package cn.edu.bjut.textmining.chapter2;

import cn.edu.bjut.textmining.util.FileUtil;
import opennlp.tools.lemmatizer.DictionaryLemmatizer;

import java.io.File;
import java.nio.file.Path;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.List;

/** 使用 OpenNLP 根据“词形 + 词性”查询词元。 */
public class OpenNlpLemmatizerDemo {
    public static void main(String[] args) {
        try {
            File dictionaryFile = FileUtil.path(
                    "data", "chapter2", "en-lemmatizer.dict").toFile();
            DictionaryLemmatizer lemmatizer = new DictionaryLemmatizer(dictionaryFile);

            Path sampleFile = FileUtil.path(
                    "data", "chapter2", "lemmatization-samples.tsv");
            List<String> lines = FileUtil.readNonEmptyUtf8Lines(sampleFile);
            List<String> tokenList = new ArrayList<String>();
            List<String> tagList = new ArrayList<String>();
            for (String line : lines) {
                String[] parts = line.split("\\t");
                if (parts.length != 2) {
                    throw new IllegalArgumentException("词形还原样本格式错误：" + line);
                }
                tokenList.add(parts[0]);
                tagList.add(parts[1]);
            }

            String[] tokens = tokenList.toArray(new String[tokenList.size()]);
            String[] tags = tagList.toArray(new String[tagList.size()]);
            String[] lemmas = lemmatizer.lemmatize(tokens, tags);

            System.out.println("词形: " + Arrays.toString(tokens));
            System.out.println("词性: " + Arrays.toString(tags));
            System.out.println("词元: " + Arrays.toString(lemmas));
    
        } catch (Exception e) {
            e.printStackTrace();
        }
    }
}