一、基础知识
Emoji 实际上是 UTF-8 (Unicode) 字符集上的特殊字符，多数基本 Emoji 都被分配到 Unicode 编码表 1 号平面的 U+1F300–1F6FF 和 U+1F900–1FAFF 两个区域，由2个字符组成。
肤色修饰：大多数与人相关的 Emoji 默认是黄色的，所以后来引入了五个新码点作为修饰符：
U+1F3FB、U+1F3FC、U+1F3FD、U+1F3FE、U+1F3FF。肤色修饰符追加到现有的 Emoji 后形成新的样式：
U+1F44B(👋 ) +U+1F3FD= 👋🏽符号变体或组合：一个普通的字后连接一个或多个变体、组合标识(字符)，组合形成的 Emoji ：
U+25C0+U+FE0F= ◀️
U+27A1+U+FE0F= ➡️
1+U+FE0F+U+20E3= 1️⃣国旗：每个国旗由2个地区标识符组合而成，地区标识符的对应码点范围为
U+1F1E6~U+1F1FF，等同于2个指定范围的普通 Emoji 字符组成。
U+1F1E8+U+1F1F3= 🇨🇳零宽度连接符（ZWJ）：多个基础 Emoji 通过零宽度连接符（
U+200D）形成的复杂 Emoji：
👩+U+200D+🌾= 👩🌾
👩+U+200D+❤️+U+200D+👩= 👩❤️👩
👨+U+200D+❤️+U+200D+💋+U+200D+👨= 👨❤️💋👨序列：一个基础 Emoji 加上多个标签字符 (
U+E0020~U+E007F)并以 Tag Cancel(U+E007)结尾，组合形成一个复杂 Emoji：
U+1F3F4(🏴) +U+E0067+U+E0062+U+E0065+U+E006E+U+E0067+U+E007F= 🏴特殊符号：
特殊符号只有1个字符，有些符号在某些环境下会被当做Emoj处理：⏯、⏫、⏹;
Unicode 只是约定了码点到 emoji 的映射关系，并没有约定 Emoji 图形，每个 Emoji 字体文件可以按照自己的想法设计 Emoji。
二、解决方案
- 除了一些特殊符号形式的 Emoji，其他Emoji至少有2个字符，所以先根据第二个字符类型判断是否为Emoji，使用
Character.UnicodeBlock.of和Character.getType方法判定每个字符的类型。 - 通过第二个字符类型判断当前2个字符为 Emoji 后：
1）判断是否有后续修饰
2）判断处理国旗类型；判断处理肤色修饰；判断处理 Emoji 序列标签；判断处理零宽度连接符；判断处理连续变体、组合标识；按照普通 Emoji 处理； - 处理单字符的特殊符号，这一类型内有的属于 Emoji，有的不是，目前全部简单的按照普通 Emoji 处理；
三、完整代码
package com.zpf.tool;
import java.util.List;
public class EmojiUtil {
public static boolean isEmojiNationalFlag(int codePoint) {
return codePoint >= 127462 && codePoint <= 127487;
}
// String str = new String(new int[]{0x1F44B, 0x1F3FD}, 0, 2);
public static boolean isEmojiSkinColor(int codePoint) {
return codePoint >= 127995 && codePoint <= 127999;
}
// String str = new String(new int[]{0x1F3F4, 0xE0067, 0xE0062, 0xE0065, 0xE006E, 0xE0067, 0xE007F}, 0, 7);
public static boolean isEmojiTagEnd(int codePoint) {
return codePoint == 917631;
}
public static boolean isEmojiTagSpec(int codePoint) {
return codePoint >= 917536 && codePoint <= 917630;
}
public static boolean isEmojiDecorateBlock(Character.UnicodeBlock block) {
if (block == null) {
return false;
}
return block.equals(Character.UnicodeBlock.VARIATION_SELECTORS)
|| block.equals(Character.UnicodeBlock.VARIATION_SELECTORS_SUPPLEMENT)
|| block.equals(Character.UnicodeBlock.COMBINING_HALF_MARKS)
|| block.equals(Character.UnicodeBlock.COMBINING_MARKS_FOR_SYMBOLS)
|| block.equals(Character.UnicodeBlock.COMBINING_DIACRITICAL_MARKS)
|| block.equals(Character.UnicodeBlock.COMBINING_DIACRITICAL_MARKS_SUPPLEMENT);
}
public static void pickAllEmoji(CharSequence data, StringBuilder removeResult, List<String> emojiList) {
if (removeResult == null && emojiList == null) {
return;
}
if (removeResult != null) {
removeResult.delete(0, removeResult.length());
}
if (emojiList != null) {
emojiList.clear();
}
if (data == null || data.length() == 0) {
return;
}
StringBuilder emojiBuilder = new StringBuilder();
int i = 0;
int j;
Character.UnicodeBlock block;
while (i < data.length()) {
if (i + 1 < data.length()) {
block = Character.UnicodeBlock.of(data.charAt(i + 1));
if (isEmojiDecorateBlock(block) || Character.UnicodeBlock.LOW_SURROGATES.equals(block)) {
if (i + 2 >= data.length()) {
emojiBuilder.append(data, i, i + 2);
break;
}
j = handleNationalFlag(data, i, emojiBuilder, emojiList);
if (i != j) {
i = j;
continue;
}
j = handleHumanSkin(data, i, emojiBuilder, emojiList);
if (i != j) {
i = j;
continue;
}
j = handleTagSequence(data, i, emojiBuilder, emojiList);
if (i != j) {
i = j;
continue;
}
emojiBuilder.append(data, i, i + 2);
i = handleNextChar(data, i + 2, emojiBuilder, emojiList);
continue;
}
}
recordEmoji(emojiBuilder, emojiList);
int type = Character.getType(data.charAt(i));
if (type == (int) Character.OTHER_SYMBOL) {//特殊符号一律按照Emoji处理
if (emojiList != null) {
emojiList.add(String.valueOf(data.charAt(i)));
}
} else if (removeResult != null) {
removeResult.append(data.charAt(i));
}
i++;
}
recordEmoji(emojiBuilder, emojiList);
}
private static int handleNextChar(CharSequence data, int i, StringBuilder emojiBuilder, List<String> emojiList) {
if (i >= data.length()) {
return i;
}
char nextChar = data.charAt(i);
if (nextChar == '\u200D') {//零宽度连接符
emojiBuilder.append(nextChar);
return i + 1;
}
int j = i;
Character.UnicodeBlock block;
while (j < data.length()) {
nextChar = data.charAt(j);
block = Character.UnicodeBlock.of(nextChar);
if (isEmojiDecorateBlock(block)) {
emojiBuilder.append(nextChar);
j++;
} else {
break;
}
}
if (i != j) {
recordEmoji(emojiBuilder, emojiList);
}
return j;
}
private static int handleNationalFlag(CharSequence data, int i, StringBuilder emojiBuilder, List<String> emojiList) {
int codePoint = Character.codePointAt(data, i);
if (isEmojiNationalFlag(codePoint)) {//处理国旗类型
recordEmoji(emojiBuilder, emojiList);//提交未处理
if (i + 3 < data.length()) {
codePoint = Character.codePointAt(data, i + 2);
if (isEmojiNationalFlag(codePoint)) {
emojiBuilder.append(data, i, i + 4);
recordEmoji(emojiBuilder, emojiList);
i = i + 4;
}
}
i = i + 2;
}
return i;
}
private static int handleHumanSkin(CharSequence data, int i, StringBuilder emojiBuilder, List<String> emojiList) {
if (i + 3 >= data.length()) {
return i;
}
int codePoint = Character.codePointAt(data, i + 2);
if (isEmojiSkinColor(codePoint)) {//肤色修饰
emojiBuilder.append(data, i, i + 4);
recordEmoji(emojiBuilder, emojiList);
i = i + 4;
}
return i;
}
private static int handleTagSequence(CharSequence data, int i, StringBuilder emojiBuilder, List<String> emojiList) {
if (i + 3 >= data.length()) {
return i;
}
int codePoint = Character.codePointAt(data, i + 2);
if (isEmojiTagSpec(codePoint)) {
emojiBuilder.append(data, i, i + 4);
i = i + 4;
while (i < data.length()) {
codePoint = Character.codePointAt(data, i);
if (isEmojiTagSpec(codePoint)) {
emojiBuilder.append(data, i, i + 2);
i = i + 2;
} else if (isEmojiTagEnd(codePoint)) {
emojiBuilder.append(data, i, i + 2);
recordEmoji(emojiBuilder, emojiList);
i = i + 2;
break;
} else { //error
break;
}
}
emojiBuilder.delete(0, emojiBuilder.length());
} else if (isEmojiTagEnd(codePoint)) {
emojiBuilder.append(data, i, i + 4);
recordEmoji(emojiBuilder, emojiList);
i = i + 4;
}
return i;
}
private static void recordEmoji(StringBuilder builder, List<String> emojiList) {
if (builder != null && builder.length() > 0) {
if (emojiList != null) {
emojiList.add(builder.toString());
}
builder.delete(0, builder.length());
}
}
}
2024.03.13
