Playing with HashMaps

gistfile1.txt

19 wikipedia
10 featured
9 free
9 more
8 page
7 articles
7 other
6 english
6 her
6 from

Webseite.java

// good read: http://www.novixys.com/blog/java-hashmap-examples/

//package webseite;

import java.io.BufferedReader;
import java.io.ByteArrayInputStream;
import java.io.IOException;
import java.io.InputStream;
import java.io.InputStreamReader;
import java.nio.charset.StandardCharsets;
import java.util.HashMap;
import org.jsoup.Jsoup;
import org.jsoup.nodes.Document;

public class Webseite {
  public static void main(String[] args) throws IOException {
    Document doc = Jsoup.connect("http://en.wikipedia.org/").get();
    HashMap<String, Integer> mapFreq = new HashMap<>();

    HashMap<String,Integer> mapNegative  = new HashMap<>();
    mapNegative.put("a", 0);
    mapNegative.put("an", 0);
    mapNegative.put("and", 0);
    mapNegative.put("as", 0);
    mapNegative.put("at", 0);
    mapNegative.put("by", 0);
    mapNegative.put("d630", 0);
    mapNegative.put("for", 0);
    mapNegative.put("in", 0);
    mapNegative.put("is", 0);
    mapNegative.put("it", 0);
    mapNegative.put("its", 0);
    mapNegative.put("of", 0);
    mapNegative.put("on", 0);
    mapNegative.put("s", 0);
    mapNegative.put("than", 0);
    mapNegative.put("that", 0);
    mapNegative.put("the", 0);
    mapNegative.put("this", 0);
    mapNegative.put("to", 0);
    mapNegative.put("was", 0);
    mapNegative.put("with", 0);

    InputStream inputStream = new ByteArrayInputStream(
        doc
            .body()
            .text()
            .toLowerCase()
            .replaceAll("([\\p{Punct}]|[/–·])+", " ")
            .replaceAll("\\s+", "\n")
            //.replaceAll("[0-9]+", "NUMBER")
            .getBytes(StandardCharsets.UTF_8));
    BufferedReader reader = new BufferedReader(new InputStreamReader(inputStream));
    String line;

    while ((line = reader.readLine()) != null) {
      if (mapFreq.containsKey(line)) {
        mapFreq.put(line, mapFreq.get(line) + 1);
      } else {
        if (! mapNegative.containsKey(line)) {
          mapFreq.put(line, 1);
        }
      }
    }
    //mapFreq.remove("NUMBER");

    mapFreq
        .entrySet()
        .stream()
        .sorted((x, y) -> y.getValue().compareTo(x.getValue()))
        .limit(10)
        .forEach(x -> System.out.println(x.getValue() + " " + x.getKey()));
  }
}