What standard do language codes of the form "zh-Hans" belong to?

Question

Through the REST API of an application, I receive language codes of the following form: ll-Xxxx.

• two lowercase letters languages (looks like ISO 639-1),
• a dash,
• a code going up to four letters, starting with an uppercase letter (looks like an ISO 639-3 macrolanguage code).

Some examples:

az-Arab Azerbaijani in the Arabic script
az-Cyrl Azerbaijani in the Cyrillic script
az-Latn Azerbaijani in the Latin script

sr-Cyrl Serbian in the Cyrillic script
sr-Latn Serbian in the Latin script

uz-Cyrl Uzbek in the Cyrillic script
uz-Latn Uzbek in the Latin script

zh-Hans Chinese in the simplified script
zh-Hant Chinese in the traditional script

From what I found online:

[ISO 639-1] is the first part of the ISO 639 series of international standards for language codes. Part 1 covers the registration of two-letter codes.

and

ISO 639-3 is an international standard for language codes. In defining some of its language codes, some are defined as macrolanguages [...]

Now I need to write a piece of code to verify that I receive a valid language code.
But since what I receive is a mix of 639-1 (2 letters language) and 639-3 (macrolanguage), what standard am I supposed to stick with ? Are these code belonging to some sort of mixed up (perhaps common) standard ?

Answer 1

The current reference for identifying languages is IETF BCP 47, which combines IETF RFC 5646 and RFC 4647.

Codes of the form ll-Xxxx combine an ISO 639-1 language code (two letters) and an ISO 15924 script code (four letters). BCP 47 recommends that language codes be written in lower case and that script codes be written "lowercase with the initial letter capitalized", but this is basically for readability.

BCP 47 also recommends that the language code should be the shortest available ISO 639 tag. So if a language is represented in both ISO 639-1 (two letters) and ISO 639-3 (three letters), than you should use the ISO 639-1.

Answer 2

Following RFC-5646 (at page 4) a language tag can be written with the following form : [language]-[script].

• language (2 or 3 letters) is the shortest ISO 639 code
• script (4 letters) is a ISO 15924 code (see also RFC section)