%Bigint.prototype.toString% clarification

[bathos]

Initially the How “implementation-defined” is %Number.prototype.toString% meant to be? issue concerned both %Number.prototype.toString% and %BigInt.prototype.toString%, which followed a common pattern that appeared to permit a degree of behavior divergence among implementations that seemed like it might not be necessary/desirable/intentional.

@michaelficarra brought the question to an editor call which (I’m inferring) resolved to make de facto / implied behavior requirements explicit for %Number.prototype.toString%. This is being achieved by giving Number::toString a new radix parameter and updated algorithmic steps that explain the derivation of non-decimal representations, as can be seen in his corresponding PR (not yet merged at the time of writing this).

As that PR and the related discussion concerned %Number.prototype.toString% / Number::toString only, the “How implementation-defined...” issue should now be regarded as only concerning the Number case too while this new issue captures the related question of whether/how %BigInt.prototype.toString% / BigInt::toString might also be clarified/narrowed. Per @michaelficarra:

Agree, we could do something similar about %Bigint.prototype.toString%. [...] I don't think they should share an algorithm, but we can at least make a similar clarification.