Possible bug in Lazarus 4.4?

[Thaddy]

EDIT: Thaddy missed, that the "original" code also sets Result to i if Result<0

Saw that at once ...

You saw that wrong..... My intention was to fix the logic and I did. Simply recompile Lazarus for proof. The OR was incorrect and what broke it.

If you do not recompile you will never see it is correct.
The reverse logic from Zvoni also compiles and renders the same result.
The second error was passing implicitly -1 to the call, because -1 was used as index where it should have been the index, not [result], but i.

[BrunoK]

EDIT: Thaddy missed, that the "original" code also sets Result to i if Result<0

Saw that at once ...

You saw that wrong..... My intention was to fix the logic and I did. Simply recompile Lazarus for proof. The OR was incorrect and what broke it.

You know, Thaddy, you edit all your posta to hide the erroneous code you suggest. It's a constant behaviour of yours.

[Alex.Machado]

Thank you all for your responses.

What IS actually the purpose of this Function?

According to the history in github repo, the (faulty) method present in 4.4 was introduced in order to:

ide: fixed getting deepest designer base class

So I believe this is the purpose.

BTW, I rebuilt Lazarus 4.4 with the new function in trunk (as mentioned by @Bytebites) and it works, or at least doesn't give me any exceptions.

Cheers,