Compate Apache Common Lang StringUtils split performance with Google Guava Splitter split method.

StringUtilsTest.java

import com.google.common.base.Splitter;
import org.apache.commons.lang.RandomStringUtils;
import org.apache.commons.lang.math.RandomUtils;
import org.apache.commons.lang3.StringUtils;
import org.junit.AfterClass;
import org.junit.BeforeClass;
import org.junit.Test;
import org.springframework.util.StopWatch;

import java.util.Iterator;
import java.util.LinkedList;
import java.util.List;
import java.util.regex.Pattern;

import static org.junit.Assert.*;

public class StringUtilsTest
{
  // ++++++++++ 'Settings' ++++++++++
  
  // how many different Strings
  static int testDataCount = 5000;
  
  // how many "parts" in one string
  static int testDataBlockCount = 100;
  
  // the separator 
  static String separator = "1234";
  
  // fixed size of one part
  static int fixedSize = 23;
  
  // or maximum random size of a part
  static Integer randomMaxSize = null;
  
  // ++++++++++

  static List<String> testData = new LinkedList<String>();

  static List<Integer> testDataSize = new LinkedList<Integer>();

  static StopWatch stopWatch = new StopWatch();

  @BeforeClass
  public static void generateTestData() {
    for (int i = 0; i < testDataCount; i++) {
      String data = "";
      for (int j = 0; j < testDataBlockCount; j++) {
        int size = fixedSize;
        if (randomMaxSize != null) {
          size = (RandomUtils.nextInt() % randomMaxSize) + 1;
        }
        data += RandomStringUtils.random(size, "ABCDEFGHIJKLMNOPQRSTUVXYZ");
        data += separator;
      }
      data += RandomStringUtils.random(23, "ABCDEFGHIJKLMNOPQRSTUVXYZ");
      testData.add(data);
      testDataSize.add(data.split(separator).length);
    }
  }

  @AfterClass
  public static void printTestSummery() {
    System.out.println(stopWatch.prettyPrint());
  }

  @Test
  public void javaSplit() throws Exception {
    Pattern p = Pattern.compile(separator);
    stopWatch.start("javaSplit");
    Iterator<Integer> ref = testDataSize.iterator();
    for (String data : testData) {
      String[] elements = p.split(data);
      assertEquals((int) ref.next(), elements.length);
      for (String element : elements) {
        assertTrue(element.length() < testDataBlockCount);
      }
    }
    stopWatch.stop();
  }

  @Test
  public void apacheCommonLangSplit() throws Exception {
    stopWatch.start("Apache Common Lang Split");
    for (String data : testData) {
      String[] elements = StringUtils.split(data, separator);
      for (String element : elements) {
        assertTrue(element.length() < testDataBlockCount);
      }
    }
    stopWatch.stop();
  }

  @Test
  public void guavaSplitterSplit() throws Exception {
    stopWatch.start("Google Guava Splitter");
    Splitter spiltter = Splitter.onPattern(separator);
    Iterator<Integer> ref = testDataSize.iterator();
    for (String data : testData) {
      Iterable<String> elements = spiltter.split(data);
      int count = 0;
      for (String element : elements) {
        count++;
        assertTrue(element.length() < testDataBlockCount);
      }
      assertEquals((int) ref.next(), count);
    }
    stopWatch.stop();
  }
}