fix: generated stub of quest::ICharacterConditionType

[poirierlouis]

No description provided.

[psiberx]

The problem with that field is that it has different names in different classes. Dumper picks a random one. The objectRef is the most common name that makes sense for all classes. This is why we have include/RED4ext/Scripting/Natives/questICharacterConditionType.hpp override for generated header.

[poirierlouis]

Oh, so if I were to generate again, it would randomly pick another name?
We can keep objectRef but add a comment to let us know any changes there must be ignored?

[psiberx]

We could add it to .gitignore :) @wopss wdyt?

[wopss]

I find objectRef to make more sense than attackerRef.

We could add it to .gitignore :) @wopss wdyt?

Do you mean to add the whole file to .gitignore? If so, why?

[psiberx]

Yes. If we add it to .gitignore, the file will stay in the repo in the current state, and git will ignore future changes, when it randomly switches to attackerRef for someone.

[wopss]

If we do that, then the RED4EXT_ASSERT_SIZE might not be updated in the future. I don't see any issue with it being updated once in awhile.

[psiberx]

Good point about assert. Alternatively this case can be handled by dumper. I don't see any issue with updating the file either. But if it's not a mass update like after game patch, it's a waste of time trying to figure out why the field is suddenly changed, why SDK wasn't updated, etc. Not the first time it's happening.

[wopss]

What about skipping the stub in the generated file?

[poirierlouis]

Otherwise, I can definitely add a simple // 38 - ignore changes, don't commit (see PR#220).

You mean only keep #include and discard generated code in comments?

[psiberx]

What about skipping the stub in the generated file?

Just for this file?