feat: Add function to validate an identifier

[wraithgar]

Cleanup and linting of #351
Closes #350

Authored by @boxcee

[nlf]

the functionality here looks good. i do think the function name identifier is a little vague, but i don't have any alternatives to suggest so i'm going to approve this as-is

[wraithgar]

the codebase calls this a prereleaseidentifier which is quite a mouthful

[ljharb]

The bin (which is user-exposed, unlike the codebase) calls it --preid - perhaps preID as short for preleaseIdentifier?

[wraithgar]

Looking a little deeper at this, the impetus was #349 which this PR doesn't address. The line we've been using on these PRs is that if they were ready to go we could land them, but if they needed more work we'd close them and hope that folks who wanted the feature would revisit the work. I'm gonna close this in light of #349 which is the real bug to be fixed here.

[ljharb]

@wraithgar i have a local branch ready to fix #349 once this lands :-/

[wraithgar]

Can you implement it without having to export a new function?

[ljharb]

Certainly I can, but then consumers won't have an easily accessible predicate function to help explain why they got null. What's the argument against adding this?

[boxcee]

As I understand the closing comment, this PR needs more work. I cannot see any requests to change anything though, except perhaps the naming.

[wraithgar]

In rebasing old PRs the line was drawn at "is this ready" and since it didn't address the underlying issue and folks wanted to bikeshed the name it was deemed "not ready" and closed.

If there were someone actively making a new PR and responding to feedback it would likely be landed.