8303018: Unicode Emoji Properties

Reviewed-by: prr, erikj, rriggs

src/java.base/share/classes/java/lang/Character.java

@@ -10781,6 +10781,113 @@ class Character implements java.io.Serializable, Comparable<Character>, Constabl
         return CharacterData.of(codePoint).isIdentifierIgnorable(codePoint);
     }
 
+    /**
+     * Determines if the specified character (Unicode code point) is an Emoji.
+     * <p>
+     * A character is considered to be an Emoji if and only if it has the {@code Emoji}
+     * property, defined in
+     * <a href="https://unicode.org/reports/tr51/#Emoji_Properties_and_Data_Files">
+     * Unicode Emoji (Technical Standard #51)</a>.
+     *
+     * @param   codePoint the character (Unicode code point) to be tested.
+     * @return  {@code true} if the character is an Emoji;
+     *          {@code false} otherwise.
+     * @since   21
+     */
+    public static boolean isEmoji(int codePoint) {
+        return CharacterData.of(codePoint).isEmoji(codePoint);
+    }
+
+    /**
+     * Determines if the specified character (Unicode code point) has the
+     * Emoji Presentation property by default.
+     * <p>
+     * A character is considered to have the Emoji Presentation property if and
+     * only if it has the {@code Emoji_Presentation} property, defined in
+     * <a href="https://unicode.org/reports/tr51/#Emoji_Properties_and_Data_Files">
+     * Unicode Emoji (Technical Standard #51)</a>.
+     *
+     * @param   codePoint the character (Unicode code point) to be tested.
+     * @return  {@code true} if the character has the Emoji Presentation
+     *          property; {@code false} otherwise.
+     * @since   21
+     */
+    public static boolean isEmojiPresentation(int codePoint) {
+        return CharacterData.of(codePoint).isEmojiPresentation(codePoint);
+    }
+
+    /**
+     * Determines if the specified character (Unicode code point) is an
+     * Emoji Modifier.
+     * <p>
+     * A character is considered to be an Emoji Modifier if and only if it has
+     * the {@code Emoji_Modifier} property, defined in
+     * <a href="https://unicode.org/reports/tr51/#Emoji_Properties_and_Data_Files">
+     * Unicode Emoji (Technical Standard #51)</a>.
+     *
+     * @param   codePoint the character (Unicode code point) to be tested.
+     * @return  {@code true} if the character is an Emoji Modifier;
+     *          {@code false} otherwise.
+     * @since   21
+     */
+    public static boolean isEmojiModifier(int codePoint) {
+        return CharacterData.of(codePoint).isEmojiModifier(codePoint);
+    }
+
+    /**
+     * Determines if the specified character (Unicode code point) is an
+     * Emoji Modifier Base.
+     * <p>
+     * A character is considered to be an Emoji Modifier Base if and only if it has
+     * the {@code Emoji_Modifier_Base} property, defined in
+     * <a href="https://unicode.org/reports/tr51/#Emoji_Properties_and_Data_Files">
+     * Unicode Emoji (Technical Standard #51)</a>.
+     *
+     * @param   codePoint the character (Unicode code point) to be tested.
+     * @return  {@code true} if the character is an Emoji Modifier Base;
+     *          {@code false} otherwise.
+     * @since   21
+     */
+    public static boolean isEmojiModifierBase(int codePoint) {
+        return CharacterData.of(codePoint).isEmojiModifierBase(codePoint);
+    }
+
+    /**
+     * Determines if the specified character (Unicode code point) is an
+     * Emoji Component.
+     * <p>
+     * A character is considered to be an Emoji Component if and only if it has
+     * the {@code Emoji_Component} property, defined in
+     * <a href="https://unicode.org/reports/tr51/#Emoji_Properties_and_Data_Files">
+     * Unicode Emoji (Technical Standard #51)</a>.
+     *
+     * @param   codePoint the character (Unicode code point) to be tested.
+     * @return  {@code true} if the character is an Emoji Component;
+     *          {@code false} otherwise.
+     * @since   21
+     */
+    public static boolean isEmojiComponent(int codePoint) {
+        return CharacterData.of(codePoint).isEmojiComponent(codePoint);
+    }
+
+    /**
+     * Determines if the specified character (Unicode code point) is
+     * an Extended Pictographic.
+     * <p>
+     * A character is considered to be an Extended Pictographic if and only if it has
+     * the {@code Extended_Pictographic} property, defined in
+     * <a href="https://unicode.org/reports/tr51/#Emoji_Properties_and_Data_Files">
+     * Unicode Emoji (Technical Standard #51)</a>.
+     *
+     * @param   codePoint the character (Unicode code point) to be tested.
+     * @return  {@code true} if the character is an Extended Pictographic;
+     *          {@code false} otherwise.
+     * @since   21
+     */
+    public static boolean isExtendedPictographic(int codePoint) {
+        return CharacterData.of(codePoint).isExtendedPictographic(codePoint);
+    }
+
     /**
      * Converts the character argument to lowercase using case
      * mapping information from the UnicodeData file.

src/java.base/share/classes/java/lang/CharacterData.java

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2006, 2020, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2006, 2023, Oracle and/or its affiliates. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -38,6 +38,12 @@ abstract class CharacterData {
     abstract boolean isUnicodeIdentifierStart(int ch);
     abstract boolean isUnicodeIdentifierPart(int ch);
     abstract boolean isIdentifierIgnorable(int ch);
+    abstract boolean isEmoji(int ch);
+    abstract boolean isEmojiPresentation(int ch);
+    abstract boolean isEmojiModifier(int ch);
+    abstract boolean isEmojiModifierBase(int ch);
+    abstract boolean isEmojiComponent(int ch);
+    abstract boolean isExtendedPictographic(int ch);
     abstract int toLowerCase(int ch);
     abstract int toUpperCase(int ch);
     abstract int toTitleCase(int ch);

src/java.base/share/classes/java/lang/CharacterData00.java.template

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2003, 2021, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2003, 2023, Oracle and/or its affiliates. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -120,6 +120,30 @@ class CharacterData00 extends CharacterData {
         return ((props & $$maskIdentifierInfo) == $$valueIgnorable);
     }
 
+    boolean isEmoji(int ch) {
+        return (getPropertiesEx(ch) & $$maskEmoji) != 0;
+    }
+
+    boolean isEmojiPresentation(int ch) {
+        return (getPropertiesEx(ch) & $$maskEmojiPresentation) != 0;
+    }
+
+    boolean isEmojiModifier(int ch) {
+        return (getPropertiesEx(ch) & $$maskEmojiModifier) != 0;
+    }
+
+    boolean isEmojiModifierBase(int ch) {
+        return (getPropertiesEx(ch) & $$maskEmojiModifierBase) != 0;
+    }
+
+    boolean isEmojiComponent(int ch) {
+        return (getPropertiesEx(ch) & $$maskEmojiComponent) != 0;
+    }
+
+    boolean isExtendedPictographic(int ch) {
+        return (getPropertiesEx(ch) & $$maskExtendedPictographic) != 0;
+    }
+
     int toLowerCase(int ch) {
         int mapChar = ch;
         int val = getProperties(ch);

src/java.base/share/classes/java/lang/CharacterData01.java.template

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2003, 2021, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2003, 2023, Oracle and/or its affiliates. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -117,6 +117,30 @@ class CharacterData01 extends CharacterData {
         return ((props & $$maskIdentifierInfo) == $$valueIgnorable);
     }
 
+    boolean isEmoji(int ch) {
+        return (getPropertiesEx(ch) & $$maskEmoji) != 0;
+    }
+
+    boolean isEmojiPresentation(int ch) {
+        return (getPropertiesEx(ch) & $$maskEmojiPresentation) != 0;
+    }
+
+    boolean isEmojiModifier(int ch) {
+        return (getPropertiesEx(ch) & $$maskEmojiModifier) != 0;
+    }
+
+    boolean isEmojiModifierBase(int ch) {
+        return (getPropertiesEx(ch) & $$maskEmojiModifierBase) != 0;
+    }
+
+    boolean isEmojiComponent(int ch) {
+        return (getPropertiesEx(ch) & $$maskEmojiComponent) != 0;
+    }
+
+    boolean isExtendedPictographic(int ch) {
+        return (getPropertiesEx(ch) & $$maskExtendedPictographic) != 0;
+    }
+
     int toLowerCase(int ch) {
         int mapChar = ch;
         int val = getProperties(ch);

src/java.base/share/classes/java/lang/CharacterData02.java.template

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2003, 2021, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2003, 2023, Oracle and/or its affiliates. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -116,6 +116,30 @@ class CharacterData02 extends CharacterData {
         return ((props & $$maskIdentifierInfo) == $$valueIgnorable);
     }
 
+    boolean isEmoji(int ch) {
+        return (getPropertiesEx(ch) & $$maskEmoji) != 0;
+    }
+
+    boolean isEmojiPresentation(int ch) {
+        return (getPropertiesEx(ch) & $$maskEmojiPresentation) != 0;
+    }
+
+    boolean isEmojiModifier(int ch) {
+        return (getPropertiesEx(ch) & $$maskEmojiModifier) != 0;
+    }
+
+    boolean isEmojiModifierBase(int ch) {
+        return (getPropertiesEx(ch) & $$maskEmojiModifierBase) != 0;
+    }
+
+    boolean isEmojiComponent(int ch) {
+        return (getPropertiesEx(ch) & $$maskEmojiComponent) != 0;
+    }
+
+    boolean isExtendedPictographic(int ch) {
+        return (getPropertiesEx(ch) & $$maskExtendedPictographic) != 0;
+    }
+
     int toLowerCase(int ch) {
         int mapChar = ch;
         int val = getProperties(ch);

src/java.base/share/classes/java/lang/CharacterData03.java.template

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2020, 2021, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2020, 2023, Oracle and/or its affiliates. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -116,6 +116,30 @@ class CharacterData03 extends CharacterData {
         return ((props & $$maskIdentifierInfo) == $$valueIgnorable);
     }
 
+    boolean isEmoji(int ch) {
+        return (getPropertiesEx(ch) & $$maskEmoji) != 0;
+    }
+
+    boolean isEmojiPresentation(int ch) {
+        return (getPropertiesEx(ch) & $$maskEmojiPresentation) != 0;
+    }
+
+    boolean isEmojiModifier(int ch) {
+        return (getPropertiesEx(ch) & $$maskEmojiModifier) != 0;
+    }
+
+    boolean isEmojiModifierBase(int ch) {
+        return (getPropertiesEx(ch) & $$maskEmojiModifierBase) != 0;
+    }
+
+    boolean isEmojiComponent(int ch) {
+        return (getPropertiesEx(ch) & $$maskEmojiComponent) != 0;
+    }
+
+    boolean isExtendedPictographic(int ch) {
+        return (getPropertiesEx(ch) & $$maskExtendedPictographic) != 0;
+    }
+
     int toLowerCase(int ch) {
         int mapChar = ch;
         int val = getProperties(ch);

src/java.base/share/classes/java/lang/CharacterData0E.java.template

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2003, 2021, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2003, 2023, Oracle and/or its affiliates. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -116,6 +116,30 @@ class CharacterData0E extends CharacterData {
         return ((props & $$maskIdentifierInfo) == $$valueIgnorable);
     }
 
+    boolean isEmoji(int ch) {
+        return (getPropertiesEx(ch) & $$maskEmoji) != 0;
+    }
+
+    boolean isEmojiPresentation(int ch) {
+        return (getPropertiesEx(ch) & $$maskEmojiPresentation) != 0;
+    }
+
+    boolean isEmojiModifier(int ch) {
+        return (getPropertiesEx(ch) & $$maskEmojiModifier) != 0;
+    }
+
+    boolean isEmojiModifierBase(int ch) {
+        return (getPropertiesEx(ch) & $$maskEmojiModifierBase) != 0;
+    }
+
+    boolean isEmojiComponent(int ch) {
+        return (getPropertiesEx(ch) & $$maskEmojiComponent) != 0;
+    }
+
+    boolean isExtendedPictographic(int ch) {
+        return (getPropertiesEx(ch) & $$maskExtendedPictographic) != 0;
+    }
+
     int toLowerCase(int ch) {
         int mapChar = ch;
         int val = getProperties(ch);

src/java.base/share/classes/java/lang/CharacterDataLatin1.java.template

@@ -134,6 +134,30 @@ class CharacterDataLatin1 extends CharacterData {
         return ((props & $$maskIdentifierInfo) == $$valueIgnorable);
     }
 
+    boolean isEmoji(int ch) {
+        return (getPropertiesEx(ch) & $$maskEmoji) != 0;
+    }
+
+    boolean isEmojiPresentation(int ch) {
+        return (getPropertiesEx(ch) & $$maskEmojiPresentation) != 0;
+    }
+
+    boolean isEmojiModifier(int ch) {
+        return (getPropertiesEx(ch) & $$maskEmojiModifier) != 0;
+    }
+
+    boolean isEmojiModifierBase(int ch) {
+        return (getPropertiesEx(ch) & $$maskEmojiModifierBase) != 0;
+    }
+
+    boolean isEmojiComponent(int ch) {
+        return (getPropertiesEx(ch) & $$maskEmojiComponent) != 0;
+    }
+
+    boolean isExtendedPictographic(int ch) {
+        return (getPropertiesEx(ch) & $$maskExtendedPictographic) != 0;
+    }
+
     int toLowerCase(int ch) {
         if (ch < 'A') { // Fast path for low code points
             return ch;

src/java.base/share/classes/java/lang/CharacterDataPrivateUse.java

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2003, 2018, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2003, 2023, Oracle and/or its affiliates. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -60,6 +60,30 @@ class CharacterDataPrivateUse extends CharacterData {
         return false;
     }
 
+    boolean isEmoji(int ch) {
+        return false;
+    }
+
+    boolean isEmojiPresentation(int ch) {
+        return false;
+    }
+
+    boolean isEmojiModifier(int ch) {
+        return false;
+    }
+
+    boolean isEmojiModifierBase(int ch) {
+        return false;
+    }
+
+    boolean isEmojiComponent(int ch) {
+        return false;
+    }
+
+    boolean isExtendedPictographic(int ch) {
+        return false;
+    }
+
     int toLowerCase(int ch) {
         return ch;
     }

src/java.base/share/classes/java/lang/CharacterDataUndefined.java

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2003, 2018, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2003, 2023, Oracle and/or its affiliates. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -58,6 +58,30 @@ class CharacterDataUndefined extends CharacterData {
         return false;
     }
 
+    boolean isEmoji(int ch) {
+        return false;
+    }
+
+    boolean isEmojiPresentation(int ch) {
+        return false;
+    }
+
+    boolean isEmojiModifier(int ch) {
+        return false;
+    }
+
+    boolean isEmojiModifierBase(int ch) {
+        return false;
+    }
+
+    boolean isEmojiComponent(int ch) {
+        return false;
+    }
+
+    boolean isExtendedPictographic(int ch) {
+        return false;
+    }
+
     int toLowerCase(int ch) {
         return ch;
     }