This seems fine, if you want to go through the code more like you mentioned.

I think what I'd really like is to stop using **kwargs, but it gets pretty verbose to write out all parameters for the subclass constructors and decorators. Would make IDEs and type checkers smarter. That can come later though.

That one seems to have broken a behavior.

Besides that change, were there any others? Let me know when this is ready for merge.

My last commit broke a test. This is due to the fact that default=None is not equivalent to not providing default for binary flags:

• when you don't provide default, default is set to False
• when default=None, default is set to None.

I think this may be accidental behavior but it's not senseless: settings default=None you can force the prompt to require an explicit value to be provided (it'll be [y/n] rather than [y/N]).

Nonetheless, I noticed that if you provide default=None to a flag without a secondary option, the prompt message is broken: it won't include "[y/n]" at the end.

@davidism I think it's better to handle this issue in another PR eventually. Thus, I'd revert my last commit here. What do you think?

Sounds good to me, I'll merge this one after you remove that commit.

I reverted the commit, I didn't remove it. This is ready to be merged.

Besides that change, were there any others? Let me know when this is ready for merge.

I searched for "in attrs" and "in kwargs". There are a few other places where Click relies on these but they seem legit.