Range with NaN, Infinity

[sffc]

I think we should probably forbid NaN and Infinity from taking part in a number range format. What would be your expected behavior in these cases?

new Intl.NumberFormat().formatRange(10, NaN);
new Intl.NumberFormat().formatRange(10, Infinity);

The first case makes no sense. The second case could be an interesting application of the "greater than" pattern, but that's not yet supported in ICU.

For now, better to throw a RangeError. We can remove that restriction later if we ever needed to.

[ryzokuken]

I agree that throwing a RangeError seems rather reasonable in both cases, unless infinite ranges are explicitly supported (I don't really think we need to, though, since I don't know of any applications outside of abstract mathematics, but some use cases for infinite ranges would make me change my mind).

[FrankYFTang]

I agree with the NaN should throw RangeError, but I disagree with Infinity.

d8> a = new Intl.NumberFormat()
[object Intl.NumberFormat]
d8> a.formatRange(0, Infinity)
"0–∞"
d8> a.formatRange(-Infinity, Infinity)
"-∞ – ∞"
d8> a.formatRange(-Infinity, 3)
"-∞ – 3"
d8> a.formatRange(-5 , Infinity)
"-5 – ∞"

looks reasonable to me. Why should we prohibit that?

[sffc]

My reasoning was that we might eventually want something like "5+" or "≥5" for formatRange(5,Infinity), but ICU doesn't support that yet. But, if we adopted the current behavior, we could add these other styles via an option.

[sffc]

2020-10-08: Throw on NaN; allow Infinity and -Infinity.

[sffc]

2020-10-08 discussion: https://github.com/tc39/ecma402/blob/master/meetings/notes-2020-10-08.md#range-with-nan-infinity-12