public void TestJamo() throws IOException {
{
// CLDRTransforms.registerCldrTransforms(null, ".*(Jamo).*", out);
String name = "Latin-ConjoiningJamo";
Transliterator fromLatin = Transliterator.getInstance(name);
Transliterator toLatin = Transliterator.getInstance(name, Transliterator.REVERSE);
UnicodeSet sourceSet = getRepresentativeHangul();
logln(sourceSet.size() + "\t" + sourceSet.toPattern(false));
Transliterator nfd = Transliterator.getInstance("nfd");
UnicodeSet multiply = new UnicodeSet(sourceSet);
// for (UnicodeSetIterator it = new UnicodeSetIterator(sourceSet); it.next();) {
// for (UnicodeSetIterator it2 = new UnicodeSetIterator(sourceSet); it2.next();) {
// String source1 = it.getString() + it2.getString(); // try all combinations.
// multiply.add(source1);
// }
// }
// latin.addAll(toTarget.getSourceSet())
// .addAll(toTarget.getTargetSet())
// .addAll(fromTarget.getSourceSet())
// .addAll(fromTarget.getTargetSet());
// latin.retainAll(new UnicodeSet("[[:latin:][:common:][:inherited:]]"));
// Transliterator.DEBUG = true;
UnicodeSet specials = null; // new UnicodeSet("[{ch}]");
writeFile(name, multiply, nfd, toLatin, fromLatin, true, null, specials);
}
}
private static String pretty(String source) {
StringBuilder result = new StringBuilder();
for (int i = 0; i < source.length(); ++i) {
char c = source.charAt(i);
String color =
lead.contains(c)
? "FFcccc"
: vowel.contains(c) ? "ccFFcc" : trail.contains(c) ? "ccccFF" : "FFFFFF";
result.append("<span style='background-color: #" + color + "'>" + c + "</span>");
}
return result.toString();
}
/**
* Converts the char set cset into a Unicode set uset. Equivalent to charSetToUSet.
*
* @param set Set of 256 bit flags corresponding to a set of chars.
* @param uset USet to receive characters. Existing contents are deleted.
*/
private void convert(int set[], UnicodeSet uset) {
uset.clear();
if (!initNameSetsLengths()) {
return;
}
// build a char string with all chars that are used in character names
for (char c = 255; c > 0; c--) {
if (contains(set, c)) {
uset.add(c);
}
}
}
private static UnicodeSet getRepresentativeBoundaryHangul() {
UnicodeSet resultToAddTo = new UnicodeSet();
// U+1100 ( ᄀ ) HANGUL CHOSEONG KIYEOK
// U+1161 ( ᅡ ) HANGUL JUNGSEONG A
UnicodeSet L = new UnicodeSet("[:hst=L:]");
UnicodeSet V = new UnicodeSet("[:hst=V:]");
UnicodeSet T = new UnicodeSet("[:hst=T:]");
String prefixLV = "\u1100\u1161";
String prefixL = "\u1100";
String suffixV = "\u1161";
String nullL = "\u110B";
UnicodeSet L0 = new UnicodeSet("[\u1100\u110B]");
// do all combinations of L0 + V + nullL + V
for (UnicodeSetIterator iL0 = new UnicodeSetIterator(L0); iL0.next(); ) {
for (UnicodeSetIterator iV = new UnicodeSetIterator(V); iV.next(); ) {
for (UnicodeSetIterator iV2 = new UnicodeSetIterator(V); iV2.next(); ) {
String sample = iL0.getString() + iV.getString() + nullL + iV2.getString();
String trial = Normalizer.compose(sample, false);
if (trial.length() == 2) {
resultToAddTo.add(trial);
}
}
}
}
for (UnicodeSetIterator iL = new UnicodeSetIterator(L); iL.next(); ) {
// do all combinations of "g" + V + L + "a"
final String suffix = iL.getString() + suffixV;
for (UnicodeSetIterator iV = new UnicodeSetIterator(V); iV.next(); ) {
String sample = prefixL + iV.getString() + suffix;
String trial = Normalizer.compose(sample, false);
if (trial.length() == 2) {
resultToAddTo.add(trial);
}
}
// do all combinations of "ga" + T + L + "a"
for (UnicodeSetIterator iT = new UnicodeSetIterator(T); iT.next(); ) {
String sample = prefixLV + iT.getString() + suffix;
String trial = Normalizer.compose(sample, false);
if (trial.length() == 2) {
resultToAddTo.add(trial);
}
}
}
return resultToAddTo;
}
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);
}
}
}
private static int checkLatin(PrintWriter out, Transliterator fromLatin, Transliterator toLatin) {
int errorCount = 0;
for (UnicodeSetIterator it = new UnicodeSetIterator(latin); it.next(); ) {
String source = it.getString();
String to = fromLatin.transliterate(source);
if (latin.containsSome(to)) {
String from = toLatin.transliterate(to);
String backto = toLatin.transliterate(from);
errorCount += showItems(out, false, source, to, from, backto);
}
}
return errorCount;
}
@Override
public int next() {
int current = current();
int next = rules.next();
if (next == BreakIterator.DONE) return next;
else next += workingOffset;
char c = working.current();
int following = rules.next(); // lookahead
if (following != BreakIterator.DONE) {
following += workingOffset;
if (rules.getRuleStatus() == 0 && laoSet.contains(c) && verifyPushBack(current, next)) {
workingOffset = next - 1;
working.setText(
text.getText(), text.getStart() + workingOffset, text.getLength() - workingOffset);
return next - 1;
}
rules.previous(); // undo the lookahead
}
return next;
}
static {
laoSet = new UnicodeSet("[:Lao:]");
laoSet.compact();
laoSet.freeze();
}