Shane F. Carr

- @sffc first problem: it's unclear how clients use the resolved locale. It might be useful to figure out whether you fell back to `und` or to determine which locale...

Based on https://github.com/tc39/ecma402/issues/830#issuecomment-1711965221, there could a way for datagen to store the list of locales for which it generated data and then some API to access that list. Needs design...

Not exactly the same set of questions. This thread is about ResolvedLocale and #58 is about SupportedLocales. The solutions may overlap.

I changed https://github.com/unicode-org/icu4x/pull/4607 to be closing this issue.

- @hsivonen - In Segmenter and Collator, locales often fall back to root. Is this solution for APIs that process or consume natural language content? ICU4C retains knowledge of which...

Also see comment from @mihnita in https://github.com/unicode-org/icu4x/issues/2237#issuecomment-1201557171

I still think `retain-base-languages` should fix this, but currently it does not fill in the missing locales: ``` $ cargo run -p icu4x-datagen -- --deduplication retain-base-languages --markers collator/data@1 collator/meta@1 --locales...

- @hsivonen Assumed the thing that @sffc and @robertbastian agreed to would solve this, but seen no attention payed to since - @sffc We have since implemented SupportedLocale and ResolveLocale...

Closely related to https://github.com/unicode-org/icu4x/issues/3984. Need to decide crate assignment.

Seems like something that would be beneficial to do in 2.0. Anyone can take this and @hsivonen has left enough of a trail. Perhaps @echeran