Shane F. Carr

@robertbastian I would like to merge this as-is because: 1. The final 2.0 CLI option should indeed be named `--base-language-handling` because that is how we handle all other cases where...

Need to re-do this on top of #4710

> Probably super seeded by #4836. Things this PR does that #4836 didn't do: - Added to CHANGELOG.md - Added "tlh-001" to the options test (to test an unsupported non-base...

Want to discuss with: - @Manishearth - @robertbastian - @sffc

It should be noted that the `::format` function has an error only in DateTimeFormatter, not in TypedDateTimeFormatter, and the only error it returns is a mismatched calendar. So, any additional...

> Using `core::fmt::Error` means code like `formatted_date_time.to_string()` panics, so I don't think we should ever use it. We should GIGO, in debug mode we debug-assert, but in release mode we...

The formatters share the same code path. Neo: https://github.com/unicode-org/icu4x/blob/978ed24845299cad5a58a928ef85b7b2e8928ed8/components/datetime/src/raw/neo.rs#L411 Neo Pattern: https://github.com/unicode-org/icu4x/blob/978ed24845299cad5a58a928ef85b7b2e8928ed8/components/datetime/src/format/neo.rs#L1178C6-L1178C6 Old: https://github.com/unicode-org/icu4x/blob/978ed24845299cad5a58a928ef85b7b2e8928ed8/components/datetime/src/format/datetime.rs#L78 The best we could do without an enum would be something like `(error in icu::datetime::TypedDateTimeFormatter)`

Hmm, replumbing the formating code is a fairly big change My main concern with the Display impl is code size. We could avoid this by using a different error struct...

Or are you proposing that we produce outputs such as "(icu4x error: weekday names not loaded), 6 (icu4x error: month names not loaded) 2024" (instead of "Saturday, 6 January 2024")

I realized that this is basically the same problem space as "GMT+?" (#2245) Along those lines, we could make up pithy error strings for each field that are both user...