kam

one of those blogs

For the past several years most of my projects were based on Spring and Hibernate. And what I keep seeing is how people struggle to understand how the two are working together.

So going from project to project, I have to repeat the same spiel over and over again to get people up to speed.

Since I believe that the most effective way to LEARN is to DO, I create simple examples for people to play with / to complete / to analyze, etc..

So “Money Making Project” is one of such examples. It comfortably lives at github’s luxury apartments, so anybody can “git clone” / “fork” and play with it.

Besides making money, this project also demos things such as:

A way to structure a project

Maven based structure ( hence can be easily used by gradle ). Configuration and property files organized under “META-INF/conf”, “META-INF/props”, etc..

A way to separate Spring configs

“tx-spring-config.xml”, “persistence-spring-config.xml”, “service-spring-config.xml”, “repository-spring-config.xml” and “application-context.xml” that includes them all

Properties via PropertyPlaceholderConfigurer

<context:property-placeholder location="classpath:META-INF/props/env.properties"/>

Hibernate overall configuration file

That is injected into AnnotationSessionFactoryBean

Hibernate Named Queries

That are linked to the Hibernate overall config

<mapping resource="META-INF/conf/hibernate/mapping/startup-bank-named-queries.xml"/>

Spring’s DAO / Hibernate Exception Translation

Via @Repository and “PersistenceExceptionTranslationPostProcessor”

Simple CRUD Repository

public interface MoneyRepository {
 
    public void make( MoneyRoll moneyRoll );        // C
    public MoneyRoll find( Long id );                       // R
    public void update( MoneyRoll moneyRoll );     // U
    public void takeOut( MoneyRoll moneyRoll );    // D
}

with a Hibernate based implementation

Transaction Management with Spring AOP

Declarative, on a Service Level, using “aop:config”, “tx:advice” namespaces

Spring Testing

With SpringJUnit4ClassRunner, ContextConfiguration, etc..

Using Embeded in-memory H2 Database for Testing

<jdbc:embedded-database id="dataSource" type="H2"/>

Hibernate Logging

Most useful hibernate “log4j.logger” properties

Demoing how important Transaction is for Hibernate Sessions

“HibernateSessionNotBoundToThreadIntegrationTest” for the second (after LazyInitializationException ) most common Hibernate exception: “No Hibernate Session bound to thread, and configuration does not allow creation of non-transactional one here”

Let’s say you have an aspect defined that will be applied to a class:

<bean id="someComponent"
      class="org.dotkam.SomeComponent" />
 
<aop:config>
	<aop:aspect ref="someAspect">
		<aop:around method="someMethod"
                            pointcut="execution(public * org.dotkam.SomeComponent.*(..))" />
	</aop:aspect>
</aop:config>

and you of course have “proxy-target-class” set to “true”:

<aop:aspectj-autoproxy proxy-target-class="true"/>

Here is how to check if the aspect was applied:

import static org.junit.Assert.assertTrue;
import net.sf.cglib.proxy.Enhancer;
 
  ... ...
@Resource(name="someComponent")
SomeComponent someComponent
 
  ... ...
assertTrue( Enhancer.isEnhanced( someComponent.getClass() ) );

Behind the scenes it is enhanced with CGLIB’s Enhancer that has a convenient isEnhanced method.

Useful when developing aspects.

NOTE: The above is true given, of course, that this is THE ONLY aspect applied to this component.

sudo vi /usr/share/tomcat6/bin/catalina.sh

At the beginning of the file, after initial comment block add:

# setting JAVA_OPTS to perform
JAVA_OPTS="-Xms512m -Xmx1024m -XX:MaxPermSize=384m -Xss128k"

Restart tomcat:

sudo service tomcat6 restart

You heard about Spring Batch, but have this “uneasy” feeling about where/how/what you should start with? A year ago it was truly difficult for a non geek developer (most of the developers) to start with projects like Spring Batch, Spring Integration, Spring Webflow, Spring MVC, etc.. But now..

Now, you have no excuse: it is extremely simple to build a sample Spring project, of pretty much any “flavour”, that comes pre-built with unit and integration tests, that you can run right after those 12 seconds that you spent downloading the project, and resolving dependencies.

Does it sound helpful? Well, not really, since I just said what will happen, and how easy it is to make it happen, but there was no how part… I hate when people do that

So without further ado, here comes the “how” part:

Step 1. DON’T PANIC

After successful completion of Step 1, it is time for…

Step 2. Download Spring Tool Suite

As I already mentioned in “Spring Insight in Action – 5 Minutes From Scratch“, I like to think about Spring Tool Suite as Eclipse on Spring rocks (Spring IDE, Spring Interactive Tutorials, Exception Resolution, Grails support, Spring Insight, Spring tc Server, Spring Template Projects, and much more): http://www.springsource.com/products/springsource-tool-suite-download

Step 3. Create New Spring Template Project

After Spring Tool Suite is started (remember, it is just a Spring version of Eclipse, so it should look very familiar), press Ctrl+n to create “something” new, and start typing “spring temp…”:

As you type, Eclipse will narrow down selection to “Spring Template Project”. Select it, and click Next:

Here you can see many hot and ready to go Spring template projects that you can explore. The purpose of this article is to show how easy it is to start with Spring Batch, but for the most part, you can follow these steps to create all of the projects from the list above. (as an extra credit, try to create a Spring Integration Template project, once we done here).

Select “Simple Spring Batch Project”, and click Next:

Name your fresh and ready to go Spring Batch project, and click Finish, here is what your own Spring Batch project will look like:

You are actually done right here, but, since curiosity is what makes us humans, let’s take a look at what’s inside:

As you can see, it is a Maven structured project with all the goodies: source, tests, pom, configs, etc..

Let’s look at your first real Batch Job. Double click on “module-context.xml”:

In order to make sense of what you are looking at, you can refer to a very good and extensive Spring Batch documentation: http://static.springsource.org/spring-batch/reference/html/index.html

Step 4. Make Sure It Works

But hey, don’t just believe me that it works => make sure it does! And how would you do that? Well, again, simple: just run a test:

Now you should see some green, which actually tells you: “it works indeed”.

Now you see, that starting to work on any major Spring Project is only a couple of mouse clicks away. You can really feel how close and accessible all that knowledge is. So be brave and take it all!

Good Luck!

Recently one of my friends from work asked me to help him improve the process around code quality and developer productivity. So I compiled my thoughts and e-mailed to him, but then I realized that it may be very helpful for others who are involved in software industry. Are you? Then keep reading..

Although tools and frameworks listed here are JVM-based language focused, the approach can be definitely reused with any other environment / language / technology. So without further ado, here it goes:

The center of the code quality monitoring can be either Continuous Integration ( e.g. Cruise Control: http://cruisecontrol.sourceforge.net/ ) or Sonar ( http://sonar.codehaus.org/ ). Sometimes both.

Continuous Integration should be setup to “Intergarte Continuously” which means every time something is checked in, force a build. That is the whole purpose of “Continuous Integration”, and that is why I am really against the way Cruise Control is used on some client sites [builds on demand by pressing a build button.. grrr, back to 1990-ties].

So given that “Continuous Integration” is not misused, it will be plugged in with PMD ( http://pmd.sourceforge.net/ ) / Findbugs ( http://findbugs.sourceforge.net/ ), Checkstyle ( http://checkstyle.sourceforge.net/ ) and Test Coverage ( e.g. Cobertura: http://cobertura.sourceforge.net/ ) tools that will generate reports, and reflect the only true state of the code that is checked into the repository.

Now, as to a developer corner..

RAD is nothing more than just an Eclipse with lots of bloated (mostly unused IBM plugins). But being Eclipse it is very pluginable by nature. This means that PMD / Checkstyle / Cobertura reports can be available to developers prior checking in the code. PMD / Checkstyle at compile time, Cobertura at (test) run time. If possible, try to use something like Spring Tool Suite: http://www.springsource.com/products/sts, which is also an Eclipse, but much lighter (compare to RAD), faster and smarter “pluged-in”.

As far as testing. JUnit (http://www.junit.org/) should be aimed to cover two different separate layers of testing: Component/Unit Testing and Integration Testing. Component tests should be aimed to test exclusively content of the component with all of its dependencies mocked out ( http://mockito.org/ ). Where as Integration tests should test how well components integrate together, with all the test data staged ( in case DB is used: http://www.dbunit.org/ )

As for the test coverage, 85% to 90% is a good goal to aim for. Do not sign off, until this level of coverage is reached no matter how close your dead lines are, since if you do, that will increase amount of defects ten fold, that is just a law

If developers work on code that is / can be deployed to an Application Server, such as Tomcat, JBoss, Geronimo ( if you’re unlucky, Websphere ), consider getting JRebel ( http://www.zeroturnaround.com/jrebel/ ), it’ll boost developers productivity by.. let’s just say “a lot”

This is how I see it, and this really works for me, and projects around me.

“A popular technique for organizing the execution of complex processing flows is the “Chain of Responsibility” pattern, as described (among many other places) in the classic “Gang of Four” design patterns book. Although the fundamental API contracts required to implement this design patten are extremely simple, it is useful to have a base API that facilitates using the pattern, and (more importantly) encouraging composition of command implementations from multiple diverse sources.”

That is what Apache has to say as an intro to its Commons Chain API

It is no brainer, really, but Apache APIs do make it simpler to configure, implement and execute multiple Chains of Responsibilities. One thing, however, that feels “off”, when (if) implementing “Commons Chain” side by side with Spring, is there is no flexible, consistent (with Spring) way to setup Command objects: inject Command properties / refer to a single Command class as a bean, instead of duplicating the class name in Apache Chains, and Spring configuration, if Command object needs to be reused, etc…

Here is a simple way to utilize all the power that Apache offers + make it Spring friendly. Five components will be used in this example – 1 core class, 2 command classes, 1 test (driver) and Spring configuration file. Let’s see if you can figure out which is which :

As you can see we are going to play an ultra short session of table tennis, or how it is sometimes called “Ping Pong”. It is going to be quite short, because we have two command objects: Ping and Pong, which are going to be chained, and run (by ChainRunner) once each: “Ping -> Pong”.

First, instead of relying on Apache way to configure chains, to make it consistent with all other application beans (classes) we’ll keep chain(s) configuration in Spring:

chain-config.xml:

<?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-2.5.xsd
http://www.springframework.org/schema/util
http://www.springframework.org/schema/util/spring-util-2.5.xsd">
 
	<bean name="pingPongChain" class="org.apache.commons.chain.impl.ChainBase">
		<constructor-arg>
			<util:list>
				<ref bean="pingCommand" />
				<ref bean="pongCommand" />
			</util:list>
		</constructor-arg>
	</bean>
 
	<bean name="pingCommand" class="org.dotkam.samples.chain.command.PingCommand">
		<constructor-arg value="ping"/>
	</bean>
 
	<bean name="pongCommand" class="org.dotkam.samples.chain.command.PongCommand">
		<constructor-arg value="pong"/>
	</bean>
 
	<bean id="chainRunner" class="org.dotkam.samples.chain.ChainRunner"/>
 
</beans>

A “pingPongChain” is configured as “org.apache.commons.chain.impl.ChainBase”, and have two Commad classes, that implement “org.apache.commons.chain.Command” interface, wired in.

NOTE: The best practice is to keep Commands either stateless, or, if they have to have a state – immutable. If this practice is followed, these Commands can be safely reused throughout many chains (or even as stand alone utilities). That is the reason, in configuration above, parameters are injected at Command’s creation time via constructor injection.

In this example Commands are ultra simple for clarity:

PingCommand.java:

package org.dotkam.samples.chain.command;
 
import org.apache.commons.chain.Command;
import org.apache.commons.chain.Context;
 
public class PingCommand implements Command {
 
	private String message;
 
	public PingCommand( String message ) {
		this.message = message;
	}
 
	public boolean execute( Context context ) throws Exception {
 
		System.err.println( message );
		return false;
	}
}

and PongCommand.java:

package org.dotkam.samples.chain.command;
 
import org.apache.commons.chain.Command;
import org.apache.commons.chain.Context;
 
public class PongCommand implements Command {
 
	private String message;
 
	public PongCommand( String message ) {
		this.message = message;
	}
 
	public boolean execute( Context context ) throws Exception {
 
		System.err.println( message );
		return false;
	}
}

Now let’s look at the heart of the “Spring Chainer” – the “ChainRunner.java”:

package org.dotkam.samples.chain;
 
import org.apache.commons.chain.impl.ChainBase;
import org.apache.commons.chain.impl.ContextBase;
import org.springframework.beans.BeansException;
import org.springframework.beans.factory.BeanFactory;
import org.springframework.beans.factory.BeanFactoryAware;
 
public class ChainRunner implements BeanFactoryAware {
 
	private BeanFactory beanFactory;
 
	public void runChain( String chainName ) {
		try {
			createChain ( chainName ).execute( new ContextBase() );
		}
		catch ( Exception exc ) {
			throw new RuntimeException(
					"Chain \"" + chainName + "\": Execution failed.", exc );
		}
	}
 
	public void setBeanFactory( BeanFactory beanFactory ) throws BeansException {
		this.beanFactory = beanFactory;
	}
 
	protected ChainBase createChain( String chainName ) {
		return ( ChainBase ) this.beanFactory.getBean( chainName );
	}
}

It only does a couple of things, really.

By implementing Spring’s “BeanFactoryAware” interface, at runtime ChainRunner will have a reference to a “beanFactory” which it is going to use to obtain a reference to the requested chain via “createChain” method.

“ChainRunner” has an entry point which is a “runChain” method, that executes a chain by chain name (bean name in configuration file). For example, in the configuration shown above “pingPongChain” name can be provided.

It would ideally need to use a couple of custom runtime exceptions: e.g. ChainNotFoundException, IsNotChainException and ChainExecutionException, but we’ll keep it short here for clarity.

Alternatively, to strong type chains a bit, “ChainRunner” could take a Map of chains with keys as chain names, and corresponding “ChainBase” chain objects as values.

“ChainRunner” is pretty much everything that is needed in order to configure an Apache chain in Spring. We can run it by creating a JUnit (not really a test, just a driver):

ChainTest.java:

package org.dotkam.samples.chain;
 
import javax.annotation.Resource;
import org.junit.Test;
import org.junit.runner.RunWith;
import org.springframework.test.context.ContextConfiguration;
import org.springframework.test.context.junit4.SpringJUnit4ClassRunner;
 
@RunWith( SpringJUnit4ClassRunner.class )
@ContextConfiguration( locations={ "/conf/chain-config.xml"} )
public class ChainTest {
 
	@Resource
	ChainRunner chainRunner;
 
	@Test
	public void driveTheChain() {
		System.out.println("Starting up...      [Ok]");
		chainRunner.runChain( "pingPongChain" );
		System.out.println("Finised...          [Ok]");
	}
}

And here is the test/driver result:

Good start! Now it is time to sign up and see your name at the top of The International Table Tennis Federation (ITTF) World Ranking. Good Luck!

No, really, what is it? Are you one of those people who ever wondered why there is actually more than one? Or you work with Struts for 5 years in a row, and think it is the best, just because you know all the nasty hooks you have to implement to make it do what you actually need?

Are you one of those die hard Rails fans, who (as most of Rails developers) does not really know Java, but already has a lot to say how bad it is?

Do you work in C++, and feel absolutely confident, that the framework, your company had developed for the last 25 years, that build web pages via socket programming is the best, and the most efficient one out there?

Is your company big enough, so it has a mature team of software (what they call themselves) architects, that take open source frameworks, and write their own single wrapper framework around it that, as they think (since they wrote it), is the 8th wonder of the (ancient) world?

You, see, I am a consultant – I know many of you, since I met “you” before

This one will be a very short article about my experiences with several Java Web Frameworks out there. Here we go:

And here is why I like Spring MVC:

• Binding / Validation is done just right – clean, testable, reusable
• Multiple View options ( PDF, XML, Excel, Atom, etc… ) done easy [AbstractExcelView, AbstractFeedView, AbstractJExcelView, AbstractPdfView, AbstractUrlBasedView, AbstractXsltView ]
• Annotation based – no XML madness, and very clear when looking at the code – check the Pet Clinic in Spring 3.0 M3
• Integrates with Spring JS very nicely ( in case needed )
• Handing requests and parameters with ( Spring ) expression language – quite flexible
• In Spring 3, MVC is actually REST aware (GET, POST, PUT, DELETE)

You can download Spring STS, import sample projects, and see many examples on how to use it, but here is a very simple controller example from Spring’s Pet Clinic:

@Controller                                                 // it’s a controller
public class ClinicController {
 
      @RequestMapping("/welcome.do")           // that all there is to mapping
      public void welcomeHandler() {
      }
 
      /** @return a ModelMap with the model attributes for the view
                                 uses org.springframework.core.Conventions */
      @RequestMapping("/vets.do")
      public ModelMap vetsHandler() {
            return new ModelMap(this.clinic.getVets());
      }
 
      @RequestMapping("/owner.do")
      public ModelMap ownerHandler(@RequestParam("ownerId") int ownerId) {  // parameters are passed in easily
            return new ModelMap(this.clinic.loadOwner(ownerId));
      }
}

It all depends on project’s requirements / timeline / resources / requirements / technologies already in place / etc… But having a choice, I do choose Spring MVC – it just makes sense: easy implementaion with Spring Roo / integration with Spring’s back end / support / community / releases / etc… I also like where Wicket is going, but it feels like “it is still going…”

In any case – good luck, and remember – if it does not have to be Java, but you still like to “stay close”, I would definitely give Grails a shot.

My first computer book was “Basic for Children with Arturo”. It was 1989, the book was in Russian, and I was 10…

Back then, in USSR, to own a 8086 compatible PC was somewhat mystical, and therefore people, who knew a thing or two about microelectronics, were putting together Speccy(s) (ZX Spectrums), which were based on Zilog Z80 8-bit microprocessor and ran at 3.5 MHz.

Official ZX-Spectrum that was sold by “Sinclair Research Ltd” at that time was a bit pricey for middle class families, and therefore many cities across Europe and USSR started to come up with their own “clones“. There were over 50(!) semi-official clones that came out by the end of 1980s, begining of 1990s. But to cut cost even further, people who knew “how to solder” created hybrids of the clones at home, which came out to be very cheap.

The process to build ZX-Spectrum’s clone’s hybrid was “a bit” involved, and included many days/weeks of hacking poor, incomplete electrical circuit documentation and sleepless nights of manual calibration of the hardware. The one had to “get” a ZX-Spectrum full circuit schema:

Study it, change it to fit the budget. Go to several flee markets to first find CPU, which was a quite rare find:

empty electrical border and all other chips / transistors / capacitors / resistors / etc…

After some time the great merge of new ideas, documentation, hardware and labour, the one ended up with something like this:

ZX-Spectrum keyboard had an interesting feature – pretty much every key, depending on the combination of special keys used, could represent 3-4, sometimes even 5 different things (commands/letters/symbols/special characters/etc). This made it much smaller (“compacter”) than for example PC keyboards:

And due to its small size, it most of the time served as a ZX-Spectrum case, and fit it the Speccy’s “motherboard”:

There were also some alternatives, and some people used “an extended” keyboards, but “a single key multi-purpose” pattern was still there. Here is a snippet of the Russian ZX-Spectrum keyboard, for example:

which is “a bit extended” – has a separate cursor, and some other keys. But take a look at the individual keys – due to the additional Cyrillic letters, “an average key” is now used to represent even more (+1) values.

At this point, when most of the work with oscillograph was behind, the only true way to see if the creation works was to see it with your own eyes. The word “monitor” in a sense of a “computer monitor” was not a widely used due to its price mostly, therefore most of ZX-Spectrum people used old black and white (at first) TVs. Speccy (well it’s clone’s hybrid at least) was designed in a way that it could be pugged into a regular TV Antena’s jack:

And due to the fact that family is usually more than one person, the main TV often was not used, and people used old TVs as a Speccy “monitor” instead:

The whole process above was about 50% of the job – yea, just 50%. The other half was split on two. First quarter was spent in calibrating hundreds of little circuits, in order to get a sacred “© 1982 Sinclair Research Ltd” message at the bottom of TV (monitor):

This message brought lots and lots of joy into families once it appeared on the screen. It meant a lot – CPU was working (so that guy on the flee market who sold it to you was ok.. sigh…). ROM was working (so that guy on the flee mark…). CPU and ROM were calibrated to work together, along with RAM and some other parts.

Usually the one plugged in Speccy to TV with no RAM/ROM/CPU at first, and calibrated all the circuits to reach a documented pattern on TV (for example small squares). After that was achieved, RAM/ROM and CPU were inserted one after the other by the same technique – each had to produce a certain visual pattern. So if all the parts were inserted and the message “© 1982 Sinclair Research Ltd” appeared, job was 75% done.

The last quarter was usually spent to tune all the parts to work as designed. Most of the time Speccy would restart in 3-4 minutes of use, just because something was “out of it”. To find that “something” was an art on its own. There was no Internet, no books or literature on this – it was pure work of art!

Fortunately my father is such an artist, and by the time I was 10, I got one of those Speccy beasts that my dad and I put together. The only problem though was that although it was designed to have 64Kb of RAM, it only had 16Kb. Well, 7 months later we solder more capacitors around 4Kb RAM memory chips, and that fixed it, but for 7 months I could not really play any games, because most of them required 64Kb to be there. Hence, the “Basic for Children with Arturo” book did not get a chance to get too dusty…

Where ZX-Spectrum shined the most is of course boot time – there was none Well, everything was in a ROM chip, including Sinclair Basic.

So having spent about 4-5 months to build Speccy, I was out of luck, and could not use it. Well, that is how I saw it for the first one or two weeks. But then…. I pressed a “p” key, which by default was a Basic “print” command, and then…. I typed what was on my mind: “I want to play!”, and to my astonishment Speccy “says“: “I want to play!”.

It is probably hard to surprise anybody right now by the fact that computer prints something on the screen, when you ask him/her to, but for me, being 10, and haven’t seen anything cooler than that, it was the biggest technical discovery of all times. I picked up my Basic book, and started to type, and type, and type, and then…. type a little more. That is how I started. That, in programming sense, makes me a “Speccy Child”.

The program on Basic in Speccy looked something like this:

where every line had to have a number, and the usual increment was 10. Quickly my favourite commands became “peek”, “poke” and “randomize” which worked directly with the memory (absolute addresses). And although I felt like a memory King, something was missing. I figured out what it was after I bought a system program that exposed me to something much more powerful than Basic – it was 8-bit Assembly language, the language that Zilog Z80 talked natively. Therefore me together with Z80, could control pretty much everything in the Speccy kindom. That was awesome!

Here is an example on how you’d print a line of text to the screen in Speccy’s native language:

Cool – huh!? Another thing that was very useful at the beginning of my Assembly journey, was a documented character table ( like an “ASCII” ):

The only problem was that in order to work in Assembly, or in any other language, or with any other system program – it needed to be loaded since it was not in RAM/ROM – and that took time. No, really took time. It needed to be loaded from a tape – a regular cassette that is used in a tape recorder. In fact these very tape recorders were used to load all these neat programs in:

But not only it did take time to load, it was a constant pain to tune the head of the tape recorder for each tape bought on the flee market, since nobody followed the recording standards. A lot of times, you would have to sit and hold a screw driver against the recorder’s head for 3-5, even 10 minutes, depending on how big the program was.

Here is what the screen looked like when the program was loading from tape into RAM. You can see the zebra looking lines, they were the moving loading indicator:

But time went, flee markets grew along with demand, and we upgraded to the real “RGB deal”. It was done by soldering an external controller (kind of like modern video cards, but not really) that had an RGB jack:

Although it had 9 pins, only 5 were used: Red / Green / Blue / Sync / Ground. But we did not stop on this. Soldered another external controller, but this time for a disk drive! Oh yea – that was a huuuge deal. Looking at programs to load for seconds (well in rear occasions minutes) from these beautiful 5 inch disks!

Since we had a disk drive, Speccy needed to catch up with us, and the disk controller brought another cool thing, which was TR-DOS, where, of course, DOS stands for Disk Operating System. That is how pretty it looked:

Did you notice some colour? Good catch So now it was possible to actually browse the disk, and start programs with a button click. There were tons of system utilities with different browsers that you could start at the start up time (sort of like in MBS, but not).

Right after TR-DOS, I felt like I need more. I need more.. I need more.. RAM! And we soldered another level of RAM on top of 64Kb chips. So now my Speccy was equipped with 128Kb of RAM, which could free up some ROM real estate, which enabled ROM to fit a little menu on start up:

At this point not only I could swim in the Assembly language ocean, but also play games, and use various utilities!

First let’s look at Photoshop:

well, maybe not the Photoshop you’re used to see right now, but Art Studio rocked! I was drawing away for hours!

Now, how about CuBase/ProTools:

well, maybe not exactly the latest CuBase, but I wrote lots of tunes in Wham. Some tunes I actually recorded along with playing a real guitar – it was cool!

As to games, my favourite control was my keyboard, but for many friends that came to play, I bought a joystick:

Now we could play multiplayer games like:

Match Point (Tennis) or:

Fist (Karate) game. Or even:

Tetris2 – which was a bit different from the regular Tetris in a way it gave different patterns and challenges on different levels, and was also a multiplayer game, which was of course cool.

There were tons of other games, but two I still remember are Robocop:

and Bomb Jack:

I could play long long time in Bomb Jack, not longer than I spent writing my own little games in Assembly language of course, but still – it was good amount of Bomb Jack time.

All the above was of course some time ago… 15, or even soon 20 years. Everything is a bit different now. Speccy lives somewhere in museum behind the glass:

But I remember him/her – those were times of big discoveries for me!

Like probably right now for some other 10 year old kid who just accidentally pressed a “p” button, and discovered….. time traveling…

Good luck to you kid!

Let’s say you need to do a “local integration” test, where you code does not depend on external systems (queues/external servers/DB/etc..). Creating unit tests, mocking/stubbing everything out is all good, but sometimes you need to be able to run tests that are as close as possible to the “real world” deal, while you are in “local mode” – e.g. plane, subway, basically somewhere without access to the real external systems.

One of ways you can approach it with databases is to load schemas you need in memory, and work (test) against those schemas. This is relatively easy to do with HSQLDB, and there are many “googlable” guides on how to do it. However in this little howto, I want to show you how you can define your own HSQLDB functions, and use them as either “stubs” (or even “real deal”).

Here is an example on why you may need it. Consider this Hibernate mapping file that is used in your application:

 
<?xml version="1.0"?>
 
<!DOCTYPE hibernate-mapping PUBLIC
    "-//Hibernate/Hibernate Mapping DTD 3.0//EN"
    "http://hibernate.sourceforge.net/hibernate-mapping-3.0.dtd">
 
<hibernate-mapping
	package="org.project.customer.dto.search">
	<class name="CustomerCreditCardSearchResultDto" table="CR_CREDIT_CARD">
 
		<id name="id" column="CR_CREDIT_CARD_ID">
			<generator class="native">
                                 <param name="sequence">SQ_CR_CREDIT_CARD</param>
			</generator>
		</id>
 
                <property name="accountNumber" formula="some_pkg.decode( ACCOUNT_NUMBER )" />
                <property name="secretNumber" formula="some_pkg.decode( SECRET_NUMBER )" />
                <property name="expirationDate" column="EXPIRATION_DATE" type="date" />
                <property name="zipCode" column="ZIP_CODE" />
 
	</class>
</hibernate-mapping>

Let’s say you created a CR_CREDIT_CARD table in in-memory DB, populated it, started your test. Now when you try to read a “CustomerCreditCardSearchResultDto” object somewhere in your test, Hibernate will construct an SQL query from the mapping file above, and execute it on the in-memory DB.

However there is a problem – it is going to fail with a similar Exception:

Caused by: java.sql.SQLException: Unexpected token: DECODE in statement [select <query here> .... ]
	at org.hsqldb.jdbc.Util.throwError(Unknown Source)
	at org.hsqldb.jdbc.jdbcPreparedStatement.<init>(Unknown Source)
	at org.hsqldb.jdbc.jdbcConnection.prepareStatement(Unknown Source)
	at org.apache.commons.dbcp.DelegatingConnection.prepareStatement(DelegatingConnection.java:248)
	at org.apache.commons.dbcp.PoolingDataSource$PoolGuardConnectionWrapper.prepareStatement(PoolingDataSource.java:302)
	at org.hibernate.jdbc.AbstractBatcher.getPreparedStatement(AbstractBatcher.java:442)
	at org.hibernate.jdbc.AbstractBatcher.getPreparedStatement(AbstractBatcher.java:368)
	at org.hibernate.jdbc.AbstractBatcher.prepareQueryStatement(AbstractBatcher.java:105)
	at org.hibernate.loader.Loader.prepareQueryStatement(Loader.java:1561)
	at org.hibernate.loader.Loader.doQuery(Loader.java:661)
	at org.hibernate.loader.Loader.doQueryAndInitializeNonLazyCollections(Loader.java:224)
	at org.hibernate.loader.Loader.loadEntity(Loader.java:1785)
	... 60 more

If there is an Exception – there is a Reason for it Notice these two mappings in the Hibernate mapping file above:

<property name="accountNumber" formula="some_pkg.decode( ACCOUNT_NUMBER )" />
<property name="secretNumber" formula="some_pkg.decode( SECRET_NUMBER )" />

That says Hibernate to construct SQL that uses this “some_pkg.decode()” custom function/store procedure after (in case of SELECT) reading these values from a database. However, in you local testing environment, you do not have this function defined. It is probably defined in the DB (like Oracle/DB2/etc) itself.

But for most, if not for all, technical problems there is a solution, so don’t worry – you can define this function(s) yourself. Since it is a test, the most logical thing, since you want to abstract the code out from the external systems (and their functions) would be to define stubs to these functions.

Let’s create a utility “HsqlFunctions” Java class with static methods to be used by HSQLDB (it can only use static Java methods in its SQL, btw):

package org.project.test.util;
 
/**
 * HSQL functions to be aliased.
 *
 *    DDL example:
 *
 *    CREATE ALIAS DECODE FOR "org.project.test.util.HsqlFunctions.decodeString"
 *
 *
 * @author blog.dotkam.com
 *
 */
 
public final class HsqlFunctions
{
   private HsqlFunctions()
   {
     // static utility class - does not need to be constructed.
   }
 
    /**
     * Stub for the decode function.
     * Usually used to "please" Hibernate Mapping Files.
     *
     * @param value - String value for the column
     * @return - returns the same String value
     */
    public static String decodeString( String value )
    {
        return value;
    }
}

Now, when creating a test schema, you can ALIAS this static method as HSQLDB function like this:

     CREATE ALIAS DECODE FOR "org.project.test.util.HsqlFunctions.decodeString"

This will tell HSQLDB to call “decodeString” static Java method on the column value, every time it sees “decode( COLUMN )” in SQL.

One thing to notice, though – make sure the type that the Java methods take are exactly the same as defined by DDL (Database Schema). For example, if you pass in column value as an Object:

    public static String decodeString( Object value )
    {
        return String.valueOf (value );
    }

And the column is defined as VARCHAR in schema, HSQLDB will try to pack that String into an Object, and it will fail with a similar exception:

Caused by: java.sql.SQLException: Wrong data type: hexadecimal string with odd number of characters in statement [select ... <query here>]
	at org.hsqldb.jdbc.Util.throwError(Unknown Source)
	at org.hsqldb.jdbc.jdbcPreparedStatement.executeQuery(Unknown Source)
	at org.apache.commons.dbcp.DelegatingPreparedStatement.executeQuery(DelegatingPreparedStatement.java:93)
	at org.hibernate.jdbc.AbstractBatcher.getResultSet(AbstractBatcher.java:139)
	at org.hibernate.loader.Loader.getResultSet(Loader.java:1669)
	at org.hibernate.loader.Loader.doQuery(Loader.java:662)
	at org.hibernate.loader.Loader.doQueryAndInitializeNonLazyCollections(Loader.java:224)
	at org.hibernate.loader.Loader.loadEntity(Loader.java:1785)
	... 60 more

Below is just an excerpt on how you would set up the schema to use it with HSQLDB in your test:

    ....
 
    private static final String CREATE_ALIASES =
        "CREATE ALIAS DECODE " +
        "FOR \"org.project.test.util.HsqlFunctions.decodeString\"";
 
    private static final String CREATE_SEQUENCES =
        "DROP SEQUENCE SQ_CR_CREDIT_CARD IF EXISTS;" +
 
        "CREATE SEQUENCE SQ_CR_CREDIT_CARD " +
        "START WITH 1 INCREMENT BY 1";
 
    private static final String CREATE_TABLES =
        "DROP TABLE cr_credit_card IF EXISTS;" +
 
        "CREATE TABLE cr_credit_card (" +
            "cr_credit_id NUMERIC(15), " +
            "account_number VARCHAR(48)," +
             ....
             ....
   ....
 
        //  Creating the schema
 
        m_jdbcTemplate.execute( CREATE_ALIASES );
        m_jdbcTemplate.execute( CREATE_SEQUENCES );
        m_jdbcTemplate.execute( CREATE_TABLES );
 
        //  HSQLDB is ready to be populated with data at this point.
   ....

You can use DBUnit to create the schema above and populate it with the data. The above is just a straight forward hardcoded example.
Notice how it also creates a sequence “SQ_CR_CREDIT_CARD” to please the Hibernate, and others who might use it in the application.

Happy “local integration” testing!

Feel free to post questions/comments/suggestions, I’ll try to respond when have a free second or two

Happens to the best of us, less with system admins more with developers. But we are all human, and believe it or not we DO forget and loose passwords at least once every so often. Some time ago I wrote a tutorial on how to reset lost root password in mysql, and here is another similar tutorial on how to reset the lost domain password but this time for Sun Application Server.

Before going any further with this article, please first check “.asadminprefs” file:

cat /home/toly/.asadminprefs
 
AS_ADMIN_USER=admin
AS_ADMIN_PASSWORD=YourSecretPasswordInTextHere!

the admin password could be there

If it is not there, there are two ways to reset it:

• Reinstall or recreate the affected domain.
• Create a new dummy domain and copy its security key file over to the real domain to substitute the password.

Below is an explanation for the second approach (in case when “reinstall or recreate affected domain” is not an option):

Given:

 >   Sun App server is installed in                 "/opt/SUNWappserver"
 >   Domain to which the password is lost:   "domain1"

Step 1. Creating a new dummy domain

/opt/SUNWappserver/bin/asadmin create-domain --adminport 7070 --adminuser admin --instanceport 7071 dummy-domain
 
Please enter the admin password>password
Please enter the admin password again>password
Please enter the master password>password
Please enter the master password again>password
 
Domain dummy-domain created.

Step 2. Copy dummy-domain’s “admin-keyfile” to domain1′s “admin-keyfile”

cp /opt/SUNWappserver/domains/dummy-domain/config/admin-keyfile  /opt/SUNWappserver/domains/domain1/config/admin-keyfile

now the password for domain1 is “password” – DONE

Step 3. Deleting the dummy domain

/opt/SUNWappserver/bin/asadmin delete-domain dummy-domain
Domain dummy-domain deleted.

NOTES:

The above is true for Sun’s Application Server 8.x and later.

For Sun’s Application Server 9.x check out “change-admin-password“