CLDR : Implementer’s FAQ
Q. Why has CLDR lowercased and dropped the hyphen from the ISO-3166-2 codes, e.g. “AD-02” becomes “ad02”
It was done because the identifiers have to be used in a BCP47 context, which puts limitations on the format of identifiers.
Q. Why have the subdivisions of France that are separate countries been removed (e.g. Martinique, Reunion, etc.)
The subdivisions occur in a few places in the XML data:
- a complete list with status (validity/subdivision.xml)
- the hierarchy (from ISO) (supplemental/subdivisions.xml) and Charts
- aliases and charts
- translations (common/subdivisions/)
Q. What are the codes for subdivisions based on?
The set of codes is from ISO, but stabilized. That means that we retain codes that ISO has dropped, but mark them as deprecated. This provides a stable set of codes for developers. The complete list is in validity/subdivision.xml, and the aliases (with a mapping from older to newer codes) is in Aliases
Q. Where are the subdivision translations from?
We don't do translations for these in in the survey tool because of the volume, and because it has been lesser priority.
The English has been reviewed, as have the names for England, Scotland, and Wales in main CLDR languages (the last three because of emoji
See an issue? Want to contribute? Please open a pull request at https://github.com/unicode-org/cldr-implementers-guide (You will be asked to sign the CLA when the PR is open, which is one-click with Github)