Using ResourceBundle and MessageFormat for Error Messages

When generating error messages, two Java utility classes, ResourceBundle and MessageFormat, are extremely practical and powerful.  From the ResourceBundle JavaDoc:

Resource bundles contain locale-specific objects. When your program needs a locale-specific resource, a String for example, your program can load it from the resource bundle that is appropriate for the current user’s locale. In this way, you can write program code that is largely independent of the user’s locale isolating most, if not all, of the locale-specific information in resource bundles.

This allows you to write programs that can:

  • be easily localized, or translated, into different languages
  • handle multiple locales at once
  • be easily modified later to support even more locales

And from the MessageFormat JavaDoc:

MessageFormat provides a means to produce concatenated messages in a language-neutral way. Use this to construct messages displayed for end users.

MessageFormat takes a set of objects, formats them, then inserts the formatted strings into the pattern at the appropriate places.

This is an example of an error message resource bundle,

userAlreadyExists=A user already exists with the name {0}.
passwordInvalid=Please enter a valid password.

This is an example of how you would use this resource bundle.

protected static final ResourceBundle resourceBundle =

public boolean handleLogin(DynamoHttpServletRequest pReq, DynamoHttpServletResponse pRes) {


  // user already exists
  String errMsg1 = resourceBundle.getString("userAlreadyExists");
  errMsg1 = MessageFormat.format(errMsg1, userName);


  // password invalid
  String errMsg2 = resourceBundle.getString("passwordInvalid");


Enums in Java

Slide-together : now with cards on Flickr
(Photo: Slide-together : now with cards by fdecomite)

Enums are highly useful data types introduced in Java SE 5.0.  Though I love using them I often forget the exact syntax so this post is to remind me later how to use it.

public enum Example {

// create one using its name
Example myExample = Example.valueOf(“bar”.toUpperCase());

// if statement
if (myExample == Example.FOO) System.out.println(“FOO!”);

// switch statement
switch (myExample) {
  case FOO: System.out.println(“FOO!”);
  case BAR: System.out.println(“BAR!”);

// output as String using name

For further reading please see Java’s Enums guide and Enum Types (The Java™ Tutorials > Learning the Java Language > Classes and Objects).

Encode URI

To encode an URI you can simply use Java’s URLEncoder’s encode method which has been available since JDK 1.4.

String encodedUri;
  try {
    encodedUri = URLEncoder.encode(uri, "UTF-8");
  catch (UnsupportedEncodingException exc) {
    // this should never happen
    logger.warn("UTF-8 is not a supported encoding?  Not encoding for now...", exc);
    encodedUri = uri;

Dynamically generate sitemap.xml

sitemap.xml is a top level document on your website “for webmasters to inform search engines about pages on their sites that are available for crawling.”  Google not surprisingly has its own documentation on how to improve your site’s visibility using sitemap.xml.

Typically sitemap.xml is a static file that is hand generated.  But on large sites it makes more sense to generate this dynamically.  One way to do this is to generate it on demand using a servlet.  Here is my simple solution.  I did not include the implementation for outputPages() since that will be specific to each application server’s DB hierarchy or web server’s file structure.

public class SiteMap extends HttpServlet {

  protected static final String MIME_TYPE_XML = "application/xml";

  // XML tags
  protected static final String SITE_MAP_XML_INFO = "<?xml version=\"1.0\" encoding=\"UTF-8\"?>";
  protected static final String SITE_MAP_BEGIN =
  protected static final String SITE_MAP_END = "</urlset>";

  protected static final String LOC_BEGIN = " <loc>";
  protected static final String LOC_END = "</loc>";
  protected static final String PRIORITY_BEGIN = " <priority>";
  protected static final String PRIORITY_END = "</priority>";
  protected static final String URL_BEGIN = "<url>";
  protected static final String URL_END = "</url>";

  public void doGet(HttpServletRequest request, HttpServletResponse response) throws IOException {

    // set content type to be XML

    // get writer
    PrintWriter out = response.getWriter();

    // output header

    // output pages
    outputPages(request, out);

    // output end

  protected void outputPage(String uri, String priority, PrintWriter out, String urlStart) {
    out.println(LOC_BEGIN + urlStart + uri + LOC_END);
    out.println(PRIORITY_BEGIN + priority + PRIORITY_END);

Then you configure web.xml to use the SiteMap servlet.




I always forget how to sleep or wait in Java though it’s quite easy, just use the static method Thread.sleep.

For example:

    // sleep the filter's wait interval
    try {
      Thread.sleep(filter.getWaitInterval() * 1000);
    catch (InterruptedException exc) {
      logger.error("unexpected interrupt", exc);

Sun has a tutorial calling Pausing Execution with Sleep.

Removing a Cookie

To remove a cookie the Java API suggests getting the cookie, setting its maxAge to 0, and then adding that cookie to the response.  Digging around deeper I realized you also need to set the domain and the path to match the cookie’s domain and path.  Here is an example of how to do this.

    Cookie [] cookies = request.getCookies();
    for (Cookie cookie : cookies) {
      if (cookie.getName().equals(COOKIE_WE_WANT)) {

Note that if the domain was not set when the cookie was created then you should not set it when you try to remove it. Similarly with the path property. For example if the domain was not set at creation then the code would look like this:

    Cookie [] cookies = request.getCookies();
    for (Cookie cookie : cookies) {
      if (cookie.getName().equals(COOKIE_WE_WANT)) {

Also you should ensure that you add the cookie to the response before the response has already been committed.  Previously the above code was in a tag but that was too late to modify the response.  I moved this code to a filter and then it worked fine.

Finally you can do this in JavaScript. Doing it in JavaScript has the downside that it is done after the page is loaded. But it’s definitely helpful for testing. Here’s an example of deleting the cookie named “foo”.

document.cookie = 'foo=;expires='+new Date(0).toUTCString()+';';

In the above example I did not set the path or the domain. One will need to do that if the path and/or domain were set in the cookie at creation.

Unexplainable JSP Compilation Problem

I was getting a JSP compilation problem that I could not solve.

org.apache.jasper.JasperException: Unable to compile class for JSP
        at org.apache.jasper.servlet.JspServletWrapper.handleJspException(
        at org.apache.jasper.servlet.JspServletWrapper.service(
        at org.apache.jasper.servlet.JspServlet.serviceJspFile(
        at org.apache.jasper.servlet.JspServlet.service(

org.apache.jasper.JasperException: Unable to compile class for JSP
        at org.apache.jasper.JspCompilationContext.compile(
        at org.apache.jasper.servlet.JspServletWrapper.service(

java.lang.NumberFormatException: For input string: "${status.index}"
        at java.lang.NumberFormatException.forInputString(
        at java.lang.Integer.parseInt(
        at java.lang.Integer.valueOf(
        at org.apache.jasper.compiler.JspUtil.coerceToInt(
        at org.apache.jasper.compiler.Generator$GenerateVisitor.convertString(

The major problem was this was happening on the server but not locally.  The compilation problem was occurring for this line.

<uportal:module definition="${module}" moduleIndex="${status.index}"/>

The moduleIndex value is “${status.index}”. Locally when the JSP compiler encounters this it knows to call my setter for moduleIndex that takes a String parameter. But on the server the JSP compiler seemed to insist on using the setter for moduleIndex that takes an integer parameter, hence the compilation problem.

I finally posted on the Sun Forums and with the help of evnafets I came up with two solutions.

  1. The correct solution is to upgrade my application to use JSTL 1.1.  Then the JSTL expressions such as “${status.index}” will be evaluated by the container.  Then I don’t have to have a setter that takes a String parameter, the JSTL expression, and evaluate it myself.  I can just use a setter that takes an integer parameter.
  2. The quick, hack solution which is to get rid of the setter that takes an integer parameter.  Wherever the application passes in an integer parameter, I change it to pass in a String representation of that integer.

Upgrading to JSTL 1.1

Based on this excellent post I came up with the following instructions for upgrading from JSTL 1.1.

  1. Update URI in JSP pages to use JSTL 1.1.
    <%@ taglib prefix="c" uri="" %>
  2. Install in WEB-INF/lib the two JSTL 1.1 jars, standard.jar and jstl.jar.  You can get these from The Jakarta Site – Standard 1.1 Taglib Downloads.
  3. Update the start of web.xml to look like this.
    <?xml version="1.0" encoding="ISO-8859-1"?>
    <web-app xmlns=""

Do Not Return From a Try Block

Practical Java(TM) Programming Language Guide (Addison-Wesley Professional Computing Series)In Peter Hagar’s book, Practical Java, he recommends that you do not return from a try block.  This is because the finally block may change the return value.

Traditionally, programmers think that when they execute a return statement they immediately leave the method they are executing.  In Java, this is no longer true with the usage of finally

To avoid this pitfall, be sure you do not issue a return, break or continue statement inside of a try block. If you cannot avoid this, be sure the existence of a finally does not change the return value of the method. This particular problem can arise during maintenance of your code, even with careful design and implementation. Good comments and careful code reviews ward it off.

Practical Java Programming Language … – Google Book Search

URL.equals and hashCode make blocking Internet connections

Who knew that something as innocent as and hashCode would make blocking Internet connections?

The javadoc of URL.equals says: “Since hosts comparison requires name resolution, this operation is a blocking operation.”, but who reads the documentation of equals?  There is a general contract around equals.  Joshua Bloch writes in Effective Java: “Don’t write an equals that relies on unreliable resources” (Chapter 3, page 34). Hey Sun, as far as I know, the Internet is not reliable 😉

Eclipse and Java Blog by Michael Scharf: and hashCode make (blocking) Internet connections….