Fix: Allow to visit typeParameters in VariableDeclarator

[armano2]

This PR adds typeParameters to VariableDeclarator

typeParameters can be only present if kind=type

[platinumazure]

One small typo, otherwise LGTM. Thanks!

[platinumazure]

LGTM, thanks!

[armano2]

@platinumazure can i have estimate when this change can be released? its blocking me from fixing some of rules.

[platinumazure]

@armano2 Very sorry I missed your comment.

Is this PR still needed given the changes in #589 (i.e., is that PR separate/independent from this one)? Thanks!

[armano2]

this PR is needed only for estree v6, but its not needed for v7 because there was change that

type Foo = number

is no longer variable and now its own node TSTypeAliasDeclaration (kind=type got removed)
6aad31d#diff-32f897025e0e51a215a578a256455ab5

[platinumazure]

Thanks, makes sense.

In the PRs for upgrading typescript-estree to 7.x or higher, do we need to go back to visitor-keys.js and remove the override for variable declarations, given the decision to switch the node type for type declarations, if this PR is merged for [email protected]?

[armano2]

yes we will have to do that.

this property is no longer in AST, and it will be better if it's going to be removed when we upgrade to 7.x

[platinumazure]

@armano2 Understood. Sorry, I wasn't clear: If I merge this PR now, do you need to update #584 and #589 to ensure the property is removed from visitor-keys?

If so, then, given this extra work needed on your side: Should I merge this PR at all (and cut a patch release)? Or should we just close this and have people migrate to a new major version of this plugin (one which supports [email protected] or [email protected])?

I'm leaning towards just closing this, but if you think there's a strong need to fix this for people stuck on earlier versions of TS, I could merge this and do a release, and then request changes on #584 and #589. Let me know what you think the best way forward is here.

[armano2]

@platinumazure yes i will have to update them and i don't mind rebasing / removing it there

[armano2]

any progress on this?

[JamesHenry]

@armano2 We can close this in favour of #596, right? The node kind in question is no longer available