Should a Korean name written in hanja divide? (朴成訓 gives family 朴成訓 under the default and locales.ZH alike) #1216

Korean names have a hanja form, which official listings print beside the hangul. Han division is opt-in (rule W1, unspaced CJK division), so by default a hanja name stays whole. locales.ZH divides it only where the surname is also on the Chinese list, and a hanja form that writes the given name in hangul is read as a given name (measured on master 5f22996):

Measured (2026-11-08) on OpenSanctions kr_assembly. Of the 299 members' hanja forms:

• Wholly Han (294): all stay whole by default. Under locales.ZH, 255 divide right and 39 stay whole. Every one of the 39 is a Korean surname the Chinese list lacks: 朴 29, 禹 2, 裵 2, and one each of 愼, 夫, 兪, 千, 魚, 具.
• Mixed (5): these mix in hangul, for a native Korean given name or syllable. Four are real (千하람, 白承아, 鄭惠경, 李주鍈); 朴沖？ is a source defect. All five read given whole, with a given-or-family report.

No issue or decisions entry covers Korean hanja. The two constructed spaced rows are not in the source. They show W1 dividing a hangul word as a whole name when it stands in the family slot of a mixed-script name: under the default given-first order, 민준 is that slot.

Proposal, in rule shape:

• When: a name written in hanja, or in hanja and hangul, is parsed under a Korean pack (locales.KO, opt-in like ZH).
• Then: it reads family-first and divides after a listed hanja surname, the census surnames' hanja forms.
• Because: Han division stays opt-in for the reason W1 gives: a Korean list reads Japanese kanji names wrong as a Chinese one does. Opting in to Korean is the caller's statement that the text is Korean.

Questions

1. Is hanja input in scope at all? It appears in official listings, while writers rarely type it.
2. A KO pack, or Korean surnames (朴 above all) added to ZH's list?
3. Should a word mixing Han and hangul (千하람) count as East Asian for rule W4 (script-scoped order), so it reads family-first without a pack?
4. Is the spaced mixed-script division (朴 민준 → family 준) a bug of its own, to file apart from this question?