Parse Tangut and Nüshu sources (#856)

unicode-org · Jun 7, 2024 · 5a4870e · 5a4870e
1 parent d51c407
commit 5a4870e
Show file tree

Hide file tree

Showing 7 changed files with 43 additions and 0 deletions.
diff --git a/unicodetools/src/main/java/org/unicode/props/PropertyParsingInfo.java b/unicodetools/src/main/java/org/unicode/props/PropertyParsingInfo.java
@@ -518,6 +518,7 @@ static void parseSourceFile(
             UcdLineParser parser = new UcdLineParser(FileUtilities.in("", fullFilename));
             if (fileName.startsWith("Unihan")
                     || fileName.startsWith("Unikemet")
+                    || (fileName.endsWith("Sources") && !fileName.startsWith("Emoji"))
                     || fileName.startsWith("k")) {
                 parser.withTabs(true);
             }

diff --git a/unicodetools/src/main/java/org/unicode/props/UcdProperty.java b/unicodetools/src/main/java/org/unicode/props/UcdProperty.java
@@ -190,13 +190,15 @@ public enum UcdProperty {
     kRSKanWa(PropertyType.Miscellaneous, "cjkRSKanWa"),
     kRSKangXi(PropertyType.Miscellaneous, "cjkRSKangXi"),
     kRSKorean(PropertyType.Miscellaneous, "cjkRSKorean"),
+    kRSTUnicode(PropertyType.Miscellaneous, "kRSTUnicode"),
     kRSUnicode(
             PropertyType.Miscellaneous,
             null,
             ValueCardinality.Ordered,
             "cjkRSUnicode",
             "Unicode_Radical_Stroke",
             "URS"),
+    kReading(PropertyType.Miscellaneous, "kReading"),
     kSBGY(PropertyType.Miscellaneous, null, ValueCardinality.Unordered, "cjkSBGY"),
     kSMSZD2003Index(PropertyType.Miscellaneous, "cjkSMSZD2003Index"),
     kSMSZD2003Readings(PropertyType.Miscellaneous, "cjkSMSZD2003Readings"),
@@ -208,9 +210,11 @@ public enum UcdProperty {
             ValueCardinality.Unordered,
             "cjkSpecializedSemanticVariant"),
     kSpoofingVariant(PropertyType.Miscellaneous, "cjkSpoofingVariant"),
+    kSrc_NushuDuben(PropertyType.Miscellaneous, "kSrc_NushuDuben"),
     kStrange(PropertyType.Miscellaneous, "cjkStrange"),
     kTGH(PropertyType.Miscellaneous, null, ValueCardinality.Unordered, "cjkTGH"),
     kTGHZ2013(PropertyType.Miscellaneous, "cjkTGHZ2013"),
+    kTGT_MergedSrc(PropertyType.Miscellaneous, "kTGT_MergedSrc"),
     kTaiwanTelegraph(PropertyType.Miscellaneous, "cjkTaiwanTelegraph"),
     kTang(PropertyType.Miscellaneous, null, ValueCardinality.Unordered, "cjkTang"),
     kTotalStrokes(PropertyType.Miscellaneous, null, ValueCardinality.Ordered, "cjkTotalStrokes"),

diff --git a/unicodetools/src/main/java/org/unicode/props/UcdPropertyValues.java b/unicodetools/src/main/java/org/unicode/props/UcdPropertyValues.java
@@ -1508,11 +1508,13 @@ public static Joining_Type_Values forName(String name) {
     // kPhonetic
     // kPrimaryNumeric
     // kPseudoGB1
+    // kReading
     // kRSAdobe_Japan1_6
     // kRSJapanese
     // kRSKangXi
     // kRSKanWa
     // kRSKorean
+    // kRSTUnicode
     // kRSUnicode
     // kSBGY
     // kSemanticVariant
@@ -1521,11 +1523,13 @@ public static Joining_Type_Values forName(String name) {
     // kSMSZD2003Readings
     // kSpecializedSemanticVariant
     // kSpoofingVariant
+    // kSrc_NushuDuben
     // kStrange
     // kTaiwanTelegraph
     // kTang
     // kTGH
     // kTGHZ2013
+    // kTGT_MergedSrc
     // kTotalStrokes
     // kTraditionalVariant
     // kUnihanCore2020

diff --git a/unicodetools/src/main/resources/org/unicode/props/ExtraPropertyAliases.txt b/unicodetools/src/main/resources/org/unicode/props/ExtraPropertyAliases.txt
@@ -167,6 +167,12 @@ cjkZhuangNumeric         ; kZhuangNumeric
 # 16.0
 cjkFanqie                ; kFanqie
 
+kTGT_MergedSrc           ; kTGT_MergedSrc
+kRSTUnicode              ; kRSTUnicode
+
+kSrc_NushuDuben          ; kSrc_NushuDuben
+kReading                 ; kReading
+
 kEH_Func                 ; kEH_Func
 kEH_FVal                 ; kEH_FVal
 kEH_UniK                 ; kEH_UniK
diff --git a/unicodetools/src/main/resources/org/unicode/props/ExtraPropertyValueAliases.txt b/unicodetools/src/main/resources/org/unicode/props/ExtraPropertyValueAliases.txt
@@ -199,6 +199,11 @@ Do_Not_Emit_Type ; Deprecated               ; Deprecated
 Do_Not_Emit_Type ; Discouraged              ; Discouraged
 Do_Not_Emit_Type ; Preferred_Spelling       ; Preferred_Spelling
 
+# @missing: 0000..10FFFF; kTGT_MergedSrc ; <none>
+# @missing: 0000..10FFFF; kRSTUnicode ; <none>
+
+# @missing: 0000..10FFFF; kSrc_NushuDuben ; <none>
+# @missing: 0000..10FFFF; kReading ; <none>
 
 # @missing: 0000..10FFFF; kEH_Func ; <none>
 # @missing: 0000..10FFFF; kEH_FVal ; <none>

diff --git a/unicodetools/src/main/resources/org/unicode/props/IndexUnicodeProperties.txt b/unicodetools/src/main/resources/org/unicode/props/IndexUnicodeProperties.txt
@@ -369,6 +369,15 @@ emoji/*/emoji-zwj-sequences; RGI_Emoji_Zwj_Sequence
 
 #emoji/*/emoji-test ; Emoji_Short_Name
 
+
+FileType ; TangutSources ; PropertyValue
+TangutSources ; kTGT_MergedSrc
+TangutSources ; kRSTUnicode
+
+FileType ; NushuSources ; PropertyValue
+NushuSources ; kSrc_NushuDuben
+NushuSources ; kReading
+
 FileType ; Unikemet ; PropertyValue
 Unikemet ; kEH_Cat
 Unikemet ; kEH_Core

diff --git a/unicodetools/src/main/resources/org/unicode/text/UCD/UnicodeInvariantTest.txt b/unicodetools/src/main/resources/org/unicode/text/UCD/UnicodeInvariantTest.txt
@@ -1090,6 +1090,20 @@ Let $japaneseSimplifiedRadicals = \p{Name=/CJK RADICAL J-SIMPLIFIED/}
 In $chineseSimplifiedRadicals, Equivalent_Unified_Ideograph ∈ [\p{kRSUnicode=/^[0-9]+'\.0$/} $radicalsWithUnifiableSimplifications]
 In $japaneseSimplifiedRadicals, Equivalent_Unified_Ideograph ∈ \p{kRSUnicode=/^[0-9]+''\.0$/}
 
+# Tangut invariants
+
+Let $tangutSourcesScope = [\p{Block=/^Tangut(.Supplement)?$/} - \p{gc=Cn}]
+$tangutSourcesScope = [ [\p{gc=Lo} & \p{sc=Tangut}] - \p{name=/^TANGUT COMPONENT-/} ]
+$tangutSourcesScope = \P{kTGT_MergedSrc=@none@}
+$tangutSourcesScope = \P{kRSTUnicode=@none@}
+
+# Nüshu invariants
+
+Let $nüshuSourcesScope = [\p{Block=Nushu} - \p{gc=Cn}]
+$nüshuSourcesScope = [\p{gc=Lo} & \p{sc=Nushu}]
+$nüshuSourcesScope = \P{kSrc_NushuDuben=@none@}
+$nüshuSourcesScope = \P{kReading=@none@}
+
 # Egyptian hieroglyph invariants
 
 Let $unikemetScope = [\p{Block=/^Egyptian.Hieroglyphs/} - \p{gc=Cn}]