Style Guide and
  COLPRAC.
• Any new documentation only uses public API

Additional context

Add any other context about the problem here.

[ChrisRackauckas]

This reformats the whole repo. Right now the formatter is broken so don't worry about it. Revert that.

[EduardoMartinezLopez]

@ChrisRackauckas Thanks! I reverted the formatter commit.

[EduardoMartinezLopez]

@ChrisRackauckas Question for you. It seems like the existing code in MTKFMIExt > parseFMIVariableName() mutates the "name" input variable. Is that OK? My understanding is that this is considered bad practice as Julia passes by reference. Also if it the input is mutated, the convention is to add ! to the function name, right?

[ChrisRackauckas]

I don't see where it mutates it. It creates a view but then creates a Symbol, which is effectively an interned string. But I don't see it actually mutating the string? Put to the line that you're worried about.

Though it probably shouldn't create the symbol there for performance reasons on large models, but that's a minor detail.

[EduardoMartinezLopez]

Ok, I was concerned about the name = @view name[....] with name on the left side of an assignment. I guess that's ok with @view macro?

[ChrisRackauckas]

that just makes name be a reference to a subset of what was previously name, no mutation there.

[AayushSabharwal]

The test doesn't work

[EduardoMartinezLopez]

@AayushSabharwal what is the problem with the test on your end?