cn.hutool.extra.tokenizer.engine.analysis.AnalysisWord Maven / Gradle / Ivy

Go to download

Show more of this group Show more artifacts with this name
Show all versions of hutool-all Show documentation

Hutool是一个小而全的Java工具类库，通过静态方法封装，降低相关API的学习成本，提高工作效率，使Java拥有函数式语言般的优雅，让Java语言也可以“甜甜的”。

There is a newer version: 5.8.34

package cn.hutool.extra.tokenizer.engine.analysis;

import org.apache.lucene.analysis.tokenattributes.CharTermAttribute;
import org.apache.lucene.analysis.tokenattributes.OffsetAttribute;
import org.apache.lucene.util.Attribute;

import cn.hutool.extra.tokenizer.Word;

/**
 * Lucene-analysis分词中的一个单词包装
 * 
 * @author looly
 *
 */
public class AnalysisWord implements Word {
	private static final long serialVersionUID = 1L;

	private final Attribute word;

	/**
	 * 构造
	 * 
	 * @param word {@link CharTermAttribute}
	 */
	public AnalysisWord(CharTermAttribute word) {
		this.word = word;
	}

	@Override
	public String getText() {
		return word.toString();
	}
	
	@Override
	public int getStartOffset() {
		if(this.word instanceof OffsetAttribute) {
			return ((OffsetAttribute)this.word).startOffset();
		}
		return -1;
	}
	
	@Override
	public int getEndOffset() {
		if(this.word instanceof OffsetAttribute) {
			return ((OffsetAttribute)this.word).endOffset();
		}
		return -1;
	}

	@Override
	public String toString() {
		return getText();
	}
}