static BurmeseBreakEngine()
{
// Initialize UnicodeSets
fBurmeseWordSet = new UnicodeSet();
fMarkSet = new UnicodeSet();
fBeginWordSet = new UnicodeSet();
fBurmeseWordSet.ApplyPattern("[[:Mymr:]&[:LineBreak=SA:]]");
fBurmeseWordSet.Compact();
fMarkSet.ApplyPattern("[[:Mymr:]&[:LineBreak=SA:]&[:M:]]");
fMarkSet.Add(0x0020);
fEndWordSet = new UnicodeSet(fBurmeseWordSet);
fBeginWordSet.Add(0x1000, 0x102A); // basic consonants and independent vowels
// Compact for caching
fMarkSet.Compact();
fEndWordSet.Compact();
fBeginWordSet.Compact();
// Freeze the static UnicodeSet
fBurmeseWordSet.Freeze();
fMarkSet.Freeze();
fEndWordSet.Freeze();
fBeginWordSet.Freeze();
}
static LaoBreakEngine()
{
// Initialize UnicodeSets
fLaoWordSet = new UnicodeSet();
fMarkSet = new UnicodeSet();
fBeginWordSet = new UnicodeSet();
fLaoWordSet.ApplyPattern("[[:Laoo:]&[:LineBreak=SA:]]");
fLaoWordSet.Compact();
fMarkSet.ApplyPattern("[[:Laoo:]&[:LineBreak=SA:]&[:M:]]");
fMarkSet.Add(0x0020);
fEndWordSet = new UnicodeSet(fLaoWordSet);
fEndWordSet.Remove(0x0EC0, 0x0EC4); // prefix vowels
fBeginWordSet.Add(0x0E81, 0x0EAE); // basic consonants (including holes for corresponding Thai characters)
fBeginWordSet.Add(0x0EDC, 0x0EDD); // digraph consonants (no Thai equivalent)
fBeginWordSet.Add(0x0EC0, 0x0EC4); // prefix vowels
// Compact for caching
fMarkSet.Compact();
fEndWordSet.Compact();
fBeginWordSet.Compact();
// Freeze the static UnicodeSet
fLaoWordSet.Freeze();
fMarkSet.Freeze();
fEndWordSet.Freeze();
fBeginWordSet.Freeze();
}
static ThaiBreakEngine()
{
// Initialize UnicodeSets
fThaiWordSet = new UnicodeSet();
fMarkSet = new UnicodeSet();
fBeginWordSet = new UnicodeSet();
fSuffixSet = new UnicodeSet();
fThaiWordSet.ApplyPattern("[[:Thai:]&[:LineBreak=SA:]]");
fThaiWordSet.Compact();
fMarkSet.ApplyPattern("[[:Thai:]&[:LineBreak=SA:]&[:M:]]");
fMarkSet.Add(0x0020);
fEndWordSet = new UnicodeSet(fThaiWordSet);
fEndWordSet.Remove(0x0E31); // MAI HAN-AKAT
fEndWordSet.Remove(0x0E40, 0x0E44); // SARA E through SARA AI MAIMALAI
fBeginWordSet.Add(0x0E01, 0x0E2E); //KO KAI through HO NOKHUK
fBeginWordSet.Add(0x0E40, 0x0E44); // SARA E through SARA AI MAIMALAI
fSuffixSet.Add(THAI_PAIYANNOI);
fSuffixSet.Add(THAI_MAIYAMOK);
// Compact for caching
fMarkSet.Compact();
fEndWordSet.Compact();
fBeginWordSet.Compact();
fSuffixSet.Compact();
// Freeze the static UnicodeSet
fThaiWordSet.Freeze();
fMarkSet.Freeze();
fEndWordSet.Freeze();
fBeginWordSet.Freeze();
fSuffixSet.Freeze();
}
static KhmerBreakEngine()
{
// Initialize UnicodeSets
fKhmerWordSet = new UnicodeSet();
fMarkSet = new UnicodeSet();
fBeginWordSet = new UnicodeSet();
fKhmerWordSet.ApplyPattern("[[:Khmer:]&[:LineBreak=SA:]]");
fKhmerWordSet.Compact();
fMarkSet.ApplyPattern("[[:Khmer:]&[:LineBreak=SA:]&[:M:]]");
fMarkSet.Add(0x0020);
fEndWordSet = new UnicodeSet(fKhmerWordSet);
fBeginWordSet.Add(0x1780, 0x17B3);
fEndWordSet.Remove(0x17D2); // KHMER SIGN COENG that combines some following characters
// Compact for caching
fMarkSet.Compact();
fEndWordSet.Compact();
fBeginWordSet.Compact();
// Freeze the static UnicodeSet
fKhmerWordSet.Freeze();
fMarkSet.Freeze();
fEndWordSet.Freeze();
fBeginWordSet.Freeze();
}
static CjkBreakEngine()
{
fHangulWordSet.ApplyPattern("[\\uac00-\\ud7a3]");
fHanWordSet.ApplyPattern("[:Han:]");
fKatakanaWordSet.ApplyPattern("[[:Katakana:]\\uff9e\\uff9f]");
fHiraganaWordSet.ApplyPattern("[:Hiragana:]");
// freeze them all
fHangulWordSet.Freeze();
fHanWordSet.Freeze();
fKatakanaWordSet.Freeze();
fHiraganaWordSet.Freeze();
}
public SourceTargetUtility(ITransform <string, string> transform, Normalizer2 normalizer)
{
this.transform = transform;
if (normalizer != null)
{
// synchronized (SourceTargetUtility.class) {
// if (NFC == null) {
// NFC = Normalizer2.getInstance(null, "nfc", Mode.COMPOSE);
// for (int i = 0; i <= 0x10FFFF; ++i) {
// String d = NFC.getDecomposition(i);
// if (d == null) {
// continue;
// }
// String s = NFC.normalize(d);
// if (!CharSequences.equals(i, s)) {
// continue;
// }
// // composes
// boolean first = false;
// for (int trailing : CharSequences.codePoints(d)) {
// if (first) {
// first = false;
// } else {
// TRAILING_COMBINING.add(trailing);
// }
// }
// }
// }
// }
sourceCache = new UnicodeSet("[:^ccc=0:]");
}
else
{
sourceCache = new UnicodeSet();
}
sourceStrings = new HashSet <string>();
for (int i = 0; i <= 0x10FFFF; ++i)
{
string s = transform.Transform(UTF16.ValueOf(i));
bool added = false;
if (!CharSequences.Equals(i, s))
{
sourceCache.Add(i);
added = true;
}
if (normalizer == null)
{
continue;
}
string d = NFC.GetDecomposition(i);
if (d == null)
{
continue;
}
s = transform.Transform(d);
if (!d.Equals(s))
{
sourceStrings.Add(d);
}
if (added)
{
continue;
}
if (!normalizer.IsInert(i))
{
sourceCache.Add(i);
continue;
}
// see if any of the non-starters change s; if so, add i
// for (String ns : TRAILING_COMBINING) {
// String s2 = transform.transform(s + ns);
// if (!s2.startsWith(s)) {
// sourceCache.add(i);
// break;
// }
// }
// int endOfFirst = CharSequences.onCharacterBoundary(d, 1) ? 1 : 2;
// if (endOfFirst >= d.length()) {
// continue;
// }
// // now add all initial substrings
// for (int j = 1; j < d.length(); ++j) {
// if (!CharSequences.onCharacterBoundary(d, j)) {
// continue;
// }
// String dd = d.substring(0,j);
// s = transform.transform(dd);
// if (!dd.equals(s)) {
// sourceStrings.add(dd);
// }
// }
}
sourceCache.Freeze();
}
