Add ValidTaxonomySlugSniff

[IanDelMar]

• Adds AbstractValidSlugSniff, which can serve as a base for sniffs that check a function parameter to ensure it is a valid slug.
• Refactors ValidPostTypeSlugSniff to extend AbstractValidSlugSniff.
• Introduces ValidTaxonomySlugSniff, which also extends AbstractValidSlugSniff.
• Adds WPReservedKeywordHelper to centrally store reserved keywords.

All sniff classes inherit the public property $exclude from AbstractFunctionRestrictionsSniff.

WPReservedKeywordHelper::$terms requires a "Last updated for WordPress x.x."

[jrfnl]

Hi @IanDelMar, welcome to WPCS and thank you for your willingness to contribute.

It is generally considered a good idea to first open an issue to discuss whether a potential new sniff is desired and what to take into account, instead of straight away creating a PR (and refactoring a whole lot of pre-existing code).

In this case, the PR, in part, solves an existing, open issue - #871 -, though I get the impression you didn't see or read that issue. Am I right ?

Having said that, I think the principle of this new sniff is a good idea and the PR is a reasonable first pass (which is not unexpected as the PR is basically reusing pre-existing code with some select search-and-replace).

I've left a lot of comments inline for you to consider.

Other than that, neither the abstract nor the new Helper class have code coverage as there are no @covers tags refering to these. This needs to be fixed (as can also be seen by the nearly 2% drop in code coverage as reported by CI).

You may also want to consider updating the commit message/PR description to actually describe what problem you are trying to solve with this new sniff.
The implementation details can be seen from the code. The use case of why this sniff is a good idea to begin with and what information/documentation/handbook rules the sniff is based on and who would benefit from this sniff should be argued in the PR description.

[jrfnl]

Please make the code samples more realistic. You've clearly done a search-and-replace on the test files of the ValidPostTypeSlug sniff, but the function signature of register_taxonomy( string $taxonomy, array|string $object_type, array|string $args = array()) is not the same as that of register_post_type( string $post_type, array|string $args = array() ), which means that the fast majority of code samples used in the tests are actually invalid function calls.

Now, some invalid function calls are to be expected in sniff tests (to stress-test the sniff), but the "normal" tests, should be valid code.

[IanDelMar]

Of course

[IanDelMar]

@jrfnl Thank you for the detailed feedback.

It is generally considered a good idea to first open an issue to discuss whether a potential new sniff is desired and what to take into account

Understood. I apologise if my actions came across as rude. May I suggest adding this information to the CONTRIBUTING.md? It would have helped me better understand the expectations.

though I get the impression you didn't see or read that issue. Am I right ?

No, you are not.

I am, of course, willing to go through all the comments and make the necessary changes to the code. However, based on this comment, I feel this PR may not be the best place to address the valid slug issue for taxonomies. My intention was to share the code already implemented in the ValidPostTypeSlugSniff across other sniffs that also check slugs. I can reconsider the approach and try something different at a later stage. Please let me know if you have any preferences regarding this or how to proceed.