package cn.edu.bjut.chapter2;

import java.util.List;
import java.util.Properties;
import java.util.regex.Pattern;

import edu.stanford.nlp.ling.CoreAnnotations;
import edu.stanford.nlp.pipeline.Annotation;
import edu.stanford.nlp.pipeline.StanfordCoreNLP;
import edu.stanford.nlp.util.CoreMap;

public class SentenceSplitter {
	private static final String SPLITTER_REGEX = "(?<=[.!?])\\s+"; // 正则表达式规则

	public static String[] splitByRule(String text) {
		Pattern pattern = Pattern.compile(SPLITTER_REGEX);

		return pattern.split(text); // 切分句子
	}

	public static List<CoreMap> splitByStanfordCoreNLP(String text) {
		// Set up the Stanford NLP pipeline
		Properties props = new Properties();
		props.setProperty("annotators", "tokenize,ssplit");
		props.setProperty("outputFormat", "text");
		StanfordCoreNLP pipeline = new StanfordCoreNLP(props);
		
		// Create an Annotation object
		Annotation annotation = new Annotation(text); 
		// Annotate the document
		pipeline.annotate(annotation); 
		// Get sentences
		return annotation.get(CoreAnnotations.SentencesAnnotation.class); 
	}

	public static void main(String[] args) {
		String text = "Mr. Smith is here. This is a test! Isn't text mining "
				+ "amazing? Let's see e.g. this case.";

		String[] sentencesByRule = splitByRule(text);
		System.out.println("Sentences splitted by rules: ");
		for (String sentence : sentencesByRule) {
			System.out.println(sentence);
		}

		List<CoreMap> sentencesByStanfordCoreNLP = splitByStanfordCoreNLP(text);
		System.out.println("Sentences splitted by Stanford CoreNLP: ");
		for (CoreMap sentence : sentencesByStanfordCoreNLP) {
			System.out.println(sentence.toString());
		}
	}
}