you created too much issues, it hard to figure out what is a problem and what is a solution, please close or group related issues in one.

Please read the PRs and stop relying on the issues... I spent a lot of times doing this cleanly so at least do a little effort on your side and click on the PR to view them...

you need to decide: you want to cooperate or want to just throw your code in us. If second I will read your issue only after priority come to it - do you want to wait one more years as you did on sourceforge?

That metrics Checks are not that obvious, so if you want me to help you, please describe problem in detail and be ready to explain your changes.

in all 3 pull request you changed the same class AbstractClassCouplingCheck, I do not understand why they did not come as one update, if problem is one or am I miss smth.

I created one issue per problem because that's how I ask all contributors on my projects to do. I sent 3 PRs because they concern different topics and you may not even agree to apply all of them. In that manner you can quickly review and apply them.

The actual change seems fine to me.

But 3 issues + 3 PRs seem strange to me, too. Maybe we should explicitly ask in the contributing guide to make several minor changes to the same line of code in one PR? And if contibutor is going to submit a PR anyway - maybe it is not necessary to open the corresponding issue (EDIT: noticed that is contradicts with practice in #44)?

By means of issue description I could be sure that problem that we are trying to resolve by means of PR is real problem that belong to that Check and it is proper way to fix it.
#44 is just latest example how PR were rejected after author explained in detail his purpose.

By pressing "Merge Pull request" your code become our team responsibility and support, do not expect that we will merge anything you or any other developer provide. We would rather reject than merge suspicious changes.

Vinsent, I propose to start your issue from scratch, sorry if we are so bureaucratic.

If pull request is not fixed in 7 fays, it is will be rejected.

I've tried writing

    public java.lang.Integer _integer = new java.lang.Integer(1);

And only this line of code prevent it from being reported. So I'm for rejecting this PR based on content. Sorry for not testing it before, when I have already suggested to accept it.