The language docs at https://quarto.org/docs/authoring/language.html state "IETF language tags (BCP 47 standard), like en or en-GB" and list supported languages as prose ("...Chinese, Czech, Danish..."). Nothing tells a user writing a Chinese document which of zh / zh-TW / zh-Hant / zh-Hans will activate which UI translation. Same issue for Serbian — we ship _language-sr-Latn.yml but docs don't mention it.
This ambiguity surfaced in #14409 where a user couldn't tell whether zh-Hant or zh-TW was the "right" tag for Quarto (answer: zh-TW for our full translations, zh-Hant only for Pandoc's side).
Suggestions
We could:
- Add a table to the language docs listing each shipped translation file and the
lang: values that activate it
- Briefly explain the BCP 47 script-vs-region subtag distinction and note that both forms are valid — Quarto uses region subtags for Chinese (
zh-TW) and Portuguese (pt-BR) but script subtags for Serbian (sr-Latn), reflecting contributed files
- Note Quarto's subtag-prefix fallback (
zh-TW loads zh + zh-TW; fr-CA would load fr + fr-CA if that file existed)
The language docs at https://quarto.org/docs/authoring/language.html state "IETF language tags (BCP 47 standard), like
enoren-GB" and list supported languages as prose ("...Chinese, Czech, Danish..."). Nothing tells a user writing a Chinese document which ofzh/zh-TW/zh-Hant/zh-Hanswill activate which UI translation. Same issue for Serbian — we ship_language-sr-Latn.ymlbut docs don't mention it.This ambiguity surfaced in #14409 where a user couldn't tell whether
zh-Hantorzh-TWwas the "right" tag for Quarto (answer:zh-TWfor our full translations,zh-Hantonly for Pandoc's side).Suggestions
We could:
lang:values that activate itzh-TW) and Portuguese (pt-BR) but script subtags for Serbian (sr-Latn), reflecting contributed fileszh-TWloadszh+zh-TW;fr-CAwould loadfr+fr-CAif that file existed)