@@ -4715,17 +4715,20 @@ void UnicodeSetTest::TestToPatternOutput() {
47154715 {u" [{a}-z]" , u" [a-z]" },
47164716 {uR"( [\N{PRESENTATION FORM FOR VERTICAL RIGHT WHITE LENTICULAR BRAKCET}])" , u" [︘]" },
47174717 {uR"( [\N{bell}])" , u" [🔔]" },
4718- // Ill-formed in ICU 78 and earlier:
4718+ // Ill-formed in ICU 78 and earlier, made well-formed by ICU-23350 :
47194719 {uR"( [\N{PRESENTATION FORM FOR VERTICAL RIGHT WHITE LENTICULAR BRACKET}])" , u" [︘]" },
4720- // Loose matching: most of these were ill-formed in ICU 78 and earlier, and were made
4721- // well-formed by ICU-3736.
4720+ // Loose matching: These were ill-formed in ICU 78 and earlier, and were
4721+ // made well-formed by ICU-3736.
47224722 {uR"( [\N{Latin small ligature o-e}])" , u" [œ]" },
47234723 {uR"( [\N{Hangul jungseong O-E}])" , u" [ᆀ]" },
47244724 {uR"( [\N{Hangul jungseong O -E}])" , u" [ᆀ]" },
47254725 {uR"( [\N{Hangul jungseong OE}])" , u" [ᅬ]" },
4726- {uR"( [\N{TIBETAN LETTER -A}])" , u" [འ]" },
4726+ {uR"( [\N{Tibetan letter -a}])" , u" [འ]" },
4727+ {uR"( [\N{Tibetan letter - a}])" , u" [འ]" },
47274728 {uR"( [\N{TIBETAN_LETTER_-A}])" , u" [འ]" },
47284729 {uR"( [\N{TIBETAN LETTER-A}])" , u" [ཨ]" },
4730+ {uR"( [\N{Tibetan mark BKA- SHOG YIG MGO}])" , u" [༊]" },
4731+ {uR"( [\N{Tibetan mark BKA -SHOG-YIG-MGO}])" , u" [༊]" },
47294732 {uR"( [\N{CJK UNIFIED IDEOGRAPH-55B5}])" , u" [喵]" },
47304733 {uR"( [\N{CJK unified ideograph 5-5-b-5}])" , u" [喵]" },
47314734 {uR"( [{\N{Hangul syllable YA}\N{Hangul syllable ONG}}])" , u" [{야옹}]" },
@@ -4851,7 +4854,12 @@ void UnicodeSetTest::TestParseErrors() {
48514854 u" [:^Noncharacter_Code_Point≠No:]" ,
48524855 // This should be [\a]; tracked by ICU-8963.
48534856 uR"( [\N{BEL}])" ,
4857+ // The leading hyphen does not match the medial hyphen in the real character name.
48544858 uR"( [\N{CJK UNIFIED IDEOGRAPH -55B5}])" ,
4859+ // A medial hyphen does not match the trailing hyphen in BKA-.
4860+ uR"( [\N{Tibetan mark BKA-SHOG-YIG-MGO}])" ,
4861+ // With -- in the query, neither hyphen is medial, and two hyphens do not match one.
4862+ uR"( [\N{Tibetan mark BKA--SHOG-YIG-MGO}])" ,
48554863 }) {
48564864 UErrorCode errorCode = U_ZERO_ERROR;
48574865 const UnicodeSet set (expression, errorCode);
0 commit comments