EasyMock gotcha: you can't use argument matchers as return value

I was getting some really nasty error messages while running my unit tests. A test that used EasyMock (using 2.0) failed but it wasn't anything in the actual test, it was caused by a test that ran before it. When skipping the test case the next test case using EasyMock would fail.

The error I was getting:

java.lang.IllegalStateException: 2 matchers expected, 1 recorded.
   at org.easymock.internal.ExpectedInvocation.createMissingMatchers(ExpectedInvocation.java:41)
   at org.easymock.internal.ExpectedInvocation.(ExpectedInvocation.java:33)
   at org.easymock.internal.ExpectedInvocation.(ExpectedInvocation.java:26)
   at org.easymock.internal.RecordState.invoke(RecordState.java:63)
   at org.easymock.internal.MockInvocationHandler.invoke(MockInvocationHandler.java:24)
   at org.easymock.internal.ObjectMethodsFilter.invoke(ObjectMethodsFilter.java:45)
   at $Proxy88.handleRequest(Unknown Source)

First off, to find the offending test I build a test suite and reduced the tests run to isolate the tests where the error occurred until I had found the one causing the problem.

The code looked like this:

       JdbcOperations jdbcOperationsMock = createMock(JdbcOperations.class);

       // Expect query to use supplied alternative schema
       expect(
               jdbcOperationsMock.query(
                       eq("SELECT ARTICLE_SKU, ARTICLE_NAME FROM alt_schema_subs.tomtomtbarticle;"),
                       isA(RowMapper.class))
               )
               .andReturn(isA(List.class));

Which seems to be ok. The only problem was that I was using a argument matcher on a return value (see documentation). The solution was to supply a normal return value:

  expect(
    jdbcOperationsMock.query(
      eq("SELECT ARTICLE_SKU, ARTICLE_NAME FROM alt_schema_subs.tomtomtbarticle;"),
      isA(RowMapper.class))
    )
    .andReturn(new ArrayList<Object>());

And now the tests succeed.

Hope this was helpful. I intent to update the blog with more day-to-day problems in hope some people might find it on google.

Keeping it up

Oops, then I suddenly stopped updating the blog. This is often the case with me, start something enthusiastically which slowly dies soon after that. Keeping something on going after the start is quite hard.

Anyway, some wise words I heard today at work: "it's worse to make no decision then to make a bad decision". The logic behind that is if you make a bad decision you can't always identify that and change, you learn something. Whereas when you make no decision everything will just drop and nothing will happen. You have to keep on moving forward to keep the pace.

Criticizing is easy, doing is hard

Sometimes I fall into the trap of overly criticizing and asking annoying questions to people who are presenting their work. It's very easy to spot that little grey dot on a white piece of paper. Harder it is to create something completely clean and spotless. And how annoying it is when you've put a lot of effort into something and people only whine about the few little things that are not 100% perfect.

The best solution to participate and do the work yourself. You're not going to heavily criticize yourself. And it shows the reasons why things are not perfect, because it's hard to achieve.

That's why I'm happy with the suggestion of my manager to participate in some of the handover meetings we've been having. I was asking a lot of critical questions, because I like doing that. It's to test my own knowledge and I also like to be right. But only being critical is not the way to go. I'll be more careful.

Wordle cloud

This little toy Wordle is quite cool. Check my word cloud:

Procrastination

The last few days I've been working on a new website cms for Sanne. It's strange that I totally want to build this site for my girl. But something inside me is strongly resisting. Getting started is so hard. Once you start you get going. I still have to teach myself to break this feeling because this is painful.

This letter to a young procrastinator is very striking.

Inbox Zen

Something I've been following for a while: the inbox zero principle. It's quite simple: always have an empty inbox. When you check your mail categorize each item until the inbox is empty. You can either:

• Delete
• Delegate
• Respond
  
• Defer
• Do

The great advantage of this is: less worrying, more doing. Every time you see the rising queue in your mailbox you get demotivated by the amount of work left. If you handle everything you can right away and schedule everything else you save time you can spend on other things!

One other thing I particulary like is having a 'waiting for' folder. Here I move all the mail I have send that I'm expecting a reply to. This way there's no more need to remember if I received a reply. Once a day or so I check my 'waiting for' and send a reminder to people if I think they're taking too long. It's amazing how many people (sorry to say, especially managers) fail to reply to an explicit question in a mail.

Check out this great talk about inbox zero by Merlin Mann. He writes 47 folders and also co-hosts the hilarious podcast You Look Nice Today.

Unit testing hooray!

Slowly I'm getting more into unit testing. It's hard to enforce the discipline to write the tests during development because writing tests takes more time and managers don't like things taking time. But they should be written during development because:
a) That's when you are most into the problem and thus best able to write the test
b) It makes you think about the code you are writing, perhaps discovering bugs as you create them
c) After the fact - there is never time to write tests anyway so it won't happen

When unit tests have been written the benefits are so marvelous. You can re-run them any time and be be sure all your app is still ok. It takes away the fear of changing things because they might break, just run the tests and you are sure you didn't break anything. Beats manually re-testing by a mile.

Finally, I've been getting into EasyMock to mock out the dependencies of my classes. Works great once you get used to it. Say you have a Controller class that needs a Service to search for something:

class Controller {
  private final SearchService searcher;
  public Controller(SearchService searcher) {
    this.searcher = searcher;
  }
}

interface SearchService {
  List doSearch(String query);
}

class SearchServiceImpl implements SearchService {
  List doSearch(String query) {
    // Searching code here
  }
}

Now to unit test the Controller just mock the SearchService:

@Test
public void testController() {
  SearchService searchMock = EasyMock.createMock(SearchService.class);
  EasyMock.expect(searchMock.doSearch("something")).andReturn(isA(List.class));
  EasyMock.replay(searchMock);
  EasyMock.verify(searchMock);
}

And the test can run without an implementation needed. And as a bonus you can check if the mock is called in the expected way.

Btw, I have to find some code syntax highlighting for Blogger.

Using HibernateInterceptor to make database schema name configurable

To fit in our DTAP (Development Test Acceptance Production) cycle I needed to make our application database schema name configurable. The idea is that multiple instances of the application and database can run side-by-side on the same server so each developer and tester can have their own environment if needed.

The problem I quickly realized is that our application uses two different schemas but only one Hibernate Session and the configuration is done using annotations where the schema name for each bean is hardcoded on the @Table annotation. No chance to replace that at runtime, except maybe messing with Reflection which I'm not going to get into now.

So the solution I went with was replacing the schema name on the query just before it is send to the DB using a HibernateInterceptor as described in this blog entry.

Here's how the solution looked in short:

public class ConfigurableSchemaNameHibernateInterceptor extends EmptyInterceptor {
  public static final String DB_SCHEMA1_PATTERN = "schema1";
  public static final String DB_SCHEMA2_PATTERN = "schema2";

  @Override
  public String onPrepareStatement(String sql) {
    String prepedStatement = super.onPrepareStatement(sql);
    prepedStatement = prepedStatement.replaceAll("(?i)" + DB_SCHEMA1_PATTERN, schema1ConfiguredName);
    prepedStatement = prepedStatement.replaceAll("(?i)" + DB_SCHEMA2_PATTERN, schema1ConfiguredName);

    logger.debug("Query replaced, old sql: " + sql);
    logger.debug("Query replaced, new sql: " + prepedStatement);

    return prepedStatement;
  }
}

Maybe this is not too clean and a bit error prone (could accidently replace some genuine content..). But I learned some new things about Hibernate so I'm blogging this anyway. Next time I'll refactor it to use multiple Hibernate sessions and no harcoded schema name.

Productivity and being watched

A thing I noticed is that I am way more productive when I'm working directly with someone. If I'm sitting side to side I don't take all the long way to get somewhere. This person wants results, and wants them now. It amazes me how fast and productive I can be then. Results also motive me a great deal, encourage me to carry on.

I guess I tend to wander off sometimes while working. I discover a problem or something that can be improve. I dive into that. Soon I find something else interesting. It's a recursive process. In the end overall progress is slow.

It's just me being like this though. It's all within myself so I should be able to force myself to work faster and more goal oriented to achieve great results.

It's not also good to take the shortest route though. This usually means using things I already know instead of learning something new. And I tend to make shortcuts and make unclean solutions. That's great for quickly having something tangible but not so great so long lasting quality results. The ideal way would be in the middle somewhere.

Joining on a subquery

Pretty awesome SQL trick I just learned from Wouter. In SQL you can join on a subquery and then you can even select columns from it! The cool thing is that you can do transformations such as Group By in this subquery. This enables things I could not do before such as multiple count columns in one query.

For example:

SELECT account.IDENTIFIER, subscriptionCount.subscriptions, promotionCount.promotions
FROM tomtomtbaccount account
JOIN (SELECT account_fk, COUNT(*) AS subscriptions FROM tomtomtbsubscription GROUP BY account_fk) subscriptionCount ON account.account_id = subscriptionCount.account_fk
JOIN (SELECT account_fk, COUNT(*) AS promotions FROM tomtomtbpromotion GROUP BY account_fk) promotionCount ON account.account_id = promotionCount.account_fk
ORDER BY subscriptionCount.subscriptions
LIMIT 100;

(this query is a bit heavy though...)

Also a nice memonic Wouter provided:
if you want results side by side: JOIN
if you want results in a list: UNION

Go see people

When small problems or misunderstandings occur, just go see the people in person instead of sending mail. The mail communication can cause even more confusion. People in person are generally kinder and have more attention.

Today I was helping a colleague with setting up my application on a test environment. He was complaining so i sat with him and we quickly solved the problem. We found another problem and solved it on the spot too. Within half an hour it was all running. It would have taken a lot of more time and frustration for him to solve all the problems. Now I was giving him attention and thereby promotion myself.

Controlling your boss for fun and profit

Very good article on influencing without power:
http://blogs.msdn.com/eric_brechner/archive/2005/08/01/august-1-2005-controlling-your-boss-for-fun-and-profit.aspx

Basically: don't whine about your manager not listening. You're telling it the wrong way and do something about it if you want to make a change.

Don't be afraid to ask

Don't spend hours staring at the screen. Ask advice. People like being asked for advice, it makes them feel knowledge-able. I can learn a lot from their experience. Other people often have a different view that can bring more ideas. Even explaining the problem is a way of ordering and thinking about it. Not rarely I explain a problem and come up with a solution myself on the spot. Just because I took some distance and thought about it.

Create action lists

Write everything todo down so I don't have to think about it. Spend my braintime on actually doing something.

Testing

A good presentation on unit testing. I should really set up tests from the start. It saves a lot of work in maintenance.

http://www.masukomi.org/talks/unit_testing_talk_2/index.xul?data=slide_data.txt#page1

Humblesness

"Everyone writes bad code from time to time, but having code reviews eliminates a lot of it. Always try to be nice to your colleagues, and never think that you know better. Be humble since it's actually easy to be wrong about stuff. :-)"

http://beta.stackoverflow.com/questions/12745/how-do-you-handle-poor-quality-code-from-team-members#12761

I sometimes tend to get arrogant or think I know it all. I should be careful with this. It's easy to be wrong and not very helpful or constructive being a 'know it all'. I don't like people who act like this so shouldn't do it myself.