[swift] Feature/swift rules

[matifraga]

Before submitting a PR, please check that:

• The PR is submitted against master. The PMD team will merge back to support branches as needed.
• ./mvnw clean verify passes. This will build and test PMD, execute PMD and checkstyle rules. Check this for more info

PR Description:
This PR adds an XPath and a visitor rule for the Swift language.

[oowekyala]

Not a thorough review, just a couple of comments. Keep up the good work

[oowekyala]

Note to maintainers: We should really make it clear where we expect final modifiers on local vars & parameters in our codebase. I think it's just noise in short methods, especially when it pollutes a diff like that.

[matifraga]

There is a PMD rule that make finals required :P It's not enabled on this project though.

[ghost]

	1 Message
📖	No java rules are changed!

Generated by 🚫 Danger

[matifraga]

@jsotuyod Thanks for the review! we will add guidance for developers regarding the rules we added, and provide the examples. Regarding the ClassCastException antlr terminal nodes, are different entities than the rest of the nodes, so if you try to access one using the rule, you will retrieve an empty node. We now it's not the cleanest way to handle this, but a proper rule should never encounter this scenario, because each leaf node has a context with the information regarding the terminal node. We think this a good trade off given the amount of work that would require handling terminal nodes properly, and in doing so we don't have anything to gain.

[jsotuyod]

@tomidelucca @matifraga I took the liberty of refactoring how you handled terminal nodes.

I also turned the rule to a rulechain rule and found rule chain indexation was not working, so had to fix that. Please review yourselves before I merge.

@oowekyala I added a comment on AbstractRule#addRuleChainVisit(Class<? extends Node> nodeClass), as the implementation is far from ideal… this was a preexisting issue, so I'm not handling it on this PR, but we may want to think of a better way to get the XPath name of a node given the class name.

[matifraga]

Nice catch!

@jsotuyod I'm ok with the changes, we weren't aware of the rule chain related code on AbstractRule, sorry we missed that! I'm also ok with skipping the terminal nodes and throwing an exception instead of silencing it.