Drop redundant `Matchers.{allOf,anyOf}` containing a single `Matcher`

[oxkitsune]

Problem

Declaring a Matcher using Matchers.{allOf,anyOf}(Matcher...) containing just a single Matcher, is redundant.

Description of the proposed new feature

• Support a stylistic preference.
• Avoid a common gotcha, or potential problem.

We can rewrite

allOf(annotations(AT_LEAST_ONE, isType("org.junit.jupiter.params.provider.MethodSource")));
anyOf(annotations(AT_LEAST_ONE, isType("org.junit.jupiter.params.provider.MethodSource")));

to:

annotations(AT_LEAST_ONE, isType("org.junit.jupiter.params.provider.MethodSource"));
annotations(AT_LEAST_ONE, isType("org.junit.jupiter.params.provider.MethodSource"));

Considerations

This should support the method overloads Matchers.{allOf,anyOf}(Iterable<? extends Matcher>).

Participation

This is a nice opportunity to get your hands dirty with BugCheckers 🚀

[rickie]

A related example of what this check would flag: #335 (comment).

[EvgheniiShipilov]

I'd like to pick this one up. Anything I should be aware of?

Update: I see something similar already existing 🤔

[rickie]

I'd like to pick this one up.

Nice!

Ah I think you found the most similar RefasterAnyOfUsage check already. Another one would be the IdentityConversion check. As that check also flags enclosing method invocations that can be dropped.

Well, actually now that I'm thinking of that and looking into the tests of that check; maybe this would be as simple as adding these exact methods to IS_CONVERSION_METHOD 🤔.

(Not sure if the naming would still be correct for IS_CONVERSION_METHOD but that's another concern).

Currently not able to try it out, but might be later / tomorrow.

[rickie]

Sorry, a little later than expected, but I tried it out.

Indeed, we can use the IS_CONVERSION_METHOD and the IdentityConversion check to solve this problem.

With this information, do you have enough context to proceed with opening a PR 😉?

[EvgheniiShipilov]

I think this should be enough, thanks for looking into it!