Andy Lester

What is the old behavior, without your patch? Does it just not recognize the code? Give an error? Also, I think the test should check for more than just one...

Cases we need to test (and you may covered some of these): `sub A { my sub B {} }` = 0 errors `sub A { sub B { }...

What are you suggesting? That we modify the policy somehow? I can certainly endorse some docs explaining where explicit `return undef` is a good idea.

Agreed, I think that makes more sense.

This has been added to `bin/perlcritic` at 90375662. Thank you.

Thanks for contacting us. Sounds like a good idea to me. Docs explaining the problem will be important, since many people don't know about BOM and encodings.

I will be glad to work with anyone who has the free time to work on this. I don't have the time myself to look into it right now.

@genehack submitted a PR which has since been deleted. The change was roughly this: ``` # Mismatched name - return $self->violation( $DESC, $EXPL, $pkg_node ); + return $self->violation( $DESC, $EXPL,...

Are there other numbers besides dates that wouldn't go with the three-digit groupings? Version numbers are going to be exempted from this rule by #856.

It seems to me that any "numbers" that aren't actually used for arithmetic would be better done as strings anyway.