public static UnicodeSet getRepresentativeHangul() {
UnicodeSet extraSamples =
new UnicodeSet(
"[\uCE20{\uAD6C\uB514}{\uAD73\uC774}{\uBB34\uB837}{\uBB3C\uC5FF}{\uC544\uAE4C}{\uC544\uB530}{\uC544\uBE60}{\uC544\uC2F8}{\uC544\uC9DC}{\uC544\uCC28}{\uC545\uC0AC}{\uC545\uC2F8}{\uC546\uCE74}{\uC548\uAC00}{\uC548\uC790}{\uC548\uC9DC}{\uC548\uD558}{\uC54C\uAC00}{\uC54C\uB530}{\uC54C\uB9C8}{\uC54C\uBC14}{\uC54C\uBE60}{\uC54C\uC0AC}{\uC54C\uC2F8}{\uC54C\uD0C0}{\uC54C\uD30C}{\uC54C\uD558}{\uC555\uC0AC}{\uC555\uC2F8}{\uC558\uC0AC}{\uC5C5\uC12F\uC501}{\uC5C6\uC5C8\uC2B5}]");
UnicodeSet sourceSet = new UnicodeSet();
addRepresentativeHangul(sourceSet, 2, false);
addRepresentativeHangul(sourceSet, 3, false);
addRepresentativeHangul(sourceSet, 2, true);
addRepresentativeHangul(sourceSet, 3, true);
// add the boundary cases; we want an example of each case of V + L and one example of each case
// of T+L
UnicodeSet more = getRepresentativeBoundaryHangul();
sourceSet.addAll(more);
sourceSet.addAll(extraSamples);
return sourceSet;
}
private static void addRepresentativeHangul(
UnicodeSet resultToAddTo, int leng, boolean noFirstConsonant) {
UnicodeSet notYetSeen = new UnicodeSet();
for (char c = '\uAC00'; c < '\uD7AF'; ++c) {
String charStr = String.valueOf(c);
String decomp = Normalizer.decompose(charStr, false);
if (decomp.length() != leng) {
continue; // only take one length at a time
}
if (decomp.startsWith("ᄋ") != noFirstConsonant) {
continue;
}
if (!notYetSeen.containsAll(decomp)) {
resultToAddTo.add(c);
notYetSeen.addAll(decomp);
}
}
}