My IT world

Distributed Transactions Failure Simulation

2012-04-23T03:13:00.000-07:00

If you have many resources which are needed to be updated as one atomic operation, probably you are using XA transactions. For example, you need to change data in database and send notification about this change to external system. Important point to note - db change and notification doesn't make sense in case one of them fails for some reason. Everything could fail, we leaving in not ideal world. The question is how to test the worst case - resource failure and be sure that system is in predictable state before and after transaction. There are could be different approaches for this, but in this blog entry I'm going to describe approach I found really interesting. As for me it's one of the best approaches to ensure that transactions in application are really bullet proof. I'm talking here about "failure injection". "Injection" means that you can declaratively enforce some method to throw exception, without changing any code from service layer. It's really close to simulating real failure. All this possible with JBoss Bytemann (http://www.jboss.org/byteman). This Java agent can do much more, but it's out of scope of this post. They have support for jUnit, a special annotation which will enforce method you'll name to throw an exception. In my case I've played with db and jms XA transaction with Atomikos as transaction manager. As for me its good playground - H2 in-memory database and Apache ActiveMQ embedded in-memory broker. Friendly speaking I'm not fan of jms mocks, why do we need them if there is fully functional real JMS broker, just in memory. You even don't need to start and stop it manually. Let's take mockrunner (http://mockrunner.sourceforge.net/examplesjms.html)-

A few notes on the JMS implementation: The JMS test module is not a full blown message server, but it supports all necessary functionality to test JMS code. The implementation breaks the JMS specification in some cases. Messages are immediately forwarded to a proper receiver, if possible. Otherwise they're stored for later examination. There's no real transaction support, i.e. the framework keeps track if a transaction is committed or rolled back, but does not guarantee transaction atomicity. If you send a message in a transaction, it will be forwarded to the receiver, even if you rollback the transaction.

As you can see some special tricks are required in case of mockrunner, since it haven't atomicity guarantees. I should say that there are some tricky parts in Bytemann and Spring integration. Both of them are using @RunWith annotation for their internal infrastructure setup. Pity thing is that you can't use multiply @RunWith declarations in jUnit. So, we need to have both - @RunWith(BMUnitRunner.class) and @RunWith(SpringJUnit4ClassRunner.class)... oops. Not all so bad. It's possible to eliminate SpringJUnit4ClassRunner usage. Yep, it's not universal solution for multiply jUnit runners, but it's working in this case. To do the same as Spring class runner you can say this:

@Before
    public void setUpContext() throws Exception {
        this.testContextManager = new TestContextManager(getClass());
        this.testContextManager.prepareTestInstance(this);
}

From docs -

TestContextManager is the main entry point into the Spring TestContext Framework, which provides support for loading and accessing ApplicationContext application contexts, dependency, injection of test instances, transactional execution of test methods, etc.

So, final test looks like this:

@Test(expected = Exception.class)
@Transactional
@BMRule(name="Database goes down",
isInterface = true,
targetClass = "com.blogspot.ostas.lora.database.IUserDao",
targetMethod = "save",
action = "org.apache.log4j.Logger.getLogger(getClass())
.debug(\"DB failure simulation\");
throw new java.lang.RuntimeException(\"Simulated Database Failure\")")
public void save(){
   userService.saveAndNotify(user);
    try {
            LOGGER.info("Transaction status : "
            +jtaTransactionManager.getTransactionManager().getStatus());
        } catch (SystemException e) {
            LOGGER.error(e);
        }
    }

Here I'm enforcing DAO component of my service layer to throw runtime exception - it's simulated resource failure. Also, after this I'm checking db state using JDBC to be 100% sure that noting went wrong in case of JMS failure. As you can see, there is no changes in service layer code, no mocks, etc. Thanks for reading. Complete source code of my pet project you can find here: https://github.com/daoway/AtomikosSandbox

Atomikos distributed transactions scalability

2012-02-02T03:23:00.000-08:00

At this moment I'm working on the project where we using distributed transactions. The number of XA Resources in our case is not big at all, but for me is very interesting to know how Atomikos can scale when the number of XA Resources is large. That is why you are reading this blog post. I've desided to push this pet project on GitHub for you to be able to test/play/join the effort, etc. - https://github.com/daoway/AtomikosDistributedScalability

What do I have... I have Atomikos 3.7.1, Spring 3.1 (if you are still using 3.0 consider migration to 3.1 - it's better) and Hibernate. In this particular case I've decided to use embedded in memory H2 database to eliminate possible network overhead and to simplify setup of underlying farm of XA Resources. It's possible because H2 has XA compatible data source implementation. BTW, it can be used also for unit/integration testing instead of Oracle XA Datasource to test transactional behavior for example.

So, I'm playing with H2 XA Army. I have a lot of in-memory H2 database instances, one simple domain object and Hibernate. I want to persist this object across all nodes in XA transaction and calculate time for this effort. Then increase the number of XA resources for some Delta and repeat this procedure until I'll get OutOfMemoryException. Just kidding :) All this stuff is in-memory and RAM amount is limited, for sure, the maximum of XA resource I've played with was 2000.

Here is the result of my experiment (click on link to see Adobe Flex chart with test results) :

As you can see, replication time grows linear, together with number of XA resources participating in "distributed" transaction. If so, we can conclude that Atomikos has linear scalability.

Thanks for reading. Any comments/suggestions are welcomed.

Loading Spring Beans based on environment

2011-05-25T09:16:00.000-07:00

In complex deployment environments it makes sense to keep web application monolith war file and provide appropriate instructions for deployment team. It's possible only if your application have flexible configurability strategy, so that at least you can switch from DEV to UAT, from UAT to PROD environments and be sure that only thing has been changed is environment specific stuff. Forget about unpacking and patching war-file, it could be prohibited at all because of non-trivial distribution mechanism.

For Spring-based applications all this could look a little bit scary for a first look, you have your applicationContext.xml with placeholders, which is ok, or JNDI for enviromnent-specific resource lookup. Great, but what you are going to do if some of the parts of resources are present in one env. and absent in another ? For example +1 in dev, -1 in uat and +1 in prod ? Maybe there is not only one way to do it, but I came across the following solution for this.

It's based on combination of servlet context init params and Spring bean re-definition at runtume. Thanks to Spring it's possible. In terms of configurability servlet init params is very flexible thing. You can define one in web.xml, and then redefine it in context setup. I mean externally, so, your war file is still don't need to be patched. Other stuff is technical details for this - key point to implement this is Spring bean which implements both ServletContextAware and ApplicationContextAware interfaces. First makes possible to access Servlet context, second - Spring application context. Runtime re-defintion is not so hard - take a look here.

Complete source code you can find here.

Java code block in JIRA

2011-04-15T01:39:00.000-07:00

This is just small hint on JIRA usage.

If you want to post some source code example in JIRA issue, please use code blocks ! Other vice you'll have a chance to see fancy smiles instead of certain pieces of code. I saw it few times, very annoying...

It's not hard at all:

{code} public static void main(String[] args)  {code}

Other languages are also supported.

Precompile JSP with Maven 2

2011-02-25T03:21:00.000-08:00

There are situations where JDK is banned in production and only JRE allowed for use. In this case if you are using JSP there is only one option to use - precompile JSP. Here is example how to achieve that. Nothing very special, but I guess this can save some time and effort for somebody )

<plugin>
         <groupId>org.codehaus.mojo</groupId>
         <artifactId>jspc-maven-plugin</artifactId>
         <executions>
             <execution>
                 <id>jspc</id>
                 <goals>
                     <goal>compile</goal>
                 </goals>
             </execution>
         </executions>
     </plugin>
  
     <plugin>
         <groupId>org.apache.maven.plugins</groupId>
         <artifactId>maven-war-plugin</artifactId>
         <configuration>
             <webXml>${basedir}/target/jspweb.xml</webXml>
       <warSourceExcludes>**/*.jsp</warSourceExcludes>
         </configuration>
     </plugin>

WarSourceExcludes will not package original JSPs - really makes no sence since they are just classes and appropriate servlet mappins in web.xml now.

Mock JNDI in jUnit

2011-02-19T07:17:00.000-08:00

JNDI is very useful technology. But... not everywhere. If there is the need to deploy your application on variety of different environments, for example databases with sensitive data and you are not a person who is suitable to deploy application for a bunch of reasons - segregation of concepts, separate production team, etc - JNDI is what you need. If you are with JNDI, only thing you have to provide is JNDI names with example config, so that person responsible for deployment could figure out how to deal with it. Just imagine the mess when you need to deploy war file for ~10 different Oracle databases and the same number of WebSphere MQ queues without JNDI. Think for 1 min :)

But, in development is not so pleasant to have a deal with JNDI, since container required in this case to perform actual lookup of resources. From the other point of view not all so bad, since there is a way in 10 lines of code to mock required JNDI datasource. In fact it could be any other resource, but data source seems to be the most popular.

Key point here is the use of the calss from SpringSource -
org.springframework.mock.jndi.SimpleNamingContextBuilder

Here is the complete example how to use it -

import com.blogspot.ostas.cxf.SimpleDao;
import oracle.jdbc.pool.OracleDataSource;
import org.apache.log4j.Logger;
import org.junit.BeforeClass;
import org.junit.Test;
import org.junit.runner.RunWith;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.mock.jndi.SimpleNamingContextBuilder;
import org.springframework.test.context.ContextConfiguration;
import org.springframework.test.context.junit4.SpringJUnit4ClassRunner;

import javax.naming.NamingException;
import java.sql.SQLException;

import static org.junit.Assert.assertEquals;
import static org.junit.Assert.assertNotNull;

@RunWith(SpringJUnit4ClassRunner.class)
@ContextConfiguration(locations={"classpath:/WEB-INF/dataAccess-applicationContext.xml"})
public class DaoTestCase{
static Logger logger = Logger.getLogger(DaoTestCase.class.getName());
@BeforeClass
public static void setUp(){
    OracleDataSource ods = null;
    try {
        ods = new OracleDataSource();
    } catch (SQLException e) {
        logger.error(e);
    }
    ods.setURL("jdbc:oracle:thin:@127.0.0.1:1521/orcldb");
    ods.setUser("*");
    ods.setPassword("*");
    SimpleNamingContextBuilder builder = null;
    try {
        builder = SimpleNamingContextBuilder.emptyActivatedContextBuilder();
        builder.bind("java:comp/env/jdbc/db",ods);
    } catch (NamingException e) {
        logger.error(e);
    }
}
@Autowired
SimpleDao simpleDao;
@Test
public void testInjectBean() throws Exception
{
    logger.debug("Bean : >>> " + simpleDao);
    assertNotNull(simpleDao);
}
@Test
public void testDBconnection() throws Exception{
    assertEquals(simpleDao.getNumber(),1);
}
}

As you can see from this code, actual mock is just

SimpleNamingContextBuilder builder = null;
try {
  builder = SimpleNamingContextBuilder.emptyActivatedContextBuilder();
  builder.bind("java:comp/env/jdbc/db",ods);
 } catch (NamingException e) {
    logger.error(e);
}

ods - just plain DataSource. From Spring point jdbc/db looks llike this -

<bean id="dataSource"
class="org.springframework.jndi.JndiObjectFactoryBean">
<property name="jndiName">
    <value>java:comp/env/jdbc/db</value>
</property>
</bean>

The last thing I have to mention here is Maven2 dependency for use SimpleNamingContextBuilder class -

<dependency>
   <groupId>org.springframework</groupId>
   <artifactId>spring-test</artifactId>
   <version>3.0.3.RELEASE</version>
   <scope>test</scope>
</dependency>

Generate Maven2 Archetype from existing project

2010-02-22T13:26:00.000-08:00

It's not hard to create custom archetype in Maven 2.2 from existing project. I had multi-module .ear Maven project which I want to reuse. So, I decided to play a little with custom archetypes to create my own and don't repeat myself again and again. Fixing dependencies manually - is not the best way to spend lifetime :)

To jumpstart archetype creaton there is a special maven goal of archetype plugin archetype:create-from-project. So,

mvn clean archetype:create-from-project

Now we have prototype of our archetype in target\generated-sources\archetype\ directory.
Then...

cd target\generated-sources\archetype\
mvn install

Friendly speaking, I had to do some tweaking to successfully generate normal .ear Maven 2 project which can be compiled and run without problems. It was fixing some of dependencies at Logic and EAR modules, bit it was not hard - just specifying ${groupId} where necessary. To use this archetype after "mvn install" just type

mvn archetype:generate

and choose EnterpriseApp-archetype from the list. It will looks like....

2: local -> EnterpriseApp-archetype (EnterpriseApp-archetype)
....

in the list.

Source code is here : http://ostas-blog-src.googlecode.com/svn/trunk/JbossCargoJpaEjbWebEarArchetype/.

To build it use

svn co http://ostas-blog-src.googlecode.com/svn/trunk/JbossCargoJpaEjbWebEarArchetype/

mvn install

You'll need to specify your jboss home location at parent pom.xml file to use cargo-jboss deployment. In my case it is

<jboss.home>C:/server/Java/jboss-4.2.2.GA/</jboss.home>

If you have any comments/issues/suggestions give me to know.

Screenshot of Web Page in Python - 15 lines of code

2009-09-22T04:34:00.000-07:00

I cant say that this solution is pretty elegant, but it's short and working. But... platform dependent, because of IE. Anyway here it is :

from win32com.client import Dispatch
from time import sleep
from sys import exit
import ImageGrab

IE=Dispatch("InternetExplorer.Application")
IE.Visible=1
IE.fullScreen = 1
IE.Navigate("http://ostas.blogspot.com")
while (IE.ReadyState != 4):
 sleep(1)
img = ImageGrab.grab((0, 0, 1440, 900))
img.save('screenshot.jpg','JPEG')
IE.Quit();
exit(0);

Code is pretty simple- it launches IE in full screen mode, waiting while page being loaded, makes screenshot of whole screen resolution - 1440 x 900 in my case, and happily exits.

Google XML Pages - HelloWorld app with source code

2009-04-16T04:32:00.000-07:00

Last time I blogged was October 20, 2008... Wow ! Time is going so fast ! I'm really appreciate my readers who stayed with me and keep signing up using feedburner. You rock ! :)

So, to be closer to topic... Google XML Pages or GXP - seems to be interesting since it has been used in an army Google products - AdSense, Blogger, and Google Reader. I have some previous Apache Velocity experience, so the questions are what GXP is, what the differences between the two if any, performance ? I'll try to highlight some of this points in my next post, but for now - just hello world app.

There is a lot of information about GXP, almost all of this info - just annonce. No code, no example...
Our GXP template loooks like this :

<gxp:template name="com.google.code.projects.BlogSrc.GxpApp.HelloGxp"
        xmlns="http://www.w3.org/1999/xhtml"
        xmlns:gxp="http://google.com/2001/gxp"
        xmlns:java="http://google.com/2001/gxp/code/java"
        xmlns:call="http://google.com/2001/gxp/call"
        gxp:ispace="preserve" gxp:espace="collapse"
        content-type="text/plain">

<gxp:import class='java.util.List' />
<gxp:param name='list' java:type='List{Integer}' />

<gxp:loop type='Integer' var='listItem' iterable='list'>
Value : <gxp:eval expr='listItem'/>
</gxp:loop>

</gxp:template>

What do we have here ? Template which takes as param List with name "list", iterates through list and print values stored in it. As you can see, template code is XML, actually. But ! After compiling this XML using GXP compiler we'll get Java source file with appropriate API for calling our template. Java method for calling template looks like this:

public static
void write(
final java.lang.Appendable gxp$out,
final com.google.gxp.base.GxpContext gxp_context,
final List<Integer> list)
throws java.io.IOException
{
final java.util.Locale gxp_locale = gxp_context.getLocale();
gxp$out.append("\n");
{
boolean gxp$bool$0 = false;
for (final Integer listItem : list) {
 if (gxp$bool$0) {
   gxp$out.append(" ");
 } else {
   gxp$bool$0 = true;
 }
 gxp$out.append("\nValue : ");                                      
 com.google.gxp.text.PlaintextAppender.INSTANCE
     .append(gxp$out, gxp_context, (listItem));
 gxp$out.append("\n");
}
}
gxp$out.append("\n");
}

Please notice list and listItem variables and their types. The last step is actually put some param and get result. The name of our tempale is com.google.code.projects.BlogSrc.GxpApp.HelloGxp, the name of file is HelloGxp.gxp. Result of template compilation is a java file - HelloGxp.java in package "com.google.code.projects.BlogSrc.GxpApp". Simplest way to run tempale code is just call static write() method from generated class HelloGxp. Something like this :

package com.google.code.projects.BlogSrc.GxpApp;

import com.google.gxp.base.GxpContext;
import com.google.gxp.html.HtmlClosure;
import com.google.gxp.base.GxpClosure;

import java.io.*;
import java.util.*;

public class MyHelloWorld
{
StringWriter out = new StringWriter();
GxpContext gxpContext =
    createGxpContext(Locale.US, false);
GxpContext createGxpContext(Locale locale, boolean useXmlSyntax)
{
  return new GxpContext(locale, useXmlSyntax);
}

public void testHello() throws Exception
{
  final List<Integer> list = new ArrayList<Integer>();
  list.add(1);
  list.add(2);
  list.add(3);
  list.add(4);
  list.add(5);
  list.add(6);

  HelloGxp.write(out, gxpContext,list);
}
public static void main(String args[])
{
     final MyHelloWorld obj = new MyHelloWorld();
     try
     {
         obj.testHello();
         String s = obj.out.getBuffer().toString();
         System.out.println(s);
     }
     catch(Exception ex)
     {
         ex.printStackTrace();
     }
}
}

Complete source code you can find here.

GxpHelloWorldMavenBuild - the same project but for use with Maven 2.

Google App Engine will support Java ? Cool !

2008-10-20T04:28:00.000-07:00

Well, I don't believe yet, but it seems to be that Java will be supported on GAP !

At the recently Google Developers Day in Bangalore, Keynote speaker Prasad Ram said that Google App Engine will now support Java.

Source here and here

JBoss EJB3 Integration testing with Maven2 and Cargo

2008-10-09T01:41:00.000-07:00

One of the most boring things in EJB development is that we need container. Compile-Build-Deploy-Run_some_testing_code - this is to long cycle to repeat it frequently without some kind of automation. I'll show you one of the possible ways to simplify this stuff. There are others, but I didn't tried them (I mean jboss microcontainer here). It's not complex, but it works and probably could be portable solution for not only my case.

So, my working environment is the following : Maven 2, JBoss 4.2.2 GA, jUnit 3.8.1 and of course EJB3. Build lifecycle with Maven 2 consists of some phases, like "...compile,test,package,integration-test...". Ok, with package and compile phases all is clear. Test... we need jboss to be up and running before we can test anything. How to start and stop jboss automatically during build ? The answer is Cargo with Maven2 plugin.

Cargo is a thin wrapper around existing containers (e.g. J2EE containers). It provides different APIs to easily manipulate containers.

Now we can assign "start container" and "stop container" goals to some Maven 2 build phase. First of all, we are interested in integration-test. Also we need to skip junit tests before integration test. To achieve this I'm using this config for cargo plugin :

<plugin>
<groupId>org.codehaus.cargo</groupId>
<artifactId>cargo-maven2-plugin</artifactId>
<version>0.3-SNAPSHOT</version>
<configuration>
<container>
 <containerId>jboss4x</containerId>
  <home>${jboss.home}</home>
  <append>false</append>
</container>
<configuration>
 <type>existing</type>
 <home>${jboss.home}/server/default</home>
 <properties>
   <cargo.jboss.configuration>default</cargo.jboss.configuration>
   <cargo.rmi.port>1099</cargo.rmi.port>
   <cargo.logging>high</cargo.logging>
 </properties>
</configuration>
 <wait>false</wait>
</configuration>
 <executions>
   <execution>
     <id>start-container</id>
     <phase>pre-integration-test</phase>
     <goals>
       <goal>start</goal>
     </goals>
   </execution>
   <execution>
     <id>stop-container</id>
     <phase>post-integration-test</phase>
     <goals>
       <goal>stop</goal>
     </goals>
   </execution>
 </executions>    
</plugin>

And this for surefire - to be able to run junit tests on integration-test phase, after container started at pre-integration-test phase :

<plugin>                                      
<groupId>org.apache.maven.plugins</groupId>
<artifactId>maven-surefire-plugin</artifactId>
<configuration>                            
   <skip>true</skip>                       
</configuration>                           
<executions>                               
   <execution>                             
      <id>surefire-it</id>                 
      <phase>integration-test</phase>      
      <goals>                              
         <goal>test</goal>                 
      </goals>                             
      <configuration>                      
         <skip>false</skip>                
      </configuration>                     
   </execution>                            
</executions>                              
</plugin>

I haven't any luck with "standalone" type configuration. I should say that when"existing" jboss installation is used here, overall process is quite fast:

[INFO] -------------------------------------------
[INFO] BUILD SUCCESSFUL
[INFO] -------------------------------------------
[INFO] Total time: 42 seconds
[INFO] Finished at: Thu Oct 09 15:15:39 EEST 2008
[INFO] Final Memory: 9M/19M
[INFO] -------------------------------------------

42 seconds to compile code, package jar, copy jar to deploy dir, start jboss, run junit test, stop jboss and all this by doing just "mvn install" !

Complete source code available via svn here.

PS.
Time can be very different. My hardware is HP 6820s laptop. BTW, executing "mvn -o install" takes 23 sec :)

Build EJB3 with Maven 2 example with source code

2008-10-03T07:00:00.000-07:00

My posts (the first and the second) about how to build EJB3 using Maven 2 was very popular. This post is some kind of update, but now, with complete source code. BTW, I'm going to publish some of sources I mentioned at my blog posts, since readers are asking for them. I mean source code from this post at minimum.

So, here it is ! In my further posts I'm going to highlight some of pitfalls one can face with.

Yet another Java Puzzle

2008-08-27T02:17:00.000-07:00

No words here, just the code :)

class String
{
    public static void main(String args[])
    {
        String String  = new String();
    }
}

Wow ! DZone.com using Acegi Security !

2008-07-25T03:22:00.000-07:00

I've seen a very interesting alert box with exception message at dzone.com when tried to vote up for the post. It was something like "org.acegisecurity....". Hey, wait a minute, it is Acegi ! Really, login form looks as so

<form method="post" action="/links/j_acegi_security_check" 
  class="loginform" id="addlinkform">
<div>
 <label>Username:</label>
 <input type="text" class="txt" name="j_username" size="20"/>
</div>
<div>
 <label>Password:</label>
 <input type="password" class="txt" name="j_password" size="20"/>
 <label/>
 <a href="http://zones.dzone.com/user/password"
   class="forgotpass">
 Forgot password?</a>
</div>
<div>
 <label class="cbx">
 <input type="checkbox" name="_acegi_security_remember_me"
 checked=""/>
 remember me
 </label>
</div>
<div class="submit">
 <input type="submit" 
    onsubmit="lightbox.prototype.deactivate();" value="Login"/>
 <a rel="deactivate" class="lbAction" href="#">
 <input type="button" value="Cancel"/></a>                   
</div>
</form>

j_acegi_security_check, j_username, j_password, _acegi_security_remember_me - it seems to be really Acegi !

Struts 2 : Deprecate or remove Dojo plugin

2008-07-22T05:03:00.000-07:00

Sometimes it is very useful to read struts-dev mailing list. See here. In fact I'm very happy of that, because I was going to use some AJAX with Struts 2 and Dojo plugin. One thing I couldn't figure out is why Dojo ?? Why not Prototype ? Why not jQuery ? Well, OK. If developers using Dojo, I will do the same. Now the situtation is more clear and I'm not going to use Dojo anymore. Maybe Prototype ? :)

A little bit of dependency injection with Struts 2 and Spring 2

2008-07-18T04:04:00.000-07:00

I'm using Struts 2 + Hibernate 3 + Spring 2 for my current project. I have to admit that Struts+Spring integration is really easy. The key integration point is Struts 2 Spring plugin. I'll not repeat the docs here. Only one thing I should admit that I prefer to handle spring dependencies by myself because of little bit different version of Spring - 2.0.5 in plugin and 2.0.8 in my project. So, I have this lines in my pom.xml (Maven 2 build file)

<dependency>
<groupId>org.apache.struts</groupId>
<artifactId>struts2-spring-plugin</artifactId>
<version>2.0.11.1</version>
<exclusions>
    <exclusion>
        <groupId>org.springframework</groupId>
        <artifactId>spring-beans</artifactId>
    </exclusion>
    <exclusion>
        <groupId>org.springframework</groupId>
        <artifactId>spring-core</artifactId>
    </exclusion>
    <exclusion>
        <groupId>org.springframework</groupId>
        <artifactId>spring-context</artifactId>
    </exclusion>
    <exclusion>
        <groupId>org.springframework</groupId>
        <artifactId>spring-web</artifactId>
    </exclusion>
</exclusions>
</dependency>

and

<dependency>
<groupId>org.springframework</groupId>
<artifactId>spring-core</artifactId>
<version>${spring.version}</version>
</dependency>

etc. I'm injecting Spring powered services into Struts 2 actions, but I had never had a chance to inject some kind of static list. Now I have, so the story about it.

I have a static list of countries. I think it doesn't make sense to store this info in database. The number of countries is not that large. If so, I need to store this data in some data structure and place it in memory. It could be List or Map - it depends on situation. And init this data structure once application starts. In web app when war file being deployed, Spring's WebApplicationContext initialized. So, when we define our list as Spring bean, it will be initialized at the application start and will be ready to other manipulations. I've defined it as so :

<?xml version="1.0" encoding="UTF-8"?>
<beans xmlns="http://www.springframework.org/schema/beans"
     xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
     xmlns:util="http://www.springframework.org/schema/util"
     xsi:schemaLocation="http://www.springframework.org/schema/beans 
       http://www.springframework.org/schema/beans/spring-beans-2.0.xsd 
       http://www.springframework.org/schema/util 
       http://www.springframework.org/schema/util/spring-util-2.0.xsd">

  <util:list id="countryList" list-class="java.util.LinkedList">
      <value>item1</value>
      <value>item2</value>
      <value>item3</value>
  </util:list>

</beans>

Here is the one question on Spring forum and official docs about it. For now we have a LinkedList filled with countries. We are going to inject it into Spring action and then show it at the "view" jsp.

Please note that id of our list is countryList. In fact it is a name of the bean. It's name really important since Struts 2 Spring pluring's default autowiring strategy is by name (I'm using default autowiring type and happy with that). It means that we can use setter injection this way

public class SomeStruts2Action extends ActionSupport
{
  public List countryList;

  public void setCountryList(List countryList) {
      this.countryList = countryList;
  }
  .......
}

Please note that id and class member name with type List are equal. That is all ! But actually not all :) We have to tell where is our list definition is. In my case it's at one of jars so I have this line in web.xml :

<context-param>
   <param-name>contextConfigLocation</param-name>
   <param-value>classpath:applicationContext-myCountryList.xml,
   /WEB-INF/appContext-web.xml</param-value>
</context-param>

Have fun !

Spring Security 2.0 : Different target urls for different user roles

2008-06-05T00:24:00.000-07:00

I'm evaluating Spring Security 2.0. The nice thing - it's really easier to have deal with it, rather than Acegi Security. I had an xml config file with 250+ LOC with Acegi ! Something like Rocket Science... but it is very flexible and expendable, it is very big plus.

After all, it is still not so trivial to implement some things which I found very commonly used. But... Spring rocks ! And Spring community rocks even more. I've posted message to Spring forum here.
Thanks Luke Taylor for pointing me to the right direction.

Lets imagine we have the same login form for common user and admin and we want admin to be redirected to some kind admin control panel and common user to /home /profile or something like that. In my case customized AuthenticationProcessingFilter will do the trick. The code looks like this :

package com.mycoolcompany.App.CustomSecurity;

import javax.servlet.http.HttpServletRequest;
import javax.servlet.http.HttpSession;

import org.springframework.security.Authentication;
import org.springframework.security.AuthenticationException;
import org.springframework.security.GrantedAuthority;
import org.springframework.security.providers.
 UsernamePasswordAuthenticationToken;
import org.springframework.security.ui.webapp.
 AuthenticationProcessingFilter;

public class CustomAuthenticationProcessingFilter
extends AuthenticationProcessingFilter
{
@Override
public Authentication attemptAuthentication(HttpServletRequest request)
throws AuthenticationException {
String username = obtainUsername(request);
String password = obtainPassword(request);

if (username == null) {
 username = "";
}

if (password == null) {
  password = "";
}

username = username.trim();

UsernamePasswordAuthenticationToken authRequest =
new UsernamePasswordAuthenticationToken(username, password);

// Place the last username attempted into HttpSession for views
HttpSession session = request.getSession(false);

if (session != null || getAllowSessionCreation()) {
  request.getSession()
.setAttribute(SPRING_SECURITY_LAST_USERNAME_KEY,
escapeEntities(username));
}

// Allow subclasses to set the "details" property
setDetails(request, authRequest);

//role&URLs stuff
final Authentication auth = 
this.getAuthenticationManager().authenticate(authRequest);
final GrantedAuthority[] grantedAuthorities = auth.getAuthorities();
boolean isAdmin = false;
for(GrantedAuthority grantedAuthority : grantedAuthorities)
{
if("ROLE_SUPERVISOR".equals(grantedAuthority.toString()))
   {
     isAdmin = true;
     break;
   }
}
String outcome = null;
if(isAdmin)
{
  outcome = "/adminArea";
}
else
{
 outcome = "/someOtherUserArea";
}
//actual change of default url for user
this.setDefaultTargetUrl(outcome);

    return auth;
}
public static String escapeEntities(String s) {
StringBuffer sb = new StringBuffer();
    for (int i=0; i < s.length(); i++) {
        char c = s.charAt(i);
    
        if(c == '<') {
            sb.append("&lt;");
        } else if (c == '>') {
            sb.append("&gt;");
        } else if (c == '"') {
            sb.append("&#034;");
        } else if (c == '\'') {
            sb.append("&#039;");
        } else {
            sb.append(c);
        }
    }

    return sb.toString();
}
}

And XML config :

<?xml version="1.0" encoding="UTF-8"?>

<beans:beans xmlns="http://www.springframework.org/schema/security"
 xmlns:beans="http://www.springframework.org/schema/beans"
 xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
 xsi:schemaLocation="
 http://www.springframework.org/schema/beans 
        http://www.springframework.org/schema/beans/spring-beans-2.0.xsd
        http://www.springframework.org/schema/security 
        http://www.springframework.org/schema/security/spring-security-2.0.1.xsd">

 <http auto-config="false" 
entry-point-ref="authenticationProcessingFilterEntryPoint">
  <intercept-url 
 pattern="/secure/extreme/**"
 access="ROLE_SUPERVISOR"/>
  <intercept-url 
 pattern="/someDefaultUrl/**" 
 access="ROLE_SUPERVISOR"/>
  <intercept-url
 pattern="/secure/**" 
 access="IS_AUTHENTICATED_REMEMBERED" />
  <intercept-url
 pattern="/**" 
 access="IS_AUTHENTICATED_ANONYMOUSLY" />
  <anonymous  />
 </http>

 <!--
    Usernames/Passwords are
        rod/koala
        dianne/emu
        scott/wombat
        peter/opal
    -->
<authentication-provider>
<password-encoder hash="md5"/>
<user-service>
 <user name="rod" 
password="a564de63c2d0da68cf47586ee05984d7" 
authorities="ROLE_SUPERVISOR, ROLE_USER, ROLE_TELLER" />
<user name="dianne" 
password="65d15fe9156f9c4bbffd98085992a44e" 
authorities="ROLE_USER,ROLE_TELLER" />
<user name="scott" 
password="2b58af6dddbd072ed27ffc86725d7d3a"
authorities="ROLE_USER" />
<user name="peter" 
password="22b5c9accc6e1ba628cedc63a72d57f8" 
authorities="ROLE_USER" />
</user-service>
</authentication-provider>

<authentication-manager alias='authenticationManagerAlias'/>
<beans:bean id="authenticationProcessingFilterEntryPoint"
class="org.springframework.security.ui.webapp.
AuthenticationProcessingFilterEntryPoint">
     <beans:property 
 name="loginFormUrl"
 value="/login.jsp"/>
     <beans:property
 name="forceHttps" 
 value="false" />
 </beans:bean>
<beans:bean id="myAuthenticationProcessingFilter"
class="com.mycoolcompany.App.CustomSecurity.
CustomAuthenticationProcessingFilter">
<beans:property 
name="filterProcessesUrl" 
value="/j_spring_security_check" />
<beans:property 
name="defaultTargetUrl"
value="/someDefaultUrl/index.jsp"/>
<beans:property
name="authenticationManager"
ref="authenticationManagerAlias"/>
<custom-filter position="AUTHENTICATION_PROCESSING_FILTER"/>
</beans:bean>
</beans:beans>

The end.

Search with Spring Hibernate Lucene and Aspect Oriented Programming in action

2008-03-12T05:27:00.000-07:00

I've read a lot about AOP and what it's used for. It is useful for logging, tracing, declarative transactions management and even caching. Great ! I'm going to share with you yet another AOP use case. BTW, I found it really interesting. So, the story.

I have some entity with with field "content" which is HTML text. The number of entries is large and keep growing. It is natural that I want to find some entry which contain some text. So, I need search. Since text is in HTML format, database full text search is not the best solution possible. I've decided to try Lucene. It is a "high-performance, full-featured text search engine library". To use Lucene I need to crate an index, add my objects into it and then I will able to query index to get search results.

Since I'm using Spring, I was looking for Lucene-Spring integration. Bingo ! It is available as one of Spring modules project. This integration has two aspects - one for index management and another for performing actual search. Complete configuration :

<bean id="fsDirectory"
class="org.springmodules.lucene.index.support.FSDirectoryFactoryBean">
<property name="create" value="true"/>
<property name="location" value="file:///C:/temp/index" />
</bean>

<alias name="fsDirectory" alias="indexDirectory"/>

<bean
id="indexFactory"
class="org.springmodules.lucene.index.support.SimpleIndexFactoryBean">
<property name="create" value="true"/>
<property name="directory">
           <ref bean="indexDirectory" />
</property>
<property name="analyzer">
  <bean class="org.apache.lucene.analysis.standard.StandardAnalyzer"/>
</property>
</bean>

<bean id="indexAccessor"
class="com.mycoolcompany.app.search.LuceneIndexerService">
<property name="indexFactory">
<ref local="indexFactory" />
</property>
</bean>

<bean id="searcherFactory"
class="org.springmodules.lucene.search.factory.SimpleSearcherFactory">
<property name="directory">
  <ref bean="indexDirectory" />
</property>
</bean>

<bean id="searchService"
class="com.mycoolcompany.app.search.LuceneSearchService">
<property name="searcherFactory">
  <ref local="searcherFactory" />
</property>
<property name="analyzer">
<bean class="org.apache.lucene.analysis.standard.StandardAnalyzer" />
</property>
</bean>

I can pre-populate index by querying database and adding objects to index. But what to do when new entry has been added ? How to keep index up to date ? After some Googling I found solution with custom hibernate interceptor. It looks like this :

import java.io.Serializable;
import org.hibernate.EmptyInterceptor;
import org.hibernate.type.Type;

public class CustomHibernateInterceptor extends EmptyInterceptor
{
private static final long serialVersionUID = -1214372449390884955L;

//injected by spring
private LuceneIndexerService indexAccessor;
public LuceneIndexerService getIndexAccessor() {
   return indexAccessor;
}
public void setIndexAccessor(LuceneIndexerService indexAccessor) {
   this.indexAccessor = indexAccessor;
}   

public boolean onSave(
   Object entity,
   Serializable id,
   Object[] state,
   String[] propertyNames,
   Type[] types)
{
   if (entity instanceof Entry)
   {
       //adding to index
       indexAccessor.add((Entry)entity);
   }
   return super.onSave(entity, id, state, propertyNames, types);
}
}

Is it something wrong with this solution ? I guess no, it's working. But I have and idea to use AOP here. I guess, adding to index and persisting object to database at the same time is some king of cross-cutting concern. I'm using a Spring-Hibernate DAO for CRUD operations. My idea to resolve this issue is in using aspect. Aspect has a pointcut defined to intercept execution of my dao method addEntry(Entry entry) and "after" advice - this is the place where we can add object to index, after object has been persisted. Please notice that search service is injected into aspect :) All this stuff is very abstract but is pretty interesting to digg. Here is the source code of aspect :

@Aspect
public class IndexForSearchAspect
{
public IndexForSearchAspect()
{
    super();
}
//will be injected
private LuceneIndexerService indexAccessor;
public void setIndexAccessor(LuceneIndexerService indexAccessor)
{
  this.indexAccessor = indexAccessor;
}
//defining pointcut
@Pointcut("execution (* ....dao.BlogEntryDao.addEntry(....model.Entry))")
public void saveEntryToDB() {}

//add object to index after it has been persisted
@After("saveEntryToDB()")
public void addEntryToLuceneIndex(JoinPoint joinpoint)
{
 //getting argument of dao.addEntry()
 Entry entry = (Entry) joinpoint.getArgs()[0];
     // adding to index
 indexAccessor.add(entry);
}
}

Spring configuration for aspect :

<aop:aspectj-autoproxy/>
<bean id="searchindexAspect"
  class="com.mycoolcompany.app.search.IndexForSearchAspect">
  <!--indexer service will be injected -->
  <property name="indexAccessor">
      <ref local="indexAccessor"/>
  </property>
</bean>

I know, I've missed a lot of details here, so this post is something like show-case, not cookbook :) Any opinions are welcome !

DBUnit Boolean issue

2008-02-24T12:41:00.000-08:00

DBUnit is a nice procduct. In fact I'm not using it directly. I'm prefer going in declarative way with dbunit-maven-plugin. Why declarative way ? Well, all I need to do to populate database before executing JUnit test is to configure maven plugin and we are ready to go. The sweet thing is that data for storing in database is in xml format. XML is not SQL :) I've tested my application without any changes in HSQLDB, MySQL and Oracle without single change in data for testing. I mean no RDBMS specific SQL here, more to say, no SQL at all. Creation such datafile also is not a problem. DBUnit plugin can create it for you with the help of dbunit:export goal.

All is seems to be okay whyle you are not trying to persist Boolean value. For the first look, all is going wrong. I've spent a lot of time trying to understand WTF is going on. DBUnit does not insert boolean value, and says that column doesn't exist at all. It is damn fun when I know that column _exacly_ exists. Bug ? Feature ? /dev/hands error ? :)

As Google reported, I was not alone with this problem. And it is more a feature than a bug. Solution is here. So, adding ${dbunit.hack.Boolean} to plugin's configuration section works ! At HSQLDB profile of my parent pom.xml I have dbunit.hack.Boolean=org.dbunit.ext.hsqldb.HsqldbDataTypeFactory.

I'm happy, but how many another small bugs/features are in software, which are ready to eat wery easy 1-2-3 hours of our life ?

Tracing MyFaces with AspectJ, Tomcat and Maven 2

2007-12-04T02:17:00.000-08:00

When I started to learn Spring, I faced with AOP. It's new and interesting for me. It's really not easy to understand a new concept and to do some refactoring of my own brain while the most of documentation and books are in English (not my native language). There is not (almost) problem with English itself, but understanding a new concept is hard.

AOP is powerfull. Really. Relocating cross-cutting concerns to aspects is nice idea. Since I can intercept the methods executions of my own classes, I became very interested in how to trace any third-party libraries. In fact it's pretty cool to be able to see internals of MyFaces, Hibernate or whatever in action to get a deeper understanding what is going on behind the scene...

I found a wiki page about tracing MyFaces with AspectJ. Many thanks to Cagatay Civici for this article, really helpful. It is a starting point for my experiments. I'm using aspect source code from this article.

My environment is the following : JDK 1.5, Maven 2, Tomcat 5, MyFaces, AspectJ 1.5.

Cagatay restrict his article to statically compiled AspectJ. But as you'll see later, runtime weaving on JDK 1.5 is not so hard to implement.

The first issue I came across is how to compile AspectJ aspect with Maven 2. Bingo ! There is a Maven plugin. So, placing some Trace.aj at src\main\java will work.

<plugin>
<groupId>org.codehaus.mojo</groupId>
<artifactId>aspectj-maven-plugin</artifactId>
  <configuration>
    <outxml>true</outxml>
  </configuration>
  <executions>
   <execution>
     <goals>
       <goal>compile</goal>
         <goal>test-compile</goal>
     </goals>
   </execution>
  </executions>
</plugin>

Please note, that outxml=true. It's important since it is flag to "Generate aop.xml file for load-time weaving with default name.(/META-INF/aop.xml)".

The final part of config is agent setup for Tomcat. According to docs it's should be started with one additional JVM argument - javaagent.

Now we have tracing aspect, can compile it with Maven and know how to start Tomcat with right option. As for me, I like Cargo plugin. It is useful for integration testing for example, but in this case it is just a very nice Maven 2 plugin to put all together. With this config we can automatically download tomcat 5 distribution, unpack it, run tomcat with our -javaagent option
and deploy war. A lot of work, huh ? :)

<plugin>
<groupId>org.codehaus.cargo</groupId>
<artifactId>cargo-maven2-plugin</artifactId>
<configuration>
 <container>
  <containerId>tomcat5x</containerId>
   <zipUrlInstaller>
    <url>
    http://www.eu.apache.org/dist
    /tomcat/tomcat-5/v5.5.25/bin/apache-tomcat-5.5.25.zip
    </url>
   <installDir>
   ${java.io.tmpdir}/donloadedByCargo
   </installDir>
   </zipUrlInstaller>
   <output>
       ${project.build.directory}
   /tomcat5x.logs/container.log
   </output>
   <append>false</append>
     <log>
       ${project.build.directory}
     /tomcat5x.logs/cargo.log
     </log>
   </container>
<configuration>
 <type>standalone</type>
 <home>
 ${project.build.directory}/tomcat5x
 </home>
 <properties>
   <cargo.servlet.port>
     8080
   </cargo.servlet.port>
   <cargo.jvmargs>
     -javaagent:"${user.home}\.m2\repository
     \aspectj\aspectjweaver\1.5.3\aspectjweaver-1.5.3.jar"
   </cargo.jvmargs>
   <cargo.logging>high</cargo.logging>
 </properties>
<deployables>
 <deployable>
   <groupId>com.mycoolcompany.fun</groupId>
   <artifactId>MyFacesAspectJ</artifactId>
   <type>war</type>
   <properties>
     <context>MyFacesAspectJ</context>
   </properties>
 </deployable>
</deployables>
</configuration>
</configuration>
</plugin>

As you can see, cargo.jvmargs element contains a path to aspectjweaver-1.5.3.jar library, in my case it is in my local repository already. After all to see it in action just type "mvn clean package cargo:start" and you are ready to go !

Trace will be available in the file "C:\trace.out" as defined in aspect source code. Exploring it's contents is another story, I guess.

JSF : MyFaces and Sun RI profiles for Maven 2

2007-11-14T01:26:00.000-08:00

I've used MyFaces as implementation of JSF 1.1 specification. Why ? The main reason is great community and great support for Maven 2 - maven-archetype-myfaces for example, + Tomahawk.

And now, I'm interested in what is the difference between the two ? Any comments appreciated.

Maven 2 is my friend here, since I can use profiles - one for JSF RI, another one for MyFaces. I guess it could be useful for someone else, so here is the code :

<profiles>
<profile>
  <id>jsfri</id>
  <activation>
      <property>
          <name>jsf</name>
          <value>ri</value>
      </property>
  </activation>
  <dependencies>
  <dependency>
      <groupId>javax.faces</groupId>
      <artifactId>jsf-api</artifactId>
      <version>1.1_02</version>
      <exclusions>
          <exclusion>
              <groupId>java.servlet.servlet.jsp</groupId>
              <artifactId>jsp-api</artifactId>
          </exclusion>
          <exclusion>
              <groupId>javax.servlet</groupId>
              <artifactId>servlet-api</artifactId>
          </exclusion>
          <exclusion>
              <groupId>javax.servlet.jsp</groupId>
              <artifactId>jsp-api</artifactId>
          </exclusion>
          <exclusion>
              <groupId>javax.servlet.jsp.jstl</groupId>
              <artifactId>jstl</artifactId>
          </exclusion>
      </exclusions>
  </dependency>
  <dependency>
      <groupId>javax.faces</groupId>
      <artifactId>jsf-impl</artifactId>
      <version>1.1_02</version>
      <exclusions>
          <exclusion>
              <groupId>java.servlet.servlet.jsp</groupId>
              <artifactId>jsp-api</artifactId>
          </exclusion>
          <exclusion>
              <groupId>javax.servlet</groupId>
              <artifactId>servlet-api</artifactId>
          </exclusion>
          <exclusion>
              <groupId>javax.servlet.jsp</groupId>
              <artifactId>jsp-api</artifactId>
          </exclusion>
          <exclusion>
              <groupId>javax.servlet.jsp.jstl</groupId>
              <artifactId>jstl</artifactId>
          </exclusion>
      </exclusions>
  </dependency>
  <dependency>
      <groupId>javax.servlet</groupId>
      <artifactId>jstl</artifactId>
      <version>1.1.0</version>
  </dependency>
  </dependencies>
  <repositories>
      <repository>
          <id>java.net</id>
          <name>java.net Maven 1 Repository</name>
       <url>
 https://maven-repository.dev.java.net/nonav/repository
 </url>
          <layout>legacy</layout>
      </repository>
  </repositories>
</profile>
<profile>
  <id>myfaces</id>
  <activation>
      <property>
          <name>jsf</name>
          <value>myfaces</value>
      </property>
  </activation>
  <dependencies>
  <dependency>
      <groupId>org.apache.myfaces.core</groupId>
      <artifactId>myfaces-api</artifactId>
      <version>1.1.6-SNAPSHOT</version>
  </dependency>
  <dependency>
      <groupId>org.apache.myfaces.core</groupId>
      <artifactId>myfaces-impl</artifactId>
      <version>1.1.6-SNAPSHOT</version>
  </dependency>
  <!--
        <dependency>
            <groupId>org.apache.myfaces.tomahawk</groupId>
            <artifactId>tomahawk</artifactId>
            <version>1.1.6-SNAPSHOT</version>
        </dependency>
    -->
  </dependencies>
  <repositories>
  <repository>
      <releases>
          <enabled>false</enabled>
      </releases>
      <snapshots/>
      <id>apache-maven-snapshots</id>
      <url>
 http://people.apache.org/repo/m2-snapshot-repository
 </url>
  </repository>
  </repositories>
</profile>
</profiles>

To switch between profiles use "mvn -Djsf=ri" or "mvn -Djsf=myfaces".

JSF : hacking redirect and request scope managed bean

2007-11-12T03:45:00.000-08:00

Strting with MyFaces project and Maven 2 is very easy :

mvn archetype:create
-DarchetypeGroupId=org.apache.myfaces.maven
-DarchetypeArtifactId=maven-archetype-myfaces
-DarchetypeVersion=1.0-SNAPSHOT
-DgroupId=com.mycoolcompany.web
-DartifactId=jsfRedirect

When project is created, you can run web app (jetty is used here) :

mvn -U -PjettyConfig jetty:run

Now we have a skeleton of application. Let's try to understand how to deal with redirect.

Let's check what happens when form is submitted. Current config is

<navigation-case>
<from-outcome>success</from-outcome>
<to-view-id>/page2.jsp</to-view-id>
</navigation-case>

First request :

GET /jsfRedirect/helloWorld.jsf HTTP/1.1
Accept: */*
Accept-Language: ru
Accept-Encoding: gzip, deflate
User-Agent: Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; SV1; Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; SV1) ; .NET CLR 1.1.4322; .NET CLR 2.0.40903)
Host: localhost:8080
Connection: Keep-Alive
Cache-Control: no-cache
Cookie: JSESSIONID=zm2pcktnv9vl

Response :

HTTP/1.1 200 OK
Content-Type: text/html; charset=ISO-8859-5
Content-Language: ru
Content-Length: 7742
Server: Jetty(6.1.6rc1)

There is nothing special here.

Form submit:

POST /jsfRedirect/helloWorld.jsf HTTP/1.1
Accept: */*
Referer: http://localhost:8080/jsfRedirect/helloWorld.jsf
Accept-Language: ru
Content-Type: application/x-www-form-urlencoded
Accept-Encoding: gzip, deflate
User-Agent: Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; SV1; Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; SV1) ; .NET CLR 1.1.4322; .NET CLR 2.0.40903)
Host: localhost:8080
Content-Length: 3892
Connection: Keep-Alive
Cache-Control: no-cache
Cookie: JSESSIONID=zm2pcktnv9vl

Posted data is the following:

form:input1 testString
form:button1 press me
autoScroll 0,0
form_SUBMIT 1
form:_idcl
form:_link_hidden_
javax.faces.ViewState ___very__loong__

___very__loong__ is really very long string and I don’t want to post it, you can check it by yourself if you want.

As we can see, our submitted value were sent as form:input1 POST parameter. Please notice that at helloWorld.jsp we have assigned id="form" to h:form and id="input1" to h:inputText.

I guess org.apache.myfaces.AUTO_SCROLL is responsible for autoScroll param value.
To see it in action take a look here.

Responce is:

HTTP/1.1 200 OK
Content-Language: ru
Content-Type: text/html; charset=ISO-8859-5
Content-Length: 6781
Server: Jetty(6.1.6rc1)

....

And now let’s see what will change when redirect is used.

<navigation-case>

    <from-outcome>success</from-outcome>
<to-view-id>/page2.jsp</to-view-id>
<redirect/>
</navigation-case>

In this case form submission looks the same as before, but response is different, as expected :

HTTP/1.1 302 Found
Location: http://localhost:8080/jsfRedirect/page2.jsf
Content-Length: 0
Server: Jetty(6.1.6rc1)

"Location" http header means redirect for browser i.e. GET request. In this case the value of request scoped managed bean will be lost between requests (POST ant then GET), since HTTP is a stateless protocol.

How to avoid data loss ? Ohh.. it is a long story. To save data between requests I need to save managed bean in session. I can get access to http session object using faces context:

public class HelloWorldBacking {
....
public String sendSessionHack(){
FacesContext facesContext =
    FacesContext.getCurrentInstance();
Map session =
     facesContext.getExternalContext().getSessionMap();
session.put("myBackingBean",this);
return "successSessionHack";
}
}

Ok, but how to get this value back in view ? It is a long story...

Maybe "there is no one way to do it", but I see only one - PhaseListener.
If you know any other, please give me a sign! Code listed below will act at RESTORE_VIEW phase and then check if there is a session param with name myBackingBean, and finally assign
this value to managed bean via programmatically operations with JSF EL.

package org.apache.myfaces.blank.listener;

import javax.faces.event.PhaseListener;
import javax.faces.context.FacesContext;
import javax.faces.event.PhaseEvent;
import javax.faces.event.PhaseId;

import javax.faces.FactoryFinder;
import javax.faces.application.Application;
import javax.faces.application.ApplicationFactory;
import javax.faces.el.ValueBinding;

import org.apache.myfaces.blank.HelloWorldBacking;

import java.util.Map;

public class MyPhaseListener implements PhaseListener
{
public PhaseId getPhaseId()
{
  return PhaseId.RESTORE_VIEW;
}

public void beforePhase(PhaseEvent e)
{
FacesContext facesContext =
  FacesContext.getCurrentInstance();
Map session =
  facesContext.getExternalContext().getSessionMap();
HelloWorldBacking backingBean =
  (HelloWorldBacking)session.get("myBackingBean");

if(backingBean!=null)
{
ApplicationFactory appFactory =
(ApplicationFactory) FactoryFinder.getFactory(FactoryFinder.APPLICATION_FACTORY);
Application application =
appFactory.getApplication();
ValueBinding valueBinding =
application.createValueBinding("#{helloWorldBacking}");
valueBinding.setValue(facesContext,backingBean);
session.remove("myBackingBean");
}
}

public void afterPhase(PhaseEvent e)
{
  // don't care  
}
}

Not so trivial, isn't it? :)
BTW, to enable custom phase listener you need to add this snip of code in faces config file, examples-config.xml in our case :

<lifecycle>
<phase-listener>
org.apache.myfaces.blank.listener.MyPhaseListener
</phase-listener>
</lifecycle>

JSF : Redirect to URL with param from navigation-case

2007-11-01T01:43:00.000-07:00

I'm evaluating JSF now. I have to admit that some of JSF ideas are pretty nice, but there are many non-intuitive things which make life hard. From another point of view, it is framework, with it's own rules of the game.

From my little experience I could say that JSF doesn't like the way of passing params via GET. Of course it is possible, but I couldn't say it is trivial and easy as in Struts for example. So, here is some technical stuff.

I need to to redirect from navigation-case to URL with param which is a property of a managed bean. It looks like this :

<navigation-rule>
 <from-view-id>/add.jsp</from-view-id>
 <navigation-case>
  <from-outcome>added</from-outcome>
  <to-view-id>/view.jsf?myParam=#{myManagedBean.id}</to-view-id>
  <redirect/>
 </navigation-case>
</navigation-rule>

I've tried to figure out how to do it, but had no success here. In fact I found solution, but it's more something like a hack than solution.
So, I've asked MyFaces mailing list. Many thanks to all the people who answered me.

To keep the story short, here is solution. Custom view and navigation handlers should be added to use managed bean property as url param in navigation-case.

I think to myself... should I use core JSF or switch to Apache Orchestra, or JBoss Seam to make development with JSF less painful ? JSF is still young, 1.2 is the latest version of specification and frameworks that use JSF has many additional functionality. Who knows ?

static final String[] is not final ?

2007-09-21T07:09:00.000-07:00

Lets take final variable and try to assign new value to it. Well , we 'll have no luck here - "cannot assign a value to final variable myVar" yet at compilation stage. But having a deal with final arrays is something different. Here is an example :

class FinalArray
{
public static final String[] array = {"a", "b","c"};
 public static void main(String args[])
 {
     array[0] = "aaa";
     System.out.println(array[0]);
 }
}

Output will be "aaa". It is still not possible to do something like

String[] some_array = {"aaa", "bbb","ccc"};
array = some_array;

but possible per-element assigning new values to final array elements:

class FinalArray
{
public static final String[] array = {"a", "b","c"};
  public static void main(String args[])
  {
      String[] some_array = {"aaa", "bbb","ccc"};
      for(int i=0;i<length;i++)
      {
          array[i] = some_array[i];
      }
    
      for(String element: array)
      {
          System.out.println(element);
      }
  }
}

So, it is final, you say ?

Maven: The Definitive Guide (1.0 Alpha 2)

2007-08-27T07:31:00.000-07:00

It seems to be that this is very nice book about Maven 2. Maybe it can help me to understand Maven 2 concepts even better. I'd found link to this book at this blog. I'm agree with author, Maven 2 is a really painful sometimes. Ant and Maven concepts are very different and a lot of time is needed to switch between them.