uhyo

Thanks for the suggestion! However I'm not willing to mention file paths inside comments because they can't receive enough editor support. 🥲 As an alternative you can visit https://github.com/A-Shleifman/eslint-plugin-export-scope which...

Thanks for the report! I don't remember any technical difficulty about this, so it should be just an oversight. I'm not quite sure what we should do against `Foo[runtime_value]` though......

Hello, let me +1 this and explain another use case. Recently our database server (PostgreSQL) suddenly stopped responding to any query. This resulted in existing Prisma connections being hung up...

提案ありがとうございます。structuredCloneの型をより安全にするのはよいアイデアだと思いますが、 さすがにこの実装だとパワーがありすぎるのでどんな実装ができそうか検討します🤔

ありがとうございます。cloneできないときに返り値がneverになるのは型安全性の面でまずいと考えており、structuredCloneの呼び出し自体に型エラーが発生するような定義が望ましいと考えています。（いわゆるinvalid typeが欲しくなる……） `structuredClone(val: T)` みたいな定義になっている必要がありそうです。（実現できるのかちょっとまだ検討していませんが）

こちらお待たせしました。この定義を取り入れる方向で試していますが、次のケースで想定通りの結果が出ないようです。自分も見ていますが、修正に協力いただけると助かります 😇 ```ts type B = StructuredCloneOutput; ```

Thank you for suggestion! I see that the suggested signatures make `include` much more convenient. It would indeed start to allow some mistakes (like `strings.includes(num)`), but it wouldn't deviate from...

@graphemecluster Thank you for pointing this out. Personally I wouldn't extend this to methods like `findIndex` because we need to corrently type callback parameters. Anyway the generics approach is nice...

I tacked this and couldn't get to a perfectly working implementation. 😇 As mentioned above, we should only narrow in then-branch of `include` checks but it doesn't seem currently possible....

Thank you for sthe uggestion! Yes definitely I should do that. I'll take time 🙂