private String tokenizerToString(Tokenizer tokenizer) throws Exception {
OffsetAttribute extOffset = tokenizer.addAttribute(OffsetAttribute.class);
PositionIncrementAttribute posIncrAtt =
tokenizer.addAttribute(PositionIncrementAttribute.class);
PositionLengthAttribute posLengthAtt = tokenizer.addAttribute(PositionLengthAttribute.class);
CharTermAttribute term = tokenizer.addAttribute(CharTermAttribute.class);
TypeAttribute type = tokenizer.addAttribute(TypeAttribute.class);
SemanticClassAttribute semanticClass = tokenizer.addAttribute(SemanticClassAttribute.class);
PartOfSpeechAttribute pos = tokenizer.addAttribute(PartOfSpeechAttribute.class);
StringBuilder result = new StringBuilder();
tokenizer.reset();
while (tokenizer.incrementToken() == true) {
result.append(new String(term.buffer(), 0, term.length())).append(":");
result.append(type.type()).append(":");
result.append(pos.partOfSpeech()).append(":");
result.append(semanticClass.semanticClass()).append(":");
result.append(String.valueOf(posIncrAtt.getPositionIncrement())).append(":");
result.append(String.valueOf(posLengthAtt.getPositionLength())).append(":");
result.append(String.valueOf(extOffset.startOffset())).append(":");
result.append(String.valueOf(extOffset.endOffset()));
result.append(",");
}
tokenizer.end();
return result.toString();
}
@Override
public final boolean incrementToken() throws IOException {
while (true) {
if (curTermBuffer == null) {
if (!input.incrementToken()) {
return false;
} else {
curTermBuffer = termAtt.buffer().clone();
curTermLength = termAtt.length();
curCodePointCount = charUtils.codePointCount(termAtt);
curGramSize = minGram;
tokStart = offsetAtt.startOffset();
tokEnd = offsetAtt.endOffset();
savePosIncr += posIncrAtt.getPositionIncrement();
savePosLen = posLenAtt.getPositionLength();
}
}
if (curGramSize <= maxGram) { // if we have hit the end of our n-gram size range, quit
if (curGramSize
<= curCodePointCount) { // if the remaining input is too short, we can't generate any
// n-grams
// grab gramSize chars from front or back
clearAttributes();
offsetAtt.setOffset(tokStart, tokEnd);
// first ngram gets increment, others don't
if (curGramSize == minGram) {
posIncrAtt.setPositionIncrement(savePosIncr);
savePosIncr = 0;
} else {
posIncrAtt.setPositionIncrement(0);
}
posLenAtt.setPositionLength(savePosLen);
final int charLength =
charUtils.offsetByCodePoints(curTermBuffer, 0, curTermLength, 0, curGramSize);
termAtt.copyBuffer(curTermBuffer, 0, charLength);
curGramSize++;
return true;
}
}
curTermBuffer = null;
}
}
