@kiatng can we continue w/o DeMorgan's Law?

I cannot give my approval because in my opinion, some changes are worse than the current original code.

B/c of 6 out of +100 improvements?

Dont get me wrong, i agree with your suggestions, but i dont want to mix it up with automatic fixes.

How about to merge and verify your manual changes again?

... at the other hand ... we can leave it open. With rector added as dependency we can skip it for now.

Or submit a PR to improve Rector (which I cannot do).

I'd really like to do, but i am not so familiar with rectors internal functionality.

I think i can write simple custom rules, but in this case ... there are a lot of possibly combinations of "&&", "||" that makes it difficult to make it safe for a general rule. I guess thats the reason why your issue has been closed.

The current rule works, but - as you have seen - it does only covers some "easy" patterns.

We can add this rule, but it will only cover 2 (?) out of your 6 suggestions.

this PR makes the code much harder to read, the benefit is only for the machine, not for the humans.

Disagree. Most of the changes look better to me.

• it safes a lot of lines
• how often do you look at core code?

• how often do you look at core code?

always, since the code documentation doesn't exist

and this is not better, it requires more brain processing time for no reason at all -> more possibility to misunderstand the code

It's a negated true/false expression?

So what?

because I'm stupid I think that a negated OR is far less readable. but better make the code 1 line less but much harder to read so that only geniuses will take the time to read it.

again, the code is not for the machines, it's for the people, it should be understandable easily for the people.

only because rector has that rule it doesn't mean it should be applied.

I did not say you're stupid ... you're not.

I'd like to use rector to unify code.

For core, for extensions ...

I need to call out @sreichel comment:

how often do you look at core code?

Even though it is a simple question, it is irrelevant. The question carries with it negative implications, which are not within standards of our CoC:

... we as contributors and maintainers pledge to making participation in our project and our community a harassment-free experience for everyone, regardless of ... level of experience, education, ...

and in particular the second standard:

Being respectful of differing viewpoints and experiences

Note that our CoC was based on v1.4. The latest version is 2.1. I encourage everyone here take the time to read it and think about it. The second standard in this version includes "opinions":

Being respectful of differing opinions, viewpoints, and experiences

I urge all of us use language, English as well as PHP, that complies to the best standards.

There was no "negative implications"! It was a serious question. If you disagree with that change, it is fine.

However ...

I need to call out @sreichel comment:

how often do you look at core code?

Maybe you got me wrong?

This PR would have added 7 negated return statements.

My question was how often you face these seven particular lines during development,

When you serach the codebase for return ! you'll have +700 matches, so it is a very common pattern. Decline to add 7 more b/c of "readablility" sound bit weired.