Communication locale code — controls the language of textual data (product names, descriptions, articles, storefront UI, transactional e-mails).
Preferred format: BCP 47 language tag — language[-Script][-REGION]. Use the full tag whenever the script or region matters:
en-GB vs. en-US (British vs. American spelling)
pt-PT vs. pt-BR (European vs. Brazilian Portuguese)
zh-Hans vs. zh-Hant (Simplified vs. Traditional Chinese)
sr-Latn vs. sr-Cyrl (Latin vs. Cyrillic Serbian)
Backwards-compatible fallback: the bare two-letter ISO 639-1 code (cs, en, pl, …) is accepted indefinitely — legacy clients that only send the language subtag continue to work unchanged.
Resolution algorithm (server-side): the input is resolved against the supported locale list via the [RFC 4647 Lookup] progressive-fallback strategy — trailing subtags are stripped one by one until a supported locale is found. Example: en-GB-oxendict → en-GB → en (matched). If no subtag combination is supported, the request is rejected.
Currently supported locales: cs, en, fr, it, pl, de, sk, sv, es, zh, ja, uk, da, hu, ro, nl, pt, fi, nb, hr. Region-specific variants (e.g. en-GB, pt-BR) are accepted and resolved to their base language when the exact variant is not registered separately.