DRAFT: Temporal: Add test for formatting Temporal objects with era formatter #4634
Conversation
|
Awaiting tc39/proposal-temporal#3175 |
ptomato
left a comment
ptomato
left a comment
There was a problem hiding this comment.
Thanks!
We should have similar tests for formatRange, formatToParts, formatRangeToParts, and the toLocaleString methods of PlainDate, PlainYearMonth, PlainMonthDay, PlainTime, PlainDateTime, Instant, and ZonedDateTime.
| assert(formatter.format(new Temporal.PlainYearMonth(2025, 11, "gregory")).startsWith("11"), "formatting a PlainYearMonth should work"); | ||
| assert(formatter.format(new Temporal.PlainMonthDay(11, 4, "gregory")).startsWith("11"), "formatting a PlainMonthDay should work"); |
There was a problem hiding this comment.
Using "gregory" may fail depending on the calendar that formatter gets from the host's locale. Probably the best way to work around this would be to hardcode the formatter to "en", and include locales: [en] in the frontmatter.
There was a problem hiding this comment.
s/locales/locale apparently. Sorry!
There was a problem hiding this comment.
Actually, it looks like the linter complains about including locales: [en] in the frontmatter? Or did I do that incorrectly?
There was a problem hiding this comment.
I got it wrong, it's locale https://github.com/tc39/test262/blob/main/CONTRIBUTING.md#locale
| assert(formatter.format(new Temporal.PlainDate(2025, 11, 4)).startsWith("11"), "formatting a PlainDate should work"); | ||
| assert(formatter.format(new Temporal.PlainYearMonth(2025, 11, "gregory")).startsWith("11"), "formatting a PlainYearMonth should work"); | ||
| assert(formatter.format(new Temporal.PlainMonthDay(11, 4, "gregory")).startsWith("11"), "formatting a PlainMonthDay should work"); | ||
| assert(formatter.format(new Temporal.PlainTime(14, 46)).startsWith("A"), "formatting a PlainTime should work"); |
There was a problem hiding this comment.
I'm a bit surprised by this expectation... It does match what comes out of new Date().toLocaleTimeString('en', { era: 'narrow' }), but my reading of the spec (GetDateTimeFormat) is that era isn't copied to formatOptions when constructing [[TemporalPlainTimeFormat]].
There was a problem hiding this comment.
Changed tc39/proposal-temporal#3175 (review) accordingly and updated this test.
| assert(formatter.format(new Temporal.PlainMonthDay(11, 4, "gregory")).startsWith("11"), "formatting a PlainMonthDay should work"); | ||
| assert(formatter.format(new Temporal.PlainTime(14, 46)).startsWith("A"), "formatting a PlainTime should work"); | ||
| assert(formatter.format(new Temporal.PlainDateTime(2025, 11, 4, 14, 46)).startsWith("11"), "formatting a PlainDateTime should work"); | ||
| assert(formatter.format(new Temporal.Instant(0n)).startsWith("12"), "formatting an Instant should work"); |
There was a problem hiding this comment.
This may fail depending on the host's time zone offset. Probably the best way to work around that is pass timeZone: 'UTC' in the formatter options.
There was a problem hiding this comment.
I guess I gave you bad advice here, since the test is supposed to check what happens when you pass era with no other options.
Maybe a better approach would be to check that the output of formatter.format(new Temporal.Instant(0n)) is identical to that of new Date(0).toLocaleString(["en"], { era: "narrow" })?
2 participants
See tc39/proposal-temporal#3049