September 21, 2010

Getting rid of null parameters with a simple spring aspect


What is the most hated and at the same time the most popular exception in the world?

I bet it's the NullPointerException.

NullPointerException can mean anything, from simple “ups, I didn't think that can be null” to hours and days of debugging of third-party libraries (try using Dozer for complicated transformations, I dare you).

The funny thing is, it's trivial to get rid of all the NullPointerExceptions in your code. This triviality is a side effect of a technique called “Design by Contract”.

I won't go into much details about the theory, you can find everything you need on Wikipedia, but in the nutshell Design by Contract means:
  • each method has a precondition (what it expects before being called)
  • each method has a postcondition (what it guarantees, what is returned)
  • each class has an constraint on its state (class invariant)

So at the beginning of each method you check whether preconditions are met, at the end, whether postconditions and invariant are met, and if something's wrong you throw an exception saying what is wrong.

Using Spring's internal static methods that throw appropriate exceptions (IllegalArgumentException), it can look something like this:
import static org.springframework.util.Assert.notNull;
import static org.springframework.util.StringUtils.hasText;

public class BranchCreator {
    public Story createNewBranch(Story story, User user, String title) {
        verifyParameters(story, user, title);
        Story branch = //... the body of the class returnig an object
        verifyRetunedValue(branch);
        return branch;
    }

    private void verifyParameters(Story story, User user, String title) {
        notNull(story);
        notNull(user);
        hasText(title);
    }

    private void verifyRetunedValue(Story branch) {
        notNull(branch);
    }
}

You can also use Validate class from apache commons instead of spring's notNull/hasText.

Usually I just check preconditions and write tests for postconditions and constraints. But still, this is all boiler plate code. To move it out of your class, you can use many Design by Contract libraries, for example SpringContracts, or Contract4J. Either way you end up checking the preconditions on every public method.

And guess what? Except for Data Transfer Objects and some setters, every public method I write expects its parameters NOT to be null.

So to save us some writing of this boiler plate ocde, how about adding a simple aspect that will make it impossible in the whole application, to pass null to anything other than DTOs and setters? Without any additional libraries (I assume you are already using Spring Framework), annotations, and what else.

Why would I like to not allow for nulls in parameters? Because we have method overloading in modern languages. Seriously, how often do you want to see something like this:

Address address = AddressFactory.create(null, null, null, null);


And this is not much better either

Microsoft.Office.Interop.Excel.Workbook theWorkbook = ExcelObj.Workbooks.Open(openFileDialog.FileName, Type.Missing, Type.Missing, Type.Missing, Type.Missing, Type.Missing, Type.Missing, Type.Missing, Type.Missing, Type.Missing, Type.Missing, Type.Missing, Type.Missing, Type.Missing, Type.Missing);

The solution

So here is a simple solution: you add one class to your project and a few lines of spring IoC configuration.

The class (aspect) looks like this:
import org.aspectj.lang.JoinPoint;
import static org.springframework.util.Assert.notNull;

public class NotNullParametersAspect {
    public void throwExceptionIfParametersAreNull(JoinPoint joinPoint) {
        for(Object argument : joinPoint.getArgs()) {
            notNull(argument);
        }
    }
}

And the spring configuration is here (remember to change the namespace to your project).
<aop:config proxy-target-class="true"> 
    <aop:aspect ref="notNullParametersAspect">
        <aop:pointcut expression="execution(public * eu.solidcraft.*..*.*(..))
                          &amp;&amp; !execution(public * eu.solidcraft.*..*Dto.*(..))
                          &amp;&amp; !execution(public * eu.solidcraft.*..*.set*(..))" id="allPublicApplicationOperationsExceptDtoAndSetters"> 
            <aop:before method="throwExceptionIfParametersAreNull" pointcut-ref="allPublicApplicationOperationsExceptDtoAndSetters"></aop:before>     
        </aop:pointcut> 
 
        <task:annotation-driven>
            <bean class="eu.solidcraft.aspects.NotNullParametersAspect" id="notNullParametersAspect"></bean>
        </task:annotation-driven>
    </aop:aspect>
</aop:config>

The "&amp;&amp;" is no error, it's just && condition escaped in xml. If you do not understand aspectj pointcut definition syntaxt, here is a little cheat sheet.

And here is a test telling us that the configuration is succesfull.

public class NotNullParametersAspectIntegrationTest extends AbstractIntegrationTest {
    @Resource(name = "userFeedbackFacade")
    private UserFeedbackFacade userFeedbackFacade;

    @Test(expected = IllegalArgumentException.class)
    public void shouldThrowExceptionIfParametersAreNull() {
        //when
        userFeedbackFacade.sendFeedback(null);

        //then exception is thrown
    }

    @Test
    public void shouldNotThrowExceptionForNullParametersOnDto() {
        //when
        UserBookmarkDto userBookmarkDto = new UserBookmarkDto();
        userBookmarkDto.withChapter(null);
        StoryAncestorDto ancestorDto = new StoryAncestorDto(null, null, null, null);

        //then no exception is thrown
    }
} 
AbstractIntegrationTest is a simple class that starts the spring test context. You can use AbstractTransactionalJUnit4SpringContextTests with @ContextConfiguration(..) instead.

The catch

Ah yes, there is a catch. Since spring AOP uses either J2SE dynamic proxies basing on an interface or aspectj CGLIB proxies, every class will either need an interface (for simple proxy based aspect weaving) or a constructor without any parameters (for cglib weaving). The good news is that the constructor can be private.

September 6, 2010

Agile Skills Project at my company

Unfulfilled programmers


Erich Fromm, a famous humanist, philosopher and psychologist strongly believed that people are basically good. If he was right, then either our society is a mind-breaking dystopia or we have a great misfortune of working in a field that burns people out, because many IT people I know are more like Al Bundy than anyone else.

Why is being a couch potato something wrong? Happiness can be achieved in many different ways, but not by passive pleasures. One way of pursuing happiness is by self realization and while self realization can happen in any activity, it's makes perfect sense to have it at work, where you spend one third of your life time anyway.

But many developers I know, consider work as something boring at best, dreadful at worst. True, programming can be awful, when you have to dig deep into a terrible code base without any perspective for a change, but IT is vast and you can always find something interesting, and once you learn it, you will find a way to make money on it, either by changing position inside your company or changing your employer altogether.

Yet most unhappy IT professionals don't do anything to change their situation. The main reason for that is, because it requires a lot of learning, and learning at home is not the most beloved activity for a couch potato.

So why are developers turning into couch potatoes in the first place? Why the last thing a typical developer will do back home is learning and polishing his skills? There are plenty of reasons for that.

The three main roots of an IT couch potato


First, our work is tiresome. Nearly every job offer you can find mentions “able to work under pressure” and “flexible long working hours” in the requirements. This translates directly to the “burn-out” phenomenon.

Second, the technological landscape is changing overwhelming fast. Unless you work for a slowly adapting institution like a bank, your skills will be outdated in a few years time. Sure the deceased Sun, god help us, granted Java developers four years of relative stagnation, but that's an exception and it's going to end soon enough anyway (unless, of course, these are just convulsions before slow death of technology). You better learn and you better learn fast, or you'll have no other option than to promote yourself into management.

Third, just how long can you sit by your computer everyday? Yeah, I know, some people spend years playing WoW, Eve and alike, barely moving. I am a sinner myself, with Steam reporting over 350 hours in Modern Warfare 2, 200 hours in F.E.A.R. 2 multi, and countless months of my life wasted by Sid's Civilization. But for not-addicted, it's just simply stupid, not to mention unhealthy, to have your ass integrated with the chair. No matter how comfortable it may be. There is more to life than that.

Case Study at my company


It all started with a few SQL programmers grumbling about how they are bored to death, and how they would like to switch to OO programming. I'm not a person who waits, so next thing I did was asking our management if they could move those guys to Java/C# projects. And the management was all for it, with just one requirement: they would have to first learn our technology stack at home, not to be totally lost and unproductive. After all, the more technologies an employee know, the more valuable he is for the employer (think about switching people between projects).

A few months later and nothing has changed. I'm asking sql guys how the learning is going, and I get the answer: it hasn't started yet.

Now, I know the best way to learn something is by hands-on experience at work. After all that's why I've been changing my job a few times: to have a real world experience. It's easier to learn french if you move to Paris. And learning at home is hard because of the aforementioned reasons. The very same reasons, why you get only 650 people on a free conference, like Javarsovia.

So what can we do, then? How about we remove all the obstacles? How about we make learning at home fun, satisfying and profitable. How about we provide  motivation and feedback. How about we also solve the never-ending dissonance between employee's financial and employer's productivity expectations on the way. Sounds interesting? Let's try, then.

First: make it profitable.


Up to some point, people get motivated by money. It won't work if you are already earning enough to pay for everything you need, but in a country like Poland, to be able to build/buy yourself a house, you have to be making many times the average salary. So here, money is still a major motivator. Every year, every developer goes back to his boss and says: I want more.

Guess what, your boss wants to pay you more. No kidding. After all Henry Ford's said:
“There is one rule for industrialists and that is: make the best quality of goods possible at the lowest cost possible, paying the highest wages possible”.

Highest wages. You boss really wants to pay more. But to stay in business, the company needs you to either improve the quality or productivity. Both mean more money to the business, and more money to pay you with. If you consider that, the goal of a developer who wants to learn something new (or get better at something) is on the way to make the company more profitable. After all, this is software development – you never know what technology you gonna need tomorrow.

This works both for completely new stuff, and for learning something that company is already quite good at. If you know several technologies that the company is working with, you are more valuable, because you can handle more projects (you boss may think in terms of reallocating resources).

After thinking about all of this I went to my bosses and asked them: will you pay more to the people who learn different technologies at home? Even when they can't use them right now at work? Will you give a rise to those Oracle guys, if they learn Java?

The answer was: definitely! They actually said that every time a developer asks for a rise, they ask him back: what have you done in the last year to improve your market value? What have you learned? Because every time an employee's market value increases, the company's value increases.

Simply speaking: more skills means more money. Both for the developer and for the company. It's amazing, how often people forget about it. Making it crystal clear can give you a motivational boost to do something at home and a nice perspective. You don't have to switch your job to get a rise. You need to learn more, and they'll happily pay you more.

Second: make it easy


The main question with learning is where to start from? And for software developers it's the most important, most difficult question, because there is no way to learn everything, because spending years studying can be simply a waste of time, if the technology is dead/outdated the moment you get productive with it. What should I learn or why should I learn at all, if the risk of wasting the most precious thing in my life, my time, is so high? How do I decide what to learn.

Lets make learning safe and easy.

Your best bet is to start with something that has much longer life expectancy, something that will help you right away, no matter what a technology you have to work with in your next project. And something that is relatively simple to learn: agile skills.

These are skills of an agile developer, well established, well recognized, and not going away any time soon, because we still do not have anything on the horizon that could surpass them.

Yeah, I know, the world 'agile' is so popular nowadays, that even my grandma is agile, but lets go for a solid list of things agile. No bullshit theory, no marketing mumbo jumbo, give me a precise, distilled and refined list of things I should learn, things that will help me, things worth spending my time on.

Here it is:

The Agile Skills Project


The project is all about self improvement and learning. It's a great inventory of “ isolated, learned, practiced, and refined” agile skills, with definitions, resources, descriptions of steps to mastery and success stories. Take a look at the “Pair Programming” page, for example.

All the skills are divided into different areas: Business Value, Collaboration, Confidence, Product, Self Improvement, Supportive Culture, Technical Excellence. You even get a nice mind-map with it.

This is a single reference point for all those who do not know where to start or where to go next. All these skills are in high demand on the market and with a very long Time-To-Live. The best thing is though: no matter what technology you gonna work with tomorrow, you can benefit from them.

I took the list from the website, tidied it up a bit, refactored it for the needs of my company, and proposed it as a Request For Comment, a wiki page, where everyone gets to discuss and shape up the idea, before we give it to the management.

Soon we had a discussion. It wasn't easy to make everyone understand the concept, but after a while people joined in, and we added some more stuff.

Level up!


The Agile Skills Project is more than a simple index. It tries to create a learning ecosystem, by defining quests:
"Quests" are on-the-job experiments, self-assessments, peer-reviews, course experiences or other activities intended to help a person better apply a particular agile developer skill set.

It's a bit like a Role Playing Game. You have your quests, you do them, you get experience. For experience you get more money and new toys (technologies) to play with.  It's fun. Billions of MMORPG players cannot be wrong.

But to make that happen we need something every game has: feedback.

Third: give feedback


OK, so we have a bit of motivation (money) and a list of goals (agile skills). Who is going to give us quests, and who is going to tell us we did a good job? How will we have our feedback?

The first and most important thing, is to see the results of you actions. Otherwise you loose focus and motivation (money can only get you that far). Therefore you should create a list of quests you have done. Put it on the intranet or somewhere, where you can show it to others. It's important, because you are going to share it with your mentor.

Yes, a mentor. Choose someone from your company, someone you trust, someone you respect. It doesn't have to be an Einstein. Meet with this person once a month, during your work-time. An hour should do. Discuss with your mentor what quests you want to accomplish this month. Could be anything, reading an IT book, learning new programming language or taking another step to master one of the agile skills from the list. Tell your mentor when you'll be done. Meet together again next month, and either put the quest in your done-list, or mark it as 'failed'.

The role of the mentor is to listen to you, remove obstacles, help you choose a good path and give you feedback.

You'll be surprised by how much the meeting with your mentor motivates you. It works much better than money: you don't want to fail in the eyes of the mentor, because this is the guy you respect, and you want him to respect you as well. And once you see your constant improvement by filling the list of quests done with your mentor, it gets addictive.

Smells corporate?


How is this any different to what you can sometimes see in a corporation, with a year long plan of tasks your boss is giving you to accomplish to get your bonus?

Well, first of all, these will be your quests, chosen by you. Second, you will choose your mentor as well. Your boss usually doesn't know a thing about what you are doing. Third, it's all about your self-improvement, not meeting some company goals. You get better at something, the company gets better at something. After all a company is not much more than the people working at it.  Fourth, it's a fast feedback cycle, you do not have to wait till the end of the year to get it.

And finally, it may be a bit corporate, because I have never seen any small company doing anything like this. But even if it is, it still seems like worthwhile. Anything to get me out of the couch.

Discuss


It's a bit too early to tell whether the idea will be successful. We have just started. Fo me it is already helpfull, because with a list of quests done I have have a feeling of progress. If you'd like to discuss this, and other ways to animate software developers to do something more, I'm leading a meeting at Agile Warsaw group about it, on the 20th of September, 19:00. Feel invited.

By the way, here you have a trial of "other ways to animate" from our internal TouK Code Jam Party, we held a week ago. Doesn't look mych corporate, does it?