On Fri, 29 Jul 2022 07:40:42 GMT, Marius Hanl <mh...@openjdk.org> wrote:
>> i think *now* the code is clear without the comment!
>> (and the comment does not explain *why* it is intentional or what is the
>> intention)
>
> I would probably leave it as before. I found that code already when working
> on the OpenJFX project and while it looks strange at first glance, it was
> clear afterwards. And the comment makes this clear as well.
>
> The `== true` might confuse more now.
The combination of removing the comment in connection with adding the `== true`
has made the purpose of this line of code somewhat less clear. It needs a
comment to explain what is otherwise a pretty obscure construct. The code
relies on the fact that the body of the `assert` is executed if-and-only-if
assertions are enabled at runtime. Thus the side-effect of the assignment is
crucial to its operation. Here is a comment that might explain this:
// The following assertion check will always pass. The side-effect
// of the `assertsEnabled = true` assignment is intentional. It will
// be executed if-and-only-if assertions are enabled at runtime.
-------------
PR: https://git.openjdk.org/jfx/pull/851
