Intl Era Month Code - Test Plan

[romulocintra]

Test plan for Intl era and monthCode Proposal:

• Add feature ID to each file where it is relevant #4617
• Tests for fallback behaviour in Intl.DateTimeFormat with islamic and islamic-rgsa calendars #4613
• Verify that all required calendars are supported
• Tests for other era aliases besides bc and ad (although the Ethiopic ones are possibly getting removed from CLDR?) (Added test to verify era aliases for non-gregorian calendars. #4631)
• Test that a PlainMonthDay can be created for every valid month code in a calendar
• Test month constrain behaviour in lunisolar calendars (Hebrew M05L→M06, Chinese/Dangi remove L)
• Test that no calendar except iso8601 has week numbering (Intl.Era-monthcode: tests to ensure weekofYear only set when using "iso8601" calendar  #4632)
• Basic addition/subtraction tests for each calendar (to do: Orthodox calendars, Indian, Persian)
• Basic difference tests for each calendar (to do: East Asian lunisolar calendars, Orthodox calendars, Hebrew, Indian, Persian)
• Leap month behaviour in lunisolar calendars
• Leap year behaviour in each calendar (to do: Orthodox calendars, Hijri calendars, Hebrew, Indian, Persian)
• Tests for constraining days when adding years/months, and in with()
• Era boundary behaviour in calendars with more than one era
• Tests for behaviour at end of month (Inconsistent results in DifferenceISODate? proposal-temporal#2535) (to do: Hijri calendars, Hebrew, Indian, Persian)
• Tests for TypeError and RangeError ordering in CalendarResolveFields proposal-temporal#3166 - Error behaviour in with() and from()
• Intercalary month in Orthodox calendars
• Any relevant tests that are in staging, move them out of staging (and expand as necessary, e.g. if they check only Temporal.PlainDate.from then make a similar test for every API entry point for which the test is relevant)
• Temporal tests should not test out-of-normal-range Chinese calendar values Temporal tests should not test out-of-normal-range Chinese calendar values #4614
• Pre-1582 tests for Gregorian-like calendars

cc @ben-allen @ptomato

[sffc]

Please make sure to test all the spec logic in Intl Era Month Code, especially the AOs from Temporal that have overrides, and the prose in Temporal that deals with non Gregorian calendars.

Behavior around leap months is an important aspect, but it's just one piece of the AO.

[fabon-f]

It would be also useful to test whether buddhist, japanese, and roc calendars use same month and day to iso8601 for dates prior to 1582. (afaik ICU4C doesn't do that).

cf. tc39/ecma402#1003