public String abbreviate(String firstString, String secondString) {
if (firstString.equals(secondString)) {
return firstString;
}
buffer.setLength(0);
breaker1.setText(firstString);
breaker2.setText(secondString);
// find common initial section
// we use two offset variables, in case we want to have some kind of equivalence later.
int start1 = breaker1.first();
int start2 = breaker2.first();
while (true) {
breaker1.next();
final int current1 = breaker1.current();
if (current1 == BreakIterator.DONE) {
break;
}
breaker2.next();
final int current2 = breaker2.current();
if (current2 == BreakIterator.DONE) {
break;
}
if (!firstString.regionMatches(start1, secondString, start2, current1 - start1)) {
break;
}
start1 = current1;
start2 = current2;
}
// find common initial section
int end1 = breaker1.last();
while (true) {
breaker1.previous();
final int current1 = breaker1.current();
if (current1 == BreakIterator.DONE) {
break;
}
breaker2.previous();
final int current2 = breaker2.current();
if (current2 == BreakIterator.DONE) {
break;
}
if (!firstString.regionMatches(current1, secondString, current2, end1 - current1)) {
break;
}
end1 = current1;
}
return buffer
.append(firstString.substring(0, end1))
.append(separator)
.append(secondString.substring(start2))
.toString();
}
public void doTest() {
BreakIterator brkIter;
switch (type) {
case BreakIterator.KIND_CHARACTER:
brkIter = BreakIterator.getCharacterInstance(locale);
break;
case BreakIterator.KIND_WORD:
brkIter = BreakIterator.getWordInstance(locale);
break;
case BreakIterator.KIND_LINE:
brkIter = BreakIterator.getLineInstance(locale);
break;
case BreakIterator.KIND_SENTENCE:
brkIter = BreakIterator.getSentenceInstance(locale);
break;
default:
errln("Unsupported break iterator type " + type);
return;
}
brkIter.setText(text);
int[] foundOffsets = new int[maxOffsetCount];
int offset, foundOffsetsCount = 0;
// do forwards iteration test
while (foundOffsetsCount < maxOffsetCount
&& (offset = brkIter.next()) != BreakIterator.DONE) {
foundOffsets[foundOffsetsCount++] = offset;
}
if (!offsetsMatchExpected(foundOffsets, foundOffsetsCount)) {
// log error for forwards test
String textToDisplay = (text.length() <= 16) ? text : text.substring(0, 16);
errln(
"For type "
+ type
+ " "
+ locale
+ ", text \""
+ textToDisplay
+ "...\""
+ "; expect "
+ expectOffsets.length
+ " offsets:"
+ formatOffsets(expectOffsets, expectOffsets.length)
+ "; found "
+ foundOffsetsCount
+ " offsets fwd:"
+ formatOffsets(foundOffsets, foundOffsetsCount));
} else {
// do backwards iteration test
--foundOffsetsCount; // back off one from the end offset
while (foundOffsetsCount > 0) {
offset = brkIter.previous();
if (offset != foundOffsets[--foundOffsetsCount]) {
// log error for backwards test
String textToDisplay = (text.length() <= 16) ? text : text.substring(0, 16);
errln(
"For type "
+ type
+ " "
+ locale
+ ", text \""
+ textToDisplay
+ "...\""
+ "; expect "
+ expectOffsets.length
+ " offsets:"
+ formatOffsets(expectOffsets, expectOffsets.length)
+ "; found rev offset "
+ offset
+ " where expect "
+ foundOffsets[foundOffsetsCount]);
break;
}
}
}
}
public void TestThaiDictionaryBreakIterator() {
int position;
int index;
int result[] = {1, 2, 5, 10, 11, 12, 11, 10, 5, 2, 1, 0};
char ctext[] = {
0x0041, 0x0020, 0x0E01, 0x0E32, 0x0E23, 0x0E17, 0x0E14, 0x0E25, 0x0E2D, 0x0E07, 0x0020, 0x0041
};
String text = new String(ctext);
ULocale locale = ULocale.createCanonical("th");
BreakIterator b = BreakIterator.getWordInstance(locale);
b.setText(text);
index = 0;
// Test forward iteration
while ((position = b.next()) != BreakIterator.DONE) {
if (position != result[index++]) {
errln(
"Error with ThaiDictionaryBreakIterator forward iteration test at "
+ position
+ ".\nShould have been "
+ result[index - 1]);
}
}
// Test backward iteration
while ((position = b.previous()) != BreakIterator.DONE) {
if (position != result[index++]) {
errln(
"Error with ThaiDictionaryBreakIterator backward iteration test at "
+ position
+ ".\nShould have been "
+ result[index - 1]);
}
}
// Test invalid sequence and spaces
char text2[] = {
0x0E01, 0x0E39, 0x0020, 0x0E01, 0x0E34, 0x0E19, 0x0E01, 0x0E38, 0x0E49, 0x0E07, 0x0020,
0x0E1B, 0x0E34, 0x0E49, 0x0E48, 0x0E07, 0x0E2D, 0x0E22, 0x0E39, 0x0E48, 0x0E43, 0x0E19,
0x0E16, 0x0E49, 0x0E33
};
int expectedWordResult[] = {2, 3, 6, 10, 11, 15, 17, 20, 22};
int expectedLineResult[] = {3, 6, 11, 15, 17, 20, 22};
BreakIterator brk = BreakIterator.getWordInstance(new ULocale("th"));
brk.setText(new String(text2));
position = index = 0;
while ((position = brk.next()) != BreakIterator.DONE && position < text2.length) {
if (position != expectedWordResult[index++]) {
errln(
"Incorrect break given by thai word break iterator. Expected: "
+ expectedWordResult[index - 1]
+ " Got: "
+ position);
}
}
brk = BreakIterator.getLineInstance(new ULocale("th"));
brk.setText(new String(text2));
position = index = 0;
while ((position = brk.next()) != BreakIterator.DONE && position < text2.length) {
if (position != expectedLineResult[index++]) {
errln(
"Incorrect break given by thai line break iterator. Expected: "
+ expectedLineResult[index - 1]
+ " Got: "
+ position);
}
}
// Improve code coverage
if (brk.preceding(expectedLineResult[1]) != expectedLineResult[0]) {
errln("Incorrect preceding position.");
}
if (brk.following(expectedLineResult[1]) != expectedLineResult[2]) {
errln("Incorrect following position.");
}
int[] fillInArray = new int[2];
if (((RuleBasedBreakIterator) brk).getRuleStatusVec(fillInArray) != 1 || fillInArray[0] != 0) {
errln(
"Error: Since getRuleStatusVec is not supported in DictionaryBasedBreakIterator, it should return 1 and fillInArray[0] == 0.");
}
}