Non-first occurrence of circular class instantiation is always silently allowed

[Andarist]

🔎 Search Terms

variance class expression instantiation reference

🕗 Version & Regression Information

• This is the behavior in every version I tried

⏯ Playground Link

https://www.typescriptlang.org/play/?ts=5.6.0-dev.20240812#code/MYGwhgzhAECCA8AVaBeaA7ArgWwEYFMAnAPmgG8AoaaANzBE3wEIAuaRAbiumAAsBLEABNW0AC4BPAA74A9gDM4AOgBC8BImLEO0APS7oRQrMIRufQUIBMoyTIXK1GrTv3QIYbPmiQfuWTTeuJhi0EKy+BDoAOShRiYU3BBiYGL8wNAqqDzgUJlIhgAeYvjoQjCw2bCklNTUUmCEpWKinNwAvlztiaCQFVYFaFh4RDXcdAzMbG3UFsK20nKKsFaq6gPDBCTa5gLCNmx2S3CrTgOa2noGHl4+FUpzQtAAFGBlyo9WAJTQwaHhkRicUIxkIiWoyVS6Uy2V6eTUyHwxVK5ROVSsYzq0AaTXQLWmXGonQo3SAA

💻 Code

class A<T = number> {
  value!: T;
  child!: typeof A.B<A<T>>; // errors
  child2!: typeof A.B<A<T>>; // same as above but doesn't error

  static B = class B<T extends A = A> {
    parent!: T;
  };
}

class A2<T = number> {
  value!: T;
  child!: typeof A2.B<A2<number>>;
  child2!: typeof A2.B<A2<T>>; // same as A.child (and A.child2) but doesn't error

  static B = class B<T extends A2 = A2> {
    parent!: T;
  };
}

🙁 Actual behavior

child2 in both classes is free of errors

🙂 Expected behavior

both child2 properties should contain errors (like child in A)

Additional information about the issue

This is related to the fact that child2 reenters getVariancesWorker for the same symbol, so it returns with emptyArray that was eagerly set as links.variances on the first entry of this function. This, in turn, is a signal for structuredTypeRelatedToWorker to return Ternary.Unknown.

Since this is truly a circular situation, I think this is effectively a design limitation. I'm only reporting this for documentation purposes.

[typescript-bot]

This issue has been marked as "Not a Defect" and has seen no recent activity. It has been automatically closed for house-keeping purposes.