Add CJK ranges for Unicode 14

Manishearth · Manishearth · commit ca3f77c7204f · 2021-03-12T08:25:02.000-08:00
diff --git a/unicodetools/.classpath b/unicodetools/.classpath
@@ -1,7 +1,7 @@
 <?xml version="1.0" encoding="UTF-8"?>
 <classpath>
 	<classpathentry kind="src" path=""/>
-	<classpathentry combineaccessrules="false" exported="true" kind="src" path="/cldr-code"/>
+	<classpathentry combineaccessrules="false" kind="src" path="/cldr-code"/>
 	<classpathentry kind="lib" path="/cldr-all/cldr-code/target/test-classes"/>
 	<classpathentry kind="con" path="org.eclipse.jdt.launching.JRE_CONTAINER/org.eclipse.jdt.internal.debug.ui.launcher.StandardVMType/JavaSE-1.8">
 		<attributes>
diff --git a/unicodetools/org/unicode/text/UCD/UCD.java b/unicodetools/org/unicode/text/UCD/UCD.java
@@ -1281,6 +1281,9 @@ public static int mapToRepresentative(int ch, int rCompositeVersion) {
             if (ch <= 0x9FFC && rCompositeVersion >= 0xd0000) {
                 return CJK_BASE;
             }
+            if (ch <= 0x9FFF && rCompositeVersion >= 0xe0000) {
+                return CJK_BASE;
+            }
             if (ch < 0xAC00) {
                 return ch;
             }
@@ -1380,13 +1383,19 @@ public static int mapToRepresentative(int ch, int rCompositeVersion) {
             if (ch <= 0x2A6DD && rCompositeVersion >= 0xd0000) {
                 return CJK_B_BASE;
             }
+            if (ch <= 0x2A6DF && rCompositeVersion >= 0xe0000) {
+                return CJK_B_BASE;
+            }
             // 2A700..2B73F; CJK Unified Ideographs Extension C
             if (rCompositeVersion >= 0x50200) {
                 if (ch <= CJK_C_BASE)
                 {
                     return ch;       // Extension C first char
                 }
-                if (ch <  CJK_C_LIMIT) {
+                if (ch <= 0x2B734) {
+                    return CJK_C_BASE;
+                }
+                if (ch <= 0x2B737 && rCompositeVersion >= 0xe0000) {
                     return CJK_C_BASE;
                 }
             }
diff --git a/unicodetools/org/unicode/text/UCD/UCD_Types.java b/unicodetools/org/unicode/text/UCD/UCD_Types.java
@@ -41,7 +41,7 @@ public interface UCD_Types {
     // 4E00;<CJK Ideograph, First>;Lo;0;L;;;;;N;;;;;
     // 9FFC;<CJK Ideograph, Last>;Lo;0;L;;;;;N;;;;;
     CJK_BASE = 0x4E00,
-    CJK_LIMIT = 0x9FFC+1, // last changed in Unicode 13
+    CJK_LIMIT = 0x9FFF+1, // last changed in Unicode 14
 
     CJK_COMPAT_USED_BASE = 0xFA0E,
     CJK_COMPAT_USED_LIMIT = 0xFA2F+1,
@@ -54,12 +54,12 @@ public interface UCD_Types {
     //20000;<CJK Ideograph Extension B, First>;Lo;0;L;;;;;N;;;;;
     //2A6DD;<CJK Ideograph Extension B, Last>;Lo;0;L;;;;;N;;;;;
     CJK_B_BASE = 0x20000,
-    CJK_B_LIMIT = 0x2A6DD+1, // last changed in Unicode 13
+    CJK_B_LIMIT = 0x2A6DF+1, // last changed in Unicode 14
 
     //2A700;<CJK Ideograph Extension C, First>;Lo;0;L;;;;;N;;;;;
     //2B734;<CJK Ideograph Extension C, Last>;Lo;0;L;;;;;N;;;;;
     CJK_C_BASE = 0x2A700,
-    CJK_C_LIMIT = 0x2B734+1,
+    CJK_C_LIMIT = 0x2B737+1, // last changed in Unicode 14
 
     //2B740;<CJK Ideograph Extension D, First>;Lo;0;L;;;;;N;;;;;
     //2B81D;<CJK Ideograph Extension D, Last>;Lo;0;L;;;;;N;;;;;

Original file line number	Diff line number	Diff line change
`@@ -1281,6 +1281,9 @@ public static int mapToRepresentative(int ch, int rCompositeVersion) {`
`1281`	`1281`	`if (ch <= 0x9FFC && rCompositeVersion >= 0xd0000) {`
`1282`	`1282`	`return CJK_BASE;`
`1283`	`1283`	`}`
	`1284`	`+ if (ch <= 0x9FFF && rCompositeVersion >= 0xe0000) {`
	`1285`	`+ return CJK_BASE;`
	`1286`	`+ }`
`1284`	`1287`	`if (ch < 0xAC00) {`
`1285`	`1288`	`return ch;`
`1286`	`1289`	`}`
`@@ -1380,13 +1383,19 @@ public static int mapToRepresentative(int ch, int rCompositeVersion) {`
`1380`	`1383`	`if (ch <= 0x2A6DD && rCompositeVersion >= 0xd0000) {`
`1381`	`1384`	`return CJK_B_BASE;`
`1382`	`1385`	`}`
	`1386`	`+ if (ch <= 0x2A6DF && rCompositeVersion >= 0xe0000) {`
	`1387`	`+ return CJK_B_BASE;`
	`1388`	`+ }`
`1383`	`1389`	`// 2A700..2B73F; CJK Unified Ideographs Extension C`
`1384`	`1390`	`if (rCompositeVersion >= 0x50200) {`
`1385`	`1391`	`if (ch <= CJK_C_BASE)`
`1386`	`1392`	`{`
`1387`	`1393`	`return ch; // Extension C first char`
`1388`	`1394`	`}`
`1389`		`- if (ch < CJK_C_LIMIT) {`
	`1395`	`+ if (ch <= 0x2B734) {`
	`1396`	`+ return CJK_C_BASE;`
	`1397`	`+ }`
	`1398`	`+ if (ch <= 0x2B737 && rCompositeVersion >= 0xe0000) {`
`1390`	`1399`	`return CJK_C_BASE;`
`1391`	`1400`	`}`
`1392`	`1401`	`}`