what is codenvy exactly? well, their website states: codenvy is a cloud environment for coding, building, and debugging apps. basically, it’s an ide in the cloud (“ide as a service?”) accessible by all the major browsers . it started out as an additional feature to the exo platform in early 2009 and gained a lot of traction after the first paas (openshift) and git integration was added mid-2011. codenvy targets me as a (java) software developer to run and debug applications in their hosted cloud ide, while being able to share and collaborate during development and finally publish to a repository – e.g. git – or a number of deployment platforms – e.g. amazon, openshift or google app engine. i first encountered their booth at javaone last september, but they couldn’t demo their product right there on the spot over the wifi, because their on-line demo workspace never finished loading well i got the t-shirt instead then, but now’s the time to see what codenvy has in store as a cloud ide. signing up signing up took 3 seconds. all you have to do is go to codenvy.com , use the “sign up” button, choose an email address and a name for your workspace , confirm the email they’ll send you and you’re done. the “workspace” holds all your projects and is part of the url codenvy will create for you, like “ https://codenvy.com/ide/ . although not very clear during the registration process – which of course nowadays is usually minimalistic as can be – it seems that i’ve signed up for codenvy’s free community plan , which gives me an unlimited number of public projects. you can even start coding without registration. after confirming the registration mail, i’m in. finally i’ll end up in the browser where your (empty) workspace has been opened. empty workspace a few options a possible for here on, as seen in the figure above: create a new project from scratch – generate an empty project from predefined project types import from github – import projects from your github account clone a git repository – create a new project from any public git reposiroty browse documentation invite people – get team members on board support – questions, feedback and troubleshooting let’s… create a new project from scratch this option allows you to name the new project – e.g. “myproject”, choose a technology and a paas . the technology is a defined set of languages of frameworks to develop with. available technologies at the moment the technologies are: java jar java war java spring javascript ruby on rails python php node.js android maven multi-module at the time of writing java 1.6 is supported. available paas at the moment the available platforms are: amazon webservices (aws) elastic beanstalk savvis cloud appfrog cloudbees google app engine (gae) heroku manymo android emulator red hat’s openshift none depending on the choice of technology, or or more paas options become available. a single jar can not be deployed onto any of the platforms, leaving only the option “none” available. a java web application (war) can be deployed onto any number of platforms, except heroku and manymo. node.js can only be deployed to openshift. creating a simple jar project after having selected a jar (and no platform) one can select a project template . e.g. if webapplication (war) would have been selected, codenvy would present project templates, such as google app engine java project illustrating simple examples that use the search api , java web project with datasource usage or a demonstration of accessing amazon s3 buckets using the java sdk . the jar technology has only one project: simple jar project . after having finished the wizard, our jar project has been created in our workspace. we’ll see two views of our project: a project explorer and a package explorer. project- and package explorer what we can see is that our jar project has been given a maven pom.xml with the following content: view source print ? 01. < project xmlns = " http://maven.apache.org/pom/4.0.0 " xmlns:xsi = " http://www.w3.org/2001/xmlschema-instance " 02. xsi:schemalocation = " http://maven.apache.org/pom/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd " > 03. < modelversion >4.0.0 04. < groupid >com.codenvy.workspaceyug8g52wjwb5im13 05. < artifactid >testjarproject 06. < version >1.0-snapshot 07. < packaging >jar 08. 09. < name >sample-lib 10. 11. < properties > 12. < project.build.sourceencoding >utf-8 13. 14. 15. < dependencies > 16. < dependency > 17. < groupid >junit 18. < artifactid >junit 19. < version >3.8.1 20. < scope >test 21. 22. 23. we have a generated group id com.codenvy.workspaceyug8g52wjwb5im13 , our own artifact id and the junit dependency, which is a decent choice for many java developers use it as a testing framework. the source encoding has already been set to utf-8, which is also a sensible choice. as a convenience we’ve also been given a hello.sayhello class, so we know we’re actually in a java project say hello file & project management so what about the browser-based editor we’re working in? on top we’re seeing a few menu’s, like file, project, edit, view, run, git, paas, window, share and help . i’ll be highlighting a few. file- and project menu the file menu allows to creating folders , packages and various kind of filetypes , such as text, xml (1.0 at time of writing) , html (4.1) , css (2.0), java classes and jsp’s (2.1). although i’m in a jar project, i am still also able to create here e.g. ruby, php or python files. a very convenient feature is to upload existing files to the workspace, either separately or in zip archives. i’ve tried dropping a file onto the package explorer from the file system, but the browser (in this case, chrome) tries to open it instead the project menu allows to create new projects, either by launching the create project wizard again, but also allows for importing from github . in order to clone a repository, you’ll have to authorize codenvy to access github.com to be able to import a project. after having authorized github, codeenvy presents me with a list of projects to choose from. after having imported all necessary stuff, it somehow needs to know what kind of project i’m importing. selecting a file type after importing a project from github the project i imported didn’t give codenvy any clues as to what kind of project it is (which is right since i only had a readme.md in it), so it lists a few options to choose from. i chose the maven multi-module type after which the output window shows: [email protected]:tvinke/examples.git was successfully cloned. [info] project type updated. if you’d have a pom.xml in the root of your project, it would immediately recognize it a s a maven project. apart from going through the project > import from github option, you can also go directly to the git menu, and choose clone repository . this allows you to manually enter the remote repository uri, wanted project name and the remote name (e.g. “origin”). cloning a repository one you have pulled in a git project, the git menu allows all kinds of common operations, such as adding and removing files, committing, pushing, pulling and much more. git menu the ssh keys can be found under menu window > preferences where you can view the github.com entry, where one can view the details or delete it. also a new key can be either generated or uploaded here. sharing the project one of the unique selling points of codenvy are their collaboration possibilities which come along with any project. you can: invite other developers with read-only rights or full read-write rights to your workspace and every project in it.when you’re pair-programming like this, or co-editing a file with a colleague, you can also send each other code pointers – small shortcuts to code lines. use factories to create temporary workspaces , through cloning, off one source project (“factory”) and represent the cloning mechanism as a url which can be given to other developers. a use case might be to get a colleague quickly started on a project by providing a fully working development environment.there’s a lot more about creating factories in the docs (such as through rest), but the nice thing is that once you have a factory url, you can embed it as a button, send it through email of publish it somewhere for others! a factory url to load up e.g. their twitter bootstrap sample – as they use on their website themselves – looks like: https://codenvy.com/factory?v=1.0&pname=sample-twitterbootstrap&wname=codenvy-factories&vcs=git&vcsurl=http%3a%2f%2fcodenvy.com%2fgit%2f04%2f0f%2f7f%2fworkspacegcpv6cdxy1q34n1i%2fsample-twitterbootstrap&idcommit=c1443ecea63471f5797f172c081cd802bac6e6b0&action=openproject&ptype=javascript conclusion applications are run in the cloud nowadays, so why not create them there too? codenvy brings some interesting features, such as being able to instantly provision workspaces (through factory urls) and share projects in real-time. it supports common operations with projects, files and version control. with a slew of languages and platforms and as an ide being always accessible through the internet, it could lower the barrier to actually code anytime and anywhere. in a future post i will try and see whether or not it can actually replace my conventional desktop ide for java development.

Is Tomcat is a web server or an application server? Let me tell you how I convinced my self regarding this.

I've got an application (OK, scratty PHP script) that glues together some API things and shows them onto a dashboard for me. Recently, I updated it to use Guzzle as the consuming client, since Twitter now needs me to authenticate (I wrote about that if you're interested), and I used Composer to bring the new library in. It was very simple, so I thought I'd share it since it's quite a minimal example, and those are my favorite kind. Getting Started with Composer First of all, you'll need to either install composer, or at least get the .phar -- these examples use the composer.phar file from under the "Manual Download" section. Composer comes with instructions, for a list of command types: php composer.phar list Once you've got Composer ready to go, we can move on and configure it. State Your Dependencies All you need to do to use Composer is to create its configuration file and then tell it to do its magic. :) All I want here is the Guzzle library, so my config file is rather simple: { "require": { "guzzle/guzzle": "3.7" } } Get the Libraries From this point, Composer can do the rest by itself. It will fetch all the libraries you set in the config file, but will also get any dependencies of those libraries. To ask it to do this, just type: php composer.phar install This will download all the libraries you need and place them in a vendor directory. In the case of Guzzle, it depends on other libraries (from the Symfony project), so my vendor directory now contains: . ├── autoload.php ├── composer ├── guzzle └── symfony And there you have it: all the libraries you need are in place. Tell Your Application About Its Dependencies One more step: just add the following line into your application to make those new libraries available to it. require 'vendor/autoload.php'; There you go. :) Enjoy using Composer even in existing/legacy applications, projects that aren't built in "the one true way" or really anywhere else you don't want to have to include libraries in your repositories. Oh, and one more tip if you're just getting started: Add the vendor folder to your source control "ignore" file. Further Reading Nice post from Cal about including non-composer libs in your autoloader: http://blog.calevans.com/2013/07/21/using-3rd-party-libraries-in-composer-projects/ Composer documentation: http:// http://getcomposer.org/doc/ Further Reading Upgrading PHP Tools for Better PHP Twitter Search API Using PHP and Guzzle PSR-What?

We spend more time on reading code than writing. So if the code is more readable then obviously it will increase the developer productivity. Many people associate readability of code with coding conventions like following standard naming conventions, closing file, DB resources etc etc. When it comes to code reviews most of the people focus on these trivial things only, like checking for naming convention violations, properly releasing resources in finally block or not. Do we need "Senior Resources" in team (I hate to call a human being as a Resource) to do these things?Tools like Findbugs, PMD, Checkstyle, Sonar can do that for you. I agree that following a standard naming convention by all the team members is a good thing. But that doesn't increase the readability of code. Let us take a simple example. I would like to implement Fund Transfer usecase and following are the rules to implement: Source and target accounts should be valid accounts Check whether source account has sufficient amount Check whether source account has provision for Overdraft and check whether this transaction exceeds the overdraft limit Check for duplicate transaction with last transaction. If source, target accounts and amount is same with last transaction consider it as a duplicate transaction If everything is fine then transfer amount to target account Assume we have the following implementation for the above usecase: public class UglyMoneyTransferService { public void transferFunds(Account source, Account target, BigDecimal amount, boolean allowDuplicateTxn) throws IllegalArgumentException, RuntimeException { Connection conn = null; try { conn = DBUtils.getConnection(); PreparedStatement pstmt = conn.prepareStatement("Select * from accounts where acno = ?"); pstmt.setString(1, source.getAcno()); ResultSet rs = pstmt.executeQuery(); Account sourceAccount = null; if(rs.next()) { sourceAccount = new Account(); //populate account properties from ResultSet } if(sourceAccount == null){ throw new IllegalArgumentException("Invalid Source ACNO"); } Account targetAccount = null; pstmt.setString(1, target.getAcno()); rs = pstmt.executeQuery(); if(rs.next()) { targetAccount = new Account(); //populate account properties from ResultSet } if(targetAccount == null){ throw new IllegalArgumentException("Invalid Target ACNO"); } if(!sourceAccount.isOverdraftAllowed()) { if((sourceAccount.getBalance() - amount) < 0) { throw new RuntimeException("Insufficient Balance"); } } else { if(((sourceAccount.getBalance()+sourceAccount.getOverdraftLimit()) - amount) < 0) { throw new RuntimeException("Insufficient Balance, Exceeding Overdraft Limit"); } } AccountTransaction lastTxn = .. ; //JDBC code to obtain last transaction of sourceAccount if(lastTxn != null) { if(lastTxn.getTargetAcno().equals(targetAccount.getAcno()) && lastTxn.getAmount() == amount && !allowDuplicateTxn) { throw new RuntimeException("Duplicate transaction exception");//ask for confirmation and proceed } } sourceAccount.debit(amount); targetAccount.credit(amount); TransactionService.saveTransaction(source, target, amount); } catch(Exception e){ logger.error("",e); } finally { try { conn.close(); } catch(Exception e){ //Not everything is in your control..sometimes we have to believe in GOD/JamesGosling and proceed } } } } The above code is readable..right??...because: 1. We have followed naming conventions like camel casing variable names 2. We have all the open braces ({) on the method definition line 3. We have closed DB Connection in finally block 4. we have logged exception instead of using System.err.println() and most important, it is working as expected. So is it readable and clean code?? In my opinion absolutely not. There are many issues in the above code from readability perspective. 1. Mixing DB interaction code with business logic 2. Throwing IllegalArgumentException, RuntimeException etc from business methods instead of Business specific exceptions 3. Most importantly, the code is mixed with different levels of abstractions. Let me explain what I mean by different levels of abstractions. Firstly, from business perspective fund transfer means validating source/target accounts, checking for sufficient balance, checking for overdraft limit, checking for duplicate transaction and making the fund transfer. From technical point of view there are various tasks like fetching Account details from DB, performing all the business related checks, throwing Exceptions if there are any violations, properly closing the resources etc. But in the above code everything is mixed together. While reading the code you start looking at JDBC code and your brain is working in Technical mode and after getting Account object from ResultSet you are checking for null and throwing Exception if it is null which is Business requirement. So immediately you need to switch your mind to Business mode and think "OK, if the account is invalid we want to abort the operation immediately". Though you managed to switch between Technical and Business modes, what about making an enhancement to one perticular subtask like "Fund transfer is considred duplicate only if it matches with the last transaction that happened with in an hour only". To make that enhancement you have to go through the entire method because you haven't modularised your sub-tasks and there is no separation of concerns. Let's rewrite the above code as follows: class FundTransferTxn { private Account sourceAccount; private Account targetAccount; private BigDecimal amount; private boolean allowDuplicateTxn; //setters & getters } public class CleanMoneyTransferService { public void transferFunds(FundTransferTxn txn) { Account sourceAccount = validateAndGetAccount(txn.getSourceAccount().getAcno()); Account targetAccount = validateAndGetAccount(txn.getTargetAccount().getAcno()); checkForOverdraft(sourceAccount, txn.getAmount()); checkForDuplicateTransaction(txn); makeTransfer(sourceAccount, targetAccount, txn.getAmount()); } private Account validateAndGetAccount(String acno){ Account account = AccountDAO.getAccount(acno); if(account == null){ throw new InvalidAccountException("Invalid ACNO :"+acno); } } private void checkForOverdraft(Account account, BigDecimal amount){ if(!account.isOverdraftAllowed()){ if((account.getBalance() - amount) < 0){ throw new InsufficientBalanceException("Insufficient Balance"); } } else{ if(((account.getBalance()+account.getOverdraftLimit()) - amount) < 0){ throw new ExceedingOverdraftLimitException("Insufficient Balance, Exceeding Overdraft Limit"); } } } private void checkForDuplicateTransaction(FundTransferTxn txn){ AccountTransaction lastTxn = TransactionDAO.getLastTransaction(txn.getSourceAccount().getAcno()); if(lastTxn != null){ if(lastTxn.getTargetAcno().equals(txn.getTargetAccount().getAcno()) && lastTxn.getAmount() == txn.getAmount() && !txn.isAllowDuplicateTxn()){ throw new DuplicateTransactionException("Duplicate transaction exception"); } } } private void makeTransfer(Account source, Account target, BigDecimal amount){ sourceAccount.debit(amount); targetAccount.credit(amount); TransactionService.saveTransaction(source, target, amount); } } The above improved method do exactly what the initial verson is doing but now it looks lot better than earlier version. We have divided the entire task into sub-tasks and implemented each sub-task in a separate method. We have delegated DB interactions to DAOs We are throwing Business specific Exceptions instead of Java language Exceptions All in all we have separated the levels of abstractions. At the first level we have highest level of abstraction in transferFunds(FundTransferTxn txn) method. By looking at this method we can understand what we are doing as part of Fund Transfer operation without worrying much about implementation or technical details. At the second level we have business logic implementation methods checkForOverdraft, checkForDuplicateTransaction etc which performs business logic, again without worrying much about technical details. At the lowest level we have technical implementation details in AccountDAO and TransactionDAO which contains DB interaction logic. So the reader(future developer/maintainer) of your code can easily understand what you are doing at the high level and can dig into method which he is interested in. As I said earlier, if we have to make the change to consider the transaction as a duplicate transaction only if it happened with in an hour, we can easily understand that checkForDuplicateTransaction() is the one we have to look into and make change. Happy coding!!

I first came across Reactive Extensions about 4 years ago on Matthew Podwysocki’s blog but then haven’t heard much about it until I saw Matthew give a talk at Code Mesh a few weeks ago. It seems to have grown in popularity recently and I noticed that’s there’s now a Java version called RxJavawritten by Netflix. I thought I’d give it a try by changing some code I wrote while exploring cypher’s MERGE function to expose an Observable instead of Futures. To recap, we have 50 threads and we do 100 iterations where we create random (user, event) pairs. We create a maximum of 10 users and 50 events and the goal is to concurrently send requests for the same pairs. In the example of my other post I was throwing away the result of each query whereas here I returned the result back so I had something to subscribe to. The outline of the code looks like this: public class MergeTimeRx { public static void main( final String[] args ) throws InterruptedException, IOException { String pathToDb = "/tmp/foo"; FileUtils.deleteRecursively( new File( pathToDb ) ); GraphDatabaseService db = new GraphDatabaseFactory().newEmbeddedDatabase( pathToDb ); final ExecutionEngine engine = new ExecutionEngine( db ); int numberOfThreads = 50; int numberOfUsers = 10; int numberOfEvents = 50; int iterations = 100; Observable events = processEvents( engine, numberOfUsers, numberOfEvents, numberOfThreads, iterations ); events.subscribe( new Action1() { @Override public void call( ExecutionResult result ) { for ( Map row : result ) { } } } ); .... } } The nice thing about using RxJava is that there’s no mention of how we got our collection of ExecutionResults, it’s not important. We just have a stream of them and by calling the subscribe function on the Observablewe’ll be informed whenever another one is made available. Most of the examples I found show how to generate events from a single thread but I wanted to use a thread pool so that I could fire off lots of requests at the same time. The processEvents method ended up looking like this: private static Observable processEvents( final ExecutionEngine engine, final int numberOfUsers, final int numberOfEvents, final int numberOfThreads, final int iterations ) { final Random random = new Random(); final List userIds = generateIds( numberOfUsers ); final List eventIds = generateIds( numberOfEvents ); return Observable.create( new Observable.OnSubscribeFunc() { @Override public Subscription onSubscribe( final Observer observer ) { final ExecutorService executor = Executors.newFixedThreadPool( numberOfThreads ); List> jobs = new ArrayList<>(); for ( int i = 0; i < iterations; i++ ) { Future job = executor.submit( new Callable() { @Override public ExecutionResult call() { Integer userId = userIds.get( random.nextInt( numberOfUsers ) ); Integer eventId = eventIds.get( random.nextInt( numberOfEvents ) ); return engine.execute( "MERGE (u:User {id: {userId})\n" + "MERGE (e:Event {id: {eventId})\n" + "MERGE (u)-[:HAS_EVENT]->(e)\n" + "RETURN u, e", MapUtil.map( "userId", userId, "eventId", eventId ) ); } } ); jobs.add( job ); } for ( Future future : jobs ) { try { observer.onNext( future.get() ); } catch ( InterruptedException | ExecutionException ignored ) { } } observer.onCompleted(); executor.shutdown(); return Subscriptions.empty(); } } ); } I’m not sure if that’s the correct way of using Observables so please let me know in the comments if I’ve got it wrong. I wasn’t sure what the proper way of handling errors was. I initially had a call to observer#onError in the catch block but that means that no further events are produced which wasn’t what I wanted. The code is available as a gist if you want to play around with it. I added the following dependency to get the RxJava library: com.netflix.rxjava rxjava-core 0.15.1

I’ll review Google’s most important Big Data publications and discuss where they are (as far as they’ve disclosed).

This article is an introduction to using the specification pattern in Java. We also will see how we can combine classic specifications with JPA Criteria queries to retrieve objects from a relational database. Within this post we will use the following Poll class as an example entity for creating specifications. It represents a poll that has a start and end date. In the time between those two dates users can vote among different choices. A poll can also be locked by an administrator before the end date has been reached. In this case, a lock date will be set. @Entity public class Poll { @Id @GeneratedValue private long id; private DateTime startDate; private DateTime endDate; private DateTime lockDate; @OneToMany(cascade = CascadeType.ALL) private List votes = new ArrayList<>(); } For better readability I skipped getters, setters, JPA annotations for mapping Joda DateTime instances and fields that aren't needed in this example (like the question being asked in the poll). Now assume we have two constraints we want to implement: A poll is currently running if it is not locked and if startDate < now < endDate A poll is popular if it contains more than 100 votes and is not locked We could start by adding appropriate methods to Poll like: poll.isCurrentlyRunning(). Alternatively we could use a service method like pollService.isCurrentlyRunning(poll). However, we also want to be able to query the database to get all currently running polls. So we might add a DAO or repository method like pollRepository.findAllCurrentlyRunningPolls(). If we follow this way we implement the isCurrentlyRunning constraint two times in two different locations. Things become worse if we want to combine constraints. What if we want to query the database for a list of all popular polls that are currently running? This is where the specification pattern come in handy. When using the specification pattern we move business rules into extra classes called specifications. To get started with specifications we create a simple interface and an abstract class: public interface Specification { boolean isSatisfiedBy(T t); Predicate toPredicate(Root root, CriteriaBuilder cb); Class getType(); } abstract public class AbstractSpecification implements Specification { @Override public boolean isSatisfiedBy(T t) { throw new NotImplementedException(); } @Override public Predicate toPredicate(Root poll, CriteriaBuilder cb) { throw new NotImplementedException(); } @Override public Class getType() { ParameterizedType type = (ParameterizedType) this.getClass().getGenericSuperclass(); return (Class) type.getActualTypeArguments()[0]; } } Please ignore the AbstractSpecification class with the mysterious getType() method for a moment (we come back to it later). The central part of a specification is the isSatisfiedBy() method, which is used to check if an object satisfies the specification. toPredicate() is an additional method we use in this example to return the constraint as javax.persistence.criteria.Predicate instance which can be used to query a database. For each constraint we create a new specification class that extends AbstractSpecification and implements isSatisfiedBy() and toPredicate(). The specification implementation to check if a poll is currently running looks like this: public class IsCurrentlyRunning extends AbstractSpecification { @Override public boolean isSatisfiedBy(Poll poll) { return poll.getStartDate().isBeforeNow() && poll.getEndDate().isAfterNow() && poll.getLockDate() == null; } @Override public Predicate toPredicate(Root poll, CriteriaBuilder cb) { DateTime now = new DateTime(); return cb.and( cb.lessThan(poll.get(Poll_.startDate), now), cb.greaterThan(poll.get(Poll_.endDate), now), cb.isNull(poll.get(Poll_.lockDate)) ); } } Within isSatisfiedBy() we check if the passed object matches the constraint. In toPredicate() we construct a Predicate using JPA's CriteriaBuilder. We will use the resulting Predicate instance later to build a CriteriaQuery for querying the database. The specification for checking if a poll is popular looks similar: public class IsPopular extends AbstractSpecification { @Override public boolean isSatisfiedBy(Poll poll) { return poll.getLockDate() == null && poll.getVotes().size() > 100; } @Override public Predicate toPredicate(Root poll, CriteriaBuilder cb) { return cb.and( cb.isNull(poll.get(Poll_.lockDate)), cb.greaterThan(cb.size(poll.get(Poll_.votes)), 5) ); } } If we now want to test if a Poll instance matches one of these constraints we can use our newly created specifications: boolean isPopular = new IsPopular().isSatisfiedBy(poll); boolean isCurrentlyRunning = new IsCurrentlyRunning().isSatisfiedBy(poll); For querying the database we need to extend our DAO / repository to support specifications. This can look like the following: public class PollRepository { private EntityManager entityManager = ... public List findAllBySpecification(Specification specification) { CriteriaBuilder criteriaBuilder = entityManager.getCriteriaBuilder(); // use specification.getType() to create a Root instance CriteriaQuery criteriaQuery = criteriaBuilder.createQuery(specification.getType()); Root root = criteriaQuery.from(specification.getType()); // get predicate from specification Predicate predicate = specification.toPredicate(root, criteriaBuilder); // set predicate and execute query criteriaQuery.where(predicate); return entityManager.createQuery(criteriaQuery).getResultList(); } } Here we finally use the getType() method implemented in AbstractSpecification to createCriteriaQuery and Root instances. getType() returns the generic type of theAbstractSpecification instance defined by the subclass. For IsPopular andIsCurrentlyRunning it returns the Poll class. Without getType() we would have to create theCriteriaQuery and Root instances inside toPredicate() of every specification we create. So it is just a small helper to reduce boiler plate code inside specifications. Feel free to replace this with your own implementation if you come up with better approaches. Now we can use our repository to query the database for polls that match a certain specification: List popularPolls = pollRepository.findAllBySpecification(new IsPopular()); List currentlyRunningPolls = pollRepository.findAllBySpecification(new IsCurrentlyRunning()); At this point the specifications are the only components that contain the constraint definitions. We can use it to query the database or to check if an object fulfills the required rules. However one question remains: How do we combine two or more constraints? For example we would like to query the database for all popular polls that are still running. The answer to this is a variation of the composite design pattern called composite specifications. Using a composite specification we can combine specifications in different ways. To query the database for all running and popular pools we need to combine the isCurrentlyRunning with theisPopular specification using the logical and operation. Let's create another specification for this. We name itAndSpecification: public class AndSpecification extends AbstractSpecification { private Specification first; private Specification second; public AndSpecification(Specification first, Specification second) { this.first = first; this.second = second; } @Override public boolean isSatisfiedBy(T t) { return first.isSatisfiedBy(t) && second.isSatisfiedBy(t); } @Override public Predicate toPredicate(Root root, CriteriaBuilder cb) { return cb.and( first.toPredicate(root, cb), second.toPredicate(root, cb) ); } @Override public Class getType() { return first.getType(); } } An AndSpecification is created out of two other specifications. In isSatisfiedBy() and toPredicate()we return the result of both specifications combined by a logical and operation. We can use our new specification like this: Specification popularAndRunning = new AndSpecification<>(new IsPopular(), new IsCurrentlyRunning()); List polls = myRepository.findAllBySpecification(popularAndRunning); To improve readability we can add an and() method to the Specification interface: public interface Specification { Specification and(Specification other); // other methods } and implement it within our abstract implementation: abstract public class AbstractSpecification implements Specification { @Override public Specification and(Specification other) { return new AndSpecification<>(this, other); } // other methods } Now we can chain multiple specification by using the and() method: Specification popularAndRunning = new IsPopular().and(new IsCurrentlyRunning()); boolean isPopularAndRunning = popularAndRunning.isSatisfiedBy(poll); List polls = myRepository.findAllBySpecification(popularAndRunning); When needed we can easily extend this further with other composite specifications (for exampleOrSpecification or NotSpecification). Conclusion When using the specification pattern we move business rules in separate specification classes. These specification classes can be easily combined by using composite specifications. In general, specification improve reusability and maintainability. Additionally specifications can easily be unit tested. For more detailed information about the specification pattern I recommend this article by Eric Evans and Martin Fowler. You can find the source of this example project on GitHub.

In the directory structure of JBossAS-5.x ,there’s a directory named ‘client‘. The ‘client‘ directory consists of all the client libraries required to make communication with the server. Consider a scenario where you have an EJB application deployed on your JBoss AS Server. You want to invoke an EJB method from a Client POJO (e.g., from an Eclipse Project). In that case, you are required to include all the client libraries in your project Class-path in order to invoke EJB methods on the server. There’s a special eye-catching library file in client directory named jbossall-client.jar. This file consists of all the client classes of JBoss. Up to JBossAS-4.x, the jbossall-client.jar file was used to contain all the classes of client. Such as, However, in JBossAS-5.x and above, this jar file does not contain any actual classes, it only contains the META-INF/MANIFEST.MF, which lists the needed dependencies against the other libraries from server’s client directory. The anatomy of this jar file is shown below, The META-INF/MANIFEST.MF file contains the list of dependent jar files in Class-path as following, If you want to use this jar, you must place all the listed jar files in the same directory as jbossall-client.jar in your project classpath. The description of the readme.txt file included in jar says that,, “ This jar file contains a classpath reference to various client jar files used by jboss client applications. Each of the jar files in the following list must available in the same directory as the jbossall-client.jar, Otherwise they will not be found by the classloader. “ This means when the jbosall-client.jar file would be loaded by the JVM, this file will look up the dependent classes from the listed jars in MANIFEST.MF. If the JVM would not find the dependent classes in the classpath, it will throw the ClassNotFoundException. Well, there’r some hacks to get rid of such issues… Let the Hacking begin!!! : Hack-1: If you’re a JBoss Geek who is so sure of the minimum number of client libraries you are required to make the remote call, you can directly include those client libraries in your classpath, and exclude the jbossall-client.jar from the classpath. By this, the JVM will not cry for those dependent class files. Oh yes, you’ve got those jar file names on your fingertips. Geeky! Hack-2: Repackage your jbossall-client.jar file. Repackaging can be done in two ways. First is, open the jbossall-client.jar archive with any free archive tool such as 7-zip or WinRar, and remove the name of jar files you don’t want from the Class-path defined at META-INF/MANIFEST.MF. And just ‘Save’ the file and close it. It will automatically repackage the jar file by injecting the modified MANIFEST.MF file. If you don't want to use such tools, then the other way of repackaging this jar is using the jar command from Java. Use following steps to repackage this jar, Copy jbossall-client.jar file from JBOSS_HOME/client and place it to some convenient directory. Open command prompt or terminal and navigate to the directory where you have copied this file. Make sure your JAVA_HOME environment variable is set with bin directory in PATH, else you may have to refer to absolute path of jar command while using it. Fire command: jar -xvf jbossall-client.jar .This will extract the content of this archive. Now, go to extracted content and open META-INF/MANIFEST.MF, and remove the name of jar files you don't want to use. And save the file. Go to your command prompt or terminal again and fire the command: jar -cvfM jbossall-client.jar META-INF readme.txt . This will repackage the jar file with the same name as jbossall-client.jar. Now, once this jar file is repackaged, you can use this jar file in your classpath along with the dependencies manifested. But why.. why... why....??? : Well, you must've been asking yourself throughout the article why the heck you want to use jbossall-client.jar if you are cool enough to use other dependent jars without jbossall-client.jar at client side??? The answer to this question is: if you are not aware of the list of jar files to be included at client side, your screen will be flooded with ClassNotFoundException. And iteratively you will have to look up for those client jar files containing those class files. Hence, jbossall-client.jar provides a pretty good listing of these dependencies. Moreover, listing twenty to thirty jar files in the classpath for a client seems to be too much work on the part of your customers. Hence, I recommend the usage of this file. Sometimes people call this jbossall-client.jar file a big mess, but I find it very convenient to use it in order to impose the dependencies in a more structured way! Source: http://www.jboss.org.in

Finally the wait is over, Spring 4 is here and one of the best features is that now has a fully support for Groovy. The Spring team did a fantastic job bringing the same concept from Grails into the Spring Framework. In the past if you needed to incorporate Spring in your Groovy Applications or Scripts , normally you did something like this: @Grab('org.springframework:spring-context:4.0.0.RELEASE') import org.springframework.context.support.GenericApplicationContext import org.springframework.context.annotation.ClassPathBeanDefinitionScanner import org.springframework.beans.factory.annotation.Autowired import org.springframework.stereotype.Component @Component class Login { def authorize(User user){ if(user.credentials.username == "guest" && user.credentials.password == "guest"){ "${user.greetings} ${user.credentials.username}" }else "You are not ${user.greetings}" } } @Component class Credentials { String username = "guest" String password = "guest" } @Component class User{ @Autowired Credentials credentials String greetings = "Welcome" } def ctx = new GenericApplicationContext() new ClassPathBeanDefinitionScanner(ctx).scan('') // scan root package for components ctx.refresh() def login = ctx.getBean("login") def user = ctx.getBean("user") println login.authorize(user) One of the benefits that I see is that Groovy removes all the unnecessary boiler plate from Java. Then, if you wanted to add more flavor to your Groovy Apps using the Grail's BeanBuilder , you needed to do something like this: @Grab(group='org.slf4j', module='slf4j-simple', version='1.7.5') @Grab(group='org.grails', module='grails-web', version='2.3.4') import grails.spring.BeanBuilder import groovy.transform.ToString class Login { def authorize(User user){ if(user.credentials.username == "John" && user.credentials.password == "Doe"){ "${user.greetings} ${user.credentials.username}" }else "You are not ${user.greetings}" } } @ToString(includeNames=true) class Credentials{ String username String password } @ToString(includeNames=true) class User{ Credentials credentials String greetings } def bb = new BeanBuilder() bb.beans { login(Login) user(User){ credentials = new Credentials(username:"John", password:"Doe") greetings = 'Welcome!!' } } def ctx = bb.createApplicationContext() def u = ctx.getBean("user") println u def l = ctx.getBean("login") println l.authorize(u) And now with the new Spring 4 you can add all the Groovy flavor to your apps (without all the Grails Dependencies) and using the new GroovyBeanDefinitionReader from Spring 4: @Grab('org.springframework:spring-context:4.0.0.RELEASE') import org.springframework.context.support.GenericApplicationContext import org.springframework.beans.factory.groovy.GroovyBeanDefinitionReader import groovy.transform.ToString class Login { def authorize(User user){ if(user.credentials.username == "John" && user.credentials.password == "Doe"){ "${user.greetings} ${user.credentials.username}" }else "You are not ${user.greetings}" } } @ToString(includeNames=true) class Credentials{ String username String password } @ToString(includeNames=true) class User{ Credentials credentials String greetings } def ctx = new GenericApplicationContext() def reader = new GroovyBeanDefinitionReader(ctx) reader.beans { login(Login) user(User){ credentials = new Credentials(username:"John", password:"Doe") greetings = 'Welcome!!' } } ctx.refresh() def u = ctx.getBean("user") println u def l = ctx.getBean("login") println l.authorize(u) And that's not all, also SpringBoot gives you even more power using Groovy: //File: app.groovy import org.springframework.boot.* import org.springframework.boot.autoconfigure.* import org.springframework.stereotype.* import org.springframework.web.bind.annotation.* import org.springframework.beans.factory.annotation.* @Component class Login { def authorize(User user){ if(user.credentials.username == "John" && user.credentials.password == "Doe"){ "${user.greetings} ${user.credentials.username}" }else "You are not ${user.greetings}" } } @Component class Credentials{ String username String password } @Component class User{ Credentials credentials = new Credentials(username:"John", password:"Doe") String greetings = "Welcome!!" } @Controller @EnableAutoConfiguration class SampleController { @Autowired def login @Autowired def user @RequestMapping("/") @ResponseBody String home() { return login.authorize(user) } } To run the above code, install springboot and then just execute: $ spring run app.groovy and then go to your browser and open http://localhost:8080 and you see: Welcome!! John Congratulations to the Spring Team!! Keep grooving!! References: [1] https://spring.io/blog/2013/12/12/announcing-spring-framework-4-0-ga-release [2] http://groovy.codehaus.org/Using+Spring+Factories+with+Groovy [3] http://grails.org/doc/latest/guide/spring.html [4] http://projects.spring.io/spring-boot/ [5] https://spring.io/guides

a common and difficult problem acquiring data is extracting tables from a pdf. previously, i described how to extract the text from a pdf with pdf.js , a pdf rendering library made by mozilla labs. the rendering process requires an html canvas object, and then draws each object (character, line, rectangle, etc) on it. the easiest way to get a list of these is to to intercept all the calls pdf.js makes to drawing functions on the canvas object. (see “ self modifying javascripts ” for a similar technique). the “set” method below adds a wrapper closure to each function, which logs the call. function replace(ctx, key) { var val = ctx[key]; if (typeof(val) == "function") { ctx[key] = function() { var args = array.prototype.slice.call(arguments); console.log("called " + key + "(" + args.join(",") + ")"); return val.apply(ctx, args); } } } for (var k in context) { replace(context, k); } var rendercontext = { canvascontext: context, viewport: viewport }; page.render(rendercontext); this lets us see a series of calls: called transform(1,0,0,1,150.42,539.67) called translate(0,0) called scale(1,-1) called scale(0.752625,0.752625) called measuretext(c) called save() called scale(0.9701818181818181,1) called filltext(c,0,0) called restore() called restore() called save() called transform(1,0,0,1,150.42,539.6 we can easily retrieve the text by noting the first argument to each “filltext” call: "congregations ranked by growth and decline in membership and worship attendance, 2006 to 2011philadelphia presbytery - table 16net membership changenet worship changepercent changepercent changeworship 2006worship 2011membership 2006membership 2011abington, abington- 143(74)-13.18%(57)0(15)0.00%(22)numberrank3003001,085942anchor, wrightstown0(23)0.00%(27)-12(25)-21.43%(52)numberrank56449797arch street, philadelphia-117(71)-68.42%(117)27(5)90.00% (2)numberrank305717154aston, aston3(21)3.53%(22)-5(19)-9.43% (31)numberrank53488588beaconno reportboth yearsno reportboth yearsnumberrankbensalem, bensalem-23(39)-13.94%(62)-28(36)-28.57% (64)numberrank9870165142berean, philadelphia106(4)44.92%(4)no reportboth yearsnumberrank00236342bethany collegiate, havertown- 188(76)-42.44%(110)43(3)21.29%(7)numberrank202245443255bethel, philadelphia-13(33)-13.68%(60)-27(35)-35.06% (71)numberrank77509582bethesda, philadelphia9(18)5.56%(18)no reportboth yearsnumberrank1150162171beverly hills, upper darby-3(26)-3.03% (32)-11(24)-20.00%(48)numberrank55449996bridesburg, philadelphia0(23)0.00%(27)no reportboth yearsnumberrank004444bristol, bristolno reportboth yearsno reportboth yearsnumberrankpage 1 of 10report prepared by research services, presbyterian church (u.s.a.)1- 800-728-7228, ext #204006-oct-12" notable, this doesn’t track line endings, and not all the characters are recorded in the expected order (the first line is rendered after the second). the calls to transform, translate, and scale control where text is placed. the filltext method also takes an (x, y) parameter set that moves the individual letters between words. the exact position is a combination of successive operations, which are modeled as a stack of matrix operations. thankfully, pdf.js tracks the output of these operations as it renders, so we don’t have to recalculate it. thus, we can make a method that records the letters and their real positions. this method takes the internal context object, the type of state transition, and the arguments to the transition. this method is then called from the ‘record’ function listed above. var chars = []; var cur = {}; function record(ctx, state, args) { if (state === 'filltext') { var c = args[0]; cur.c = c; cur.x = ctx._transformmatrix[4] + args[1]; cur.y = ctx._transformmatrix[5] + args[2]; chars[chars.length] = cur; cur = {}; } } these results can be sorted by position (x and y). the sort method arranges letters by position – if they are shifted up or down a small amount, they are considered to be on one line. chars.sort( function(a, b) { var dx = b.x - a.x; var dy = b.y - a.y; if (math.abs(dy) < 0.5) { return dx * -1; } else { return dy * -1; } } ); this presents several difficulties: this doesn’t detect right-to-left text, and it’s becoming clear that we’re going to have a hard time knowing when you’re in a table and when we aren’t. to do this, we define a function which can transform the array of letters and positions into a csv style output. this tracks from letter to letter – if it sees a “large” change in y, it makes a new line. if it sees a “large” change in x, it treats it as a new column. the real challenge is defining “large” which for my test pdf were around 15 and 20, for dx and dy. function gettext(marks, ex, ey, v) { var x = marks[0].x; var y = marks[0].y; var txt = ''; for (var i = 0; i < marks.length; i++) { var c = marks[i]; var dx = c.x - x; var dy = c.y - y; if (math.abs(dy) > ey) { txt += "\"\n\""; if (marks[i+1]) { // line feed - start from position of next line x = marks[i+1].x; } } if (math.abs(dx) > ex) { txt += "\",\""; } if (v) { console.log(dx + ", " + dy); } txt += c.c; x = c.x; y = c.y; } return txt; } this algorithm doesn’t handle newlines in rows, and oddly, the columns don’t come out in the right order, but they appear to be consistently out of order. line with large spaces (e.g. an em-dash) are detected as having multiple columns, but this can be cleaned up later – here is some sample output. you can see an example below, and the final source is available on github . congregations ranked by growth and decline in m","embership and w","orship attendance, 2006 to 2011" "","philadelphia presbytery"," - table 16" "","net ","membership ","change" "","net worship ","change","percent ","change","percent ","change","worship"," 2006","worship"," 2011","membership"," 2006","membership"," 2011" "","abington, abington","-143","(74)","-13.18%(57)","0","(15)","0.00%(22)","number","rank","300","300","1,085","942" "","anchor, wrightstown","0","(23)","0.00%(27)","-12","(25)","-21.43%(52)","number","rank","56","44","97","97" "","arch street, philadelphia","-117","(71)","-68.42%","(117)","27(5)","90.00%(2)","number","rank","30","57","171","54" "","aston, aston","3","(21)","3.53%(22)","-5","(19)","-9.43%(31)","number","rank","53","48","85","88" "","beacon","no report","both years","no report","both years","number","rank" "","bensalem, bensalem","-23","(39)","-13.94%(62)","-28","(36)","-28.57%(64)","number","rank","98","70","165","142" "","berean, philadelphia","106(4)","44.92%(4)","no report","both years","number","rank","0","0","236","342" "","bethany collegiate, havertown","-188","(76)","-42.44%","(110)","43(3)","21.29%(7)","number","rank","202","245","443","255" "","bethel, philadelphia","-13","(33)","-13.68%(60)","-27","(35)","-35.06%(71)","number","rank","77","50","95","82" "","bethesda, philadelphia","9","(18)","5.56%(18)","no report","both years","number","rank","115","0","162","171" "","beverly hills, upper darby","-3","(26)","-3.03%(32)","-11","(24)","-20.00%(48)","number","rank","55","44","99","96" "","bridesburg, philadelphia","0","(23)","0.00%(27)","no report","both years","number","rank","0","0","44","44" "","bristol, bristol","no report","both years","no report","both years","number","rank" "","page 1 of 10","report prepared by research services, presbyterian church (u.s.a.)","1-800-728-7228, ext #2040","06-oct-12"

this post is not about variables in my application code (which i debug). it is about using variables in eclipse for building projects. eclipse variables allow me to make my projects ‘position independent’ whenever i cannot use a path relative to my projects or workspace. eclipse variables which variables are used where in eclipse might be sometimes not very clear. depending in which context variables are used, not everything might be available. this link for example gives a list of variables which can be used to invoke an external tool. build variables eclipse comes with many built-in variables, especially for the build system. if i want to see what variables are already defined, i can show them in the project properties, under c/c++ build > build variables with enabled option ‘show system variables’: system build variables with the ‘add…’ button i can define and add my own variables, available for that project: define a new build variable if above operation is done on a project, then the setting is for the project only. if i want to add a variable for the workspace, i can do this using the menu window > preferences : workspace build variables global system variables eclipse automatically includes the system (e.g. windows) environment variables. many dialogs have the ‘variables…’ button where i can use my variables, including the variables defined on system level: system variables system variables: one way or the other so if i want to have a variable for every workspace, one way is to define it at the system level. however, this is not a good way as this clutter the variables for every application. batch file a solution to this to create my custom batch file where i define my variables, and at the end of this batch file i launch eclipse. that way the extra variables are only for this eclipse session. cwide-env file another very nice way codewarrior eclipse offers is using the cwide-env file located in the eclipse sub-folder of the installation: cwide-env file i can define variables here, or extend existing ones: -add : add string to the variable at the end -prepend : add string to the variable at the beginning that way i can easily manipulate existing system variables or create new ones which then are used by eclipse. summary variables in eclipse help me to define paths to source files and folders outside of a project or workspace. with variables i avoid using absolute paths which would make porting projects from one machine to another difficult. i can define variables for projects, for the workspace or use system variables. with codewarrior i have a cwide-env file which is used to extend the system variables. happy variabling

Since Entity Framework now has a very nice support of performing all its actions asynchronously, the methods in the repositories in our projects will turn into asynchronous methods soon and so will be the code depending on it. Tom Fitzmacken did a nice job by putting together a tutorial on unit testing Web API 2 Controllers on official ASP.NET site. The tutorial discusses on testing synchronous action methods. The same techniques can be applied to test asynchronous action actions as well. In this post, we will see how easy it is to test asynchronous Web API action methods using MS Test. I created a simple repository interface with just one method in it. The implementation class uses Entity Framework to get a list of contacts from the database. public interface IRepository { Task> GetAllContactsAsync(); } public class Repository : IRepository { ContactsContext context = new ContactsContext(); public async Task> GetAllContactsAsync() { return await context.Contacts.ToArrayAsync(); } } Following is the ASP.NET Web API controller that uses the above repository: public class ContactsController : ApiController { IRepository repository; public ContactsController() : this(new Repository()) { } public ContactsController(IRepository _repository) { repository = _repository; } [Route("api/contacts/plain")] public async Task> GetContactsListAsync() { IEnumerable contacts; try { contacts = await repository.GetAllContactsAsync(); } catch (Exception) { throw; } return contacts; } [Route("api/contacts/httpresult")] public async Task GetContactsHttpActionResultAsync() { IEnumerable contacts; try { contacts = await repository.GetAllContactsAsync(); } catch (Exception ex) { return InternalServerError(ex); } return Ok(contacts); } } As we see, the controller has two action methods performing the same task, but the way they return the results is different. Since both of the action methods respond to HTTP GET method, I used attribute routing to distinguish them. I used poor man’s dependency injection to instantiate the repository; it can be easily replaced using an IoC container. Before writing unit tests for the above action methods, we need to create a mock repository. public class MockRepository:IRepository { List contacts; public bool FailGet { get; set; } public MockRepository() { contacts = new List() { new Contact(){Id=1, Title="Title1", PhoneNumber="1992637281", CustomerId=1}, new Contact(){Id=2, Title="Title2", PhoneNumber="9172735171", SupplierId=2}, new Contact(){Id=3, Title="Title3", PhoneNumber="8361910353", CustomerId=2}, new Contact(){Id=4, Title="Title4", PhoneNumber="7801274518", SupplierId=3} }; } public async Task> GetAllContactsAsync() { if (FailGet) { throw new InvalidOperationException(); } await Task.Delay(1000); return contacts; } } The property FailGet in the above class is used to force the mock to throw an exception. This is done just to cover more test cases. In the test class, we need a TestInitialize method to arrange the objects needed for unit testing. [TestClass] public class ContactsControllerTests { MockRepository repository; ContactsController contactsApi; [TestInitialize] public void InitializeForTests() { repository = new MockRepository(); contactsApi = new ContactsController(repository); } } Let us test the GetContactsListAsync method first. Testing this method seems to be straight forward, as it returns either a plain generic list or throws an exception. But the test method can’t just return void like other tests, as the method is asynchronous. To test an asynchronous method, the test method should also be made asynchronous and return a Task. Following test checks if the controller action returns a collection of length 4: [TestMethod] public async Task GetContacts_Should_Return_List_Of_Contacts() { var contacts = await contactsApi.GetContactsListAsync(); Assert.AreEqual(contacts.Count(), 4); } If the repository encounters an exception, the exception is re-thrown from the GetContactsListAsync method as well. This case can be checked using the ExpectedException attribute. [TestMethod] [ExpectedException(typeof(InvalidOperationException))] public async Task GetContacts_Should_Throw_Exception() { repository.FailGet = true; var contacts = await contactsApi.GetContactsListAsync(); } Now let’s test the GetContactsHttpActionResultAsync method. Though this method does the same thing as the previous method, it doesn’t return the plain .NET objects. To test this method, we need to extract the result from the IHttpActionResult object obtained from the action method. Following test checks if the action result contains a collection when the repository is able to fetch results. Return type of Ok() method used above is OkNegotiatedContentResult. IHttpActionresult has to be converted to this type to check for the result obtained: [TestMethod] public async Task GetContactsHttpActionResult_Should_Return_HttpResult_With_Contacts() { var contactsResult = await contactsApi.GetContactsHttpActionResultAsync() as OkNegotiatedContentResult>; Assert.AreEqual(contactsResult.Content.Count(), 4); } Similarly, in case of error, we are calling InternalServerError() method to return the exception for us. We need to convert the result to ExceptionResult type to be able to check the type of exception thrown. It is shown below: [TestMethod] public async Task GetContactsHttpActionResult_Should_Return_HttpResult_With_Exception() { repository.FailGet = true; var contactsResult = await contactsApi.GetContactsHttpActionResultAsync() as ExceptionResult; Assert.IsInstanceOfType(contactsResult.Exception,typeof(InvalidOperationException)); } Happy coding!

the upcoming version of jboss application server will no longer use tomcat as the integrated webserver, but instead, will replace it with undertow . the architecture of undertow is based on handlers that can be added dynamically via a builder api to the server. this approach is similar to the way of constructing a webserver in node.js . it allows developers to embed the undertow webserver easily into their applications. as the addition of features is done via the builder api, one can only add the features that are really required in one’s application. beyond that undertow supports websockets and the servlet api in version 3.1. it can be run as a blocking or non-blocking server and it is reported that first benchmarks have proven that undertow is the fastest webserver written in java. as all of this sounds very promising, so let’s try to set up a simple websocket server. as usual we start by creating a simple java project and add the undertow maven dependency: io.undertow undertow-core 1.0.0.beta20 with undertow’s builder api our buildandstartserver() method looks like this: public void buildandstartserver(int port, string host) { server = undertow.builder() .addlistener(port, host) .sethandler(getwebsockethandler()) .build(); server.start(); } we just add a listener that specifies the port and host to listen for incoming connections and afterwards add a websocket handler. as the websocket handler code is a little bit more comprehensive, i have put it into its own method: private pathhandler getwebsockethandler() { return path().addpath("/websocket", websocket(new websocketconnectioncallback() { @override public void onconnect(websockethttpexchange exchange, websocketchannel channel) { channel.getreceivesetter().set(new abstractreceivelistener() { @override protected void onfulltextmessage(websocketchannel channel, bufferedtextmessage message) { string data = message.getdata(); lastreceivedmessage = data; logger.info("received data: "+data); websockets.sendtext(data, channel, null); } }); channel.resumereceives(); } })) .addpath("/", resource(new classpathresourcemanager(websocketserver.class.getclassloader(), websocketserver.class.getpackage())) .addwelcomefiles("index.html")); } let’s go line by line through this code snippet. first of all, we add a new path: /websocket. the second argument of the addpath() methods lets us specify what kind of protocol we want to use for this path. in our case we create a new websocket. the anonymous implementation has a onconnect() method in which we set an implementation of abstractreceivelistener. here we have a convenient method onfulltextmessage() that is called when a client has sent us a text message. a call of getdata() fetches the actual message we have received. in this simple example we just echo this string back to client to validate that the roundtrip from the client to server and back works. to perform some simple manual tests we also add a second resource under the path / which serves some static html and javascript files. the directory that contains these files is given as an instance of classpathresourcemanager. the call of addwelcomefiles() tells undertow which file to server when the client asks for the path /. the index.html looks like this: our javascript code is swapped out to the websocket.js file. we use jquery and the jquery-plugin gracefulwebsocket to ease the client side development: var ws = $.gracefulwebsocket("ws://127.0.0.1:8080/websocket"); ws.onmessage = function(event) { var messagefromserver = event.data; $('#output').append('received: '+messagefromserver+''); } function send(message) { ws.send(message); } after having created a websocket object by calling $.gracefulwebsocket() we can register a callback function for incoming messages. in this method we only append the message string to the dom of the page. the send() method is just a call to gracefulwebsocket’s send() method. when we now start our application and open the url http://127.0.0.1:8080/ in our webbrowser we see the following page: entering some string and hitting the “send web socket data” button sends the message to the server, which in response echos it back to the client. now that we know that everything works as expected, we want to protect our code against regression with a junit test case. as a websocket client i have chosen the library jetty-websocket: org.eclipse.jetty jetty-websocket 8.1.0.rc5 test in the test case we build and start the websocket server to open a new connection to the websocket port. the websocket implementation of jetty-websocket allows us to implement two callback methods for the open and close events. within the open callback we send the test message to the client. the rest of the code waits for the connection to be established, closes it and asserts that the server has received the message: @test public void teststartandbuild() throws exception { subject = new websocketserver(); subject.buildandstartserver(8080, "127.0.0.1"); websocketclient client = new websocketclient(); future connectionfuture = client.open(new uri("ws://localhost:8080/websocket"), new websocket() { @override public void onopen(connection connection) { logger.info("onopen"); try { connection.sendmessage("testmessage"); } catch (ioexception e) { logger.error("failed to send message: "+e.getmessage(), e); } } @override public void onclose(int i, string s) { logger.info("onclose"); } }); websocket.connection connection = connectionfuture.get(2, timeunit.seconds); assertthat(connection, is(notnullvalue())); connection.close(); subject.stopserver(); thread.sleep(1000); assertthat(subject.lastreceivedmessage, is("testmessage")); } as usual you can find the source code on github . conclusion: undertow’s builder api makes it easy to construct a websocket server and in general an embedded webserver that fits your needs. this also eases automatic testing as you do not need any specific maven plugin that starts and stops your server before and after your integration tests. beyond that the jquery plugin jquery-graceful-websocket lets you send and receive messages over websockets with only a few lines of code.

One alternative to using SQLite on Android is to store Java objects in SharedPreferences.

Your mission-critical projects need complex event processing, realtime management and monitoring. Talend 5.4 (released in December 2013, https://www.talend.com) offers a great new feature: Talend Event Logging. It allows logging, processing and monitoring of all technical events and business data. In this article, I will focus on how to process, filter and monitor business data. You can find more details about monitoring technical events and logs (e.g. OSGi events of the ESB container) in Talend’s documentation (www.help.talend.com). This new feature is very powerful, but also extendable to fit custom requirements. You can solve and monitor much more complex scenarios than the one I describe here. Talend Event Logging with Logstash, ElasticSearch and Kibana First, let’s take a look at the components / projects which are integrated and extended into Talend’s products for implementing this new feature. logstash (http://logstash.net) is a tool for managing events and logs. You can use it to collect different (distributed) logs, parse them, and store them for later use. Speaking of searching, logstash comes with a simple, but fine web interface for searching and drilling into all of your logs. It uses ElasticSearch under the hood. So, you can easily query through all your logs for specific errors or business analytics (e.g. searching for all lines matching an unique order id). Additional to the pure collection of events, the Event Logging feature supports custom processing (e.g. custom filtering, customer data enrichment/reduction), aggregation, signing and also server side custom pre- and post-processing of events - e.g. to send them to an intrusion detection system or to any other kind of potential higher level log processing /management system. Kibana (http://www.elasticsearch.org/overview/kibana) is a browser based analytics and search interface to logstash and other timestamped data sets stored in ElasticSearch. Kibana strives to be easy to get started with, while also being flexible and powerful, just like logstash and ElasticSearch. Main difference to logstash is a much more powerful HTML5 based web interface. You can • use multiple concurrent search inputs • highlight to drill down bar charts • create line charts, stacked, unstacked, filled or unfilled, with or without points • create Pie and donut charts that compare top terms or the results of multiple queries • create custom dashboards with multiple charts • and much more… Therefore, logstash, Elasticsearch and Kibana are a perfect combination. You can use Kibana to analyze and monitor your data as you do with logstash, however, Kibana’s web interface is much more powerful and comfortable than logstash. There is a great book about logstash: “The logstash book” – for just 9.99 USD. I can really recommend this book for getting started: http://www.logstashbook.com/. For Elasticsearch, you can find several books on Amazon. Unfortunately, Kibana has no good and extensive documentation yet. I heard from its developers that this topic is addressed for Q1 2014. Integration of Event Logging and Monitoring into Talend’s Unified Platform Talend 5.4 has integrated logstash and Kibana into its Unified Platform. Talend Administration Center (TAC) is Talend’s central web application for management and monitoring. It got a new logging view: Here, you can use very flexible and powerful realtime search capabilities of Elasticsearch. Some dashboards are available by default. You can also create your own custom dashboards easily within this site thanks to Kibana. Many panel types are available, such as pie, histogram, table, hits or trends. You can analyze every technical event or business data down to the message level: By default, you see fields such as message, source, timestamp, type, and others. Of course, you can also add custom fields suitable for your business case. This way, you have a central monitoring capability which allows analyzing data on distributed clusters easily. Under the hood, this data comes from logstash. Many alternative inputs are available for logstash, such as log4j input or tcp input. For business data, I often use file input (http://logstash.net/docs/1.2.2/inputs/file) to analyze files such as CSV. Adding new inputs is very simple. You just have to add an input to your logstash configuration file of your logstash server. As I mentioned already, all this is integrated into the Talend’s Unified Platform: In this example, there are three log4j inputs and one file input. I use the file input to analyze text files in a specific directory. In this case, output is an embedded Elasticsearch instance. In production, you should use an external Elasticsearch cluster, of course. Processing such as filtering can also be configured in this file. Thus, you can process, analyze and monitor all your different inputs within one central monitoring application thanks to Kibana. Building Talend Integration Jobs, Routes and Web Services As mentioned before, you can analyze almost all data with logstash, Elasticsearch and Kibana. It does not matter if your input is technical events from a container (e.g. OSGi events) or any business data such as CSV files, log4j logs, or something else. Talend implicitly supports technical events which are created by the ESB container, by MDM, etc. However, you can also add your custom business data from your Talend jobs (integration perspective), SOAP / REST Web Services (integration perspective) or Talend routes (mediation perspective), easily: This is just one example (part of Talend’s DI demos which are included in every DI installation). The job generates some random data and stores it to a CSV file. You just have to add the configured file or directory of tFileOutputDelimited to your logstash configuration using file input (with wild cards for more complex scenarios). That’s it. You can now monitor and analyze the business data in realtime. This example showed a Talend DI Job (i.e. ETL job). However, you can also monitor your business data from SOAP / REST Web Services or Talend Routes the same way. Conclusion Your mission-critical projects need management and monitoring. Today, this is not just possible with complex and expensive tools of large vendors, but also with Talend’s Unified Platform products such as Talend ESB or Talend MDM. Under the hood, Talend integrates and extends widely used open source products: logstash, Elasticsearch and Kibana. Have fun with Talend 5.4’s new event logging and monitoring features… Best regards, Kai Wähner (@KaiWaehner) CONTENT FROM MY BLOG: http://www.kai-waehner.de/blog/2013/12/17/realtime-event-logging-complex-event-processing-cep-and-monitoring-with-talends-unified-platform-5-4-di-esb-dq-mdm-bpm-using-elasticsearch-logstash-and-kibana/

Editor's note: Be sure to check out part 2 as well. This is the fourth of an introductory series of blogs on Apache HBase. In the third part, we saw a high level view of HBase architecture . In this part, we'll use the HBase Java API to create tables, insert new data, and retrieve data by row key. We'll also see how to setup a basic table scan which restricts the columns retrieved and also uses a filter to page the results. Having just learned about HBase high-level architecture, now let's look at the Java client API since it is the way your applications interact with HBase. As mentioned earlier you can also interact with HBase via several flavors of RPC technologies like Apache Thrift plus a REST gateway, but we're going to concentrate on the native Java API. The client APIs provide both DDL (data definition language) and DML (data manipulation language) semantics very much like what you find in SQL for relational databases. Suppose we are going to store information about people in HBase, and we want to start by creating a new table. The following listing shows how to create a new table using the HBaseAdmin class. Configuration conf = HBaseConfiguration.create(); HBaseAdmin admin = new HBaseAdmin(conf); HTableDescriptor tableDescriptor = new HTableDescriptor(TableName.valueOf("people")); tableDescriptor.addFamily(new HColumnDescriptor("personal")); tableDescriptor.addFamily(new HColumnDescriptor("contactinfo")); tableDescriptor.addFamily(new HColumnDescriptor("creditcard")); admin.createTable(tableDescriptor); The people table defined in preceding listing contains three column families: personal, contactinfo, and creditcard. To create a table you create an HTableDescriptor and add one or more column families by adding HColumnDescriptor objects. You then call createTable to create the table. Now we have a table, so let's add some data. The next listing shows how to use the Put class to insert data on John Doe, specifically his name and email address (omitting proper error handling for brevity). Configuration conf = HBaseConfiguration.create(); HTable table = new HTable(conf, "people"); Put put = new Put(Bytes.toBytes("doe-john-m-12345")); put.add(Bytes.toBytes("personal"), Bytes.toBytes("givenName"), Bytes.toBytes("John")); put.add(Bytes.toBytes("personal"), Bytes.toBytes("mi"), Bytes.toBytes("M")); put.add(Bytes.toBytes("personal"), Bytes.toBytes("surame"), Bytes.toBytes("Doe")); put.add(Bytes.toBytes("contactinfo"), Bytes.toBytes("email"), Bytes.toBytes("[email protected]")); table.put(put); table.flushCommits(); table.close(); In the above listing we instantiate a Put providing the unique row key to the constructor. We then add values, which must include the column family, column qualifier, and the value all as byte arrays. As you probably noticed, the HBase API's utility Bytes class is used a lot; it provides methods to convert to and from byte[] for primitive types and strings. (Adding a static import for the toBytes() method would cut out a lot of boilerplate code.) We then put the data into the table, flush the commits to ensure locally buffered changes take effect, and finally close the table. Updating data is also done via the Put class in exactly the same manner as just shown in the prior listing. Unlike relational databases in which updates must update entire rows even if only one column changed, if you only need to update a single column then that's all you specify in the Put and HBase will only update that column. There is also a checkAndPut operation which is essentially a form of optimistic concurrency control - the operation will only put the new data if the current values are what the client says they should be. Retrieving the row we just created is accomplished using the Get class, as shown in the next listing. (From this point forward, listings will omit the boilerplate code to create a configuration, instantiate the HTable, and the flush and close calls.) Get get = new Get(Bytes.toBytes("doe-john-m-12345")); get.addFamily(Bytes.toBytes("personal")); get.setMaxVersions(3); Result result = table.get(get); The code in the previous listing instantiates a Get instance supplying the row key we want to find. Next we use addFamily to instruct HBase that we only need data from the personal column family, which also cuts down the amount of work HBase must do when reading information from disk. We also specify that we'd like up to three versions of each column in our result, perhaps so we can list historical values of each column. Finally, calling get returns a Result instance which can then be used to inspect all the column values returned. In many cases you need to find more than one row. HBase lets you do this by scanning rows, as shown in the second part which showed using a scan in the HBase shell session. The corresponding class is the Scan class. You can specify various options, such as the start and ending row key to scan, which columns and column families to include and the maximum versions to retrieve. You can also add filters, which allow you to implement custom filtering logic to further restrict which rows and columns are returned. A common use case for filters is pagination. For example, we might want to scan through all people whose last name is Smith one page (e.g. 25 people) at a time. The next listing shows how to perform a basic scan. Scan scan = new Scan(Bytes.toBytes("smith-")); scan.addColumn(Bytes.toBytes("personal"), Bytes.toBytes("givenName")); scan.addColumn(Bytes.toBytes("contactinfo"), Bytes.toBytes("email")); scan.setFilter(new PageFilter(25)); ResultScanner scanner = table.getScanner(scan); for (Result result : scanner) { // ... } In the above listing we create a new Scan that starts from the row key smith- and we then use addColumn to restrict the columns returned (thus reducing the amount of disk transfer HBase must perform) to personal:givenName and contactinfo:email. A PageFilter is set on the scan to limit the number of rows scanned to 25. (An alternative to using the page filter would be to specify a stop row key when constructing the Scan.) We then get a ResultScanner for the Scan just created, and loop through the results performing whatever actions are necessary. Since the only method in HBase to retrieve multiple rows of data is scanning by sorted row keys, how you design the row key values is very important. We'll come back to this topic later. You can also delete data in HBase using the Delete class, analogous to the Put class to delete all columns in a row (thus deleting the row itself), delete column families, delete columns, or some combination of those. Connection Handling In the above examples not much attention was paid to connection handling and RPCs (remote procedure calls). HBase provides the HConnection class which provides functionality similar to connection pool classes to share connections, for example you use the getTable() method to get a reference to an HTable instance. There is also an HConnectionManager class which is how you get instances of HConnection. Similar to avoiding network round trips in web applications, effectively managing the number of RPCs and amount of data returned when using HBase is important, and something to consider when writing HBase applications. Conclusion to Part 4 In this part we used the HBase Java API to create a people table, insert a new person, and find the newly inserted person information. We also used the Scan class to scan the people table for people with last name "Smith" and showed how to restrict the data retrieved and finally how to use a filter to limit the number of results. In the next part, we'll learn how to deal with the absence of SQL and relations when modeling schemas in HBase. References HBase web site, http://hbase.apache.org/ HBase wiki, http://wiki.apache.org/hadoop/Hbase HBase Reference Guide http://hbase.apache.org/book/book.html HBase: The Definitive Guide, http://bit.ly/hbase-definitive-guide Google Bigtable Paper, http://labs.google.com/papers/bigtable.html Hadoop web site, http://hadoop.apache.org/ Hadoop: The Definitive Guide, http://bit.ly/hadoop-definitive-guide Fallacies of Distributed Computing, http://en.wikipedia.org/wiki/Fallacies_of_Distributed_Computing HBase lightning talk slides, http://www.slideshare.net/scottleber/hbase-lightningtalk Sample code, https://github.com/sleberknight/basic-hbase-examples

In Lucene's facet module we recently added support for dynamic range faceting, to show how many hits match each of a dynamic set of ranges. For example, the Updated drill-down in the Lucene/Solr issue search application uses range facets. Another example is distance facets (< 1 km, < 2 km, etc.), where the distance is dynamically computed based on the user's current location. Price faceting might also use range facets, if the ranges cannot be established during indexing. To implement range faceting, for each hit, we first calculate the value (the distance, the age, the price) to be aggregated, and then lookup which ranges match that value and increment its counts. Today we use a simple linear search through all ranges, which has O(N) cost, where N is the number of ranges. But this is inefficient! Segment Trees There are fun data structures like segment trees and interval trees with O(log(N) + M) cost per lookup, where M is the number of ranges that match the given value. I chose to explore segment trees, as Lucene only requires looking up by a single value (interval trees can also efficiently look up all ranges overlapping a provided range) and also because all the ranges are known up front (interval trees also support dynamically adding or removing ranges). If the ranges will never overlap, you can use a simple binary search; Guava's ImmutableRangeSet takes this approach. However, Lucene's range faceting allows overlapping ranges so we can't do that. Segment trees are simple to visualize: you "project" all ranges down on top of one another, creating a one-dimensional Venn diagram, to define the elementary intervals. This classifies the entire range of numbers into a minimal number of distinct ranges, each elementary interval, such that all points in each elementary interval always match the same set of input ranges. The lookup process is then a binary search to determine which elementary interval a point belongs to, recording the matched ranges as you recurse down the tree. Consider these ranges; the lower number is inclusive and the upper number is exclusive: 0: 0 – 10 1: 0 – 20 2: 10 – 30 3: 15 – 50 4: 40 – 70 The elementary intervals (think Venn diagram!) are: -∞ – 0 0 – 10 10 – 15 15 – 20 20 – 30 30 – 40 40 – 50 50 – 70 70 – ∞ Finally, you build a binary tree on top of the elementary ranges, and then add output range indices to both internal nodes and the leaves of that tree, necessary to prevent adversarial cases that would require too much (O(N^2)) space. During lookup, as you walk down the tree, you gather up the output ranges (indices) you encounter; for our example, each elementary range is assigned the follow range indices as outputs: -∞ – 0 → 0 – 10 → 0 10 – 15 → 1, 2 15 – 20 → 1, 2, 3 20 – 30 → 2, 3 30 – 40 → 3 40 – 50 → 3, 4 50 – 70 → 4 70 – ∞ → Some ranges correspond to 1 elementary interval, while other ranges correspond to 2 or 3 or more, in general. Some, 2 in this example, may have no matching input ranges. Looking Up Matched Ranges I've pushed all sources described below to new Google code project; the code is still somewhat rough and exploratory, so there are likely exciting bugs lurking, but it does seem to work: it includes (passing!) tests and simple micro-benchmarks. I started with a basic segment tree implementation as described on the Wikipedia page, for long values, called SimpleLongRangeMultiSet; here's the recursive lookup method: private int lookup(Node node, long v, int[] answers, int upto) { if (node.outputs != null) { for(int range : node.outputs) { answers[upto++] = range; } } if (node.left != null) { if (v <= node.left.end) { upto = lookup(node.left, v, answers, upto); } else { upto = lookup(node.right, v, answers, upto); } } return upto; } This worked correctly, but I realized there must be non-trivial overhead for the recursion, checking for nulls, the for loop over the output values, etc. Next, I tried switching to parallel arrays to hold the binary tree (ArrayLongRangeMultiSet), where the left child of node N is at 2*N and the right child is at 2*N+1, but this turned out to be slower. After that I tested a code specializing implementation, first by creating dynamic Java source code from the binary tree. This eliminates the recursion and creates a single simple method that uses a series of if statements, specialized to the specific ranges, to do the binary search and record the range indices. Here's the resulting specialized code, compiled from the above ranges: void lookup(long v, int[] answers) { int upto = 0; if (v <= 19) { if (v <= 9) { if (v >= 0) { answers[upto++] = 0; answers[upto++] = 1; } } else { answers[upto++] = 1; answers[upto++] = 2; if (v >= 15) { answers[upto++] = 3; } } } else { if (v <= 39) { answers[upto++] = 3; if (v <= 29) { answers[upto++] = 2; } } else { if (v <= 49) { answers[upto++] = 3; answers[upto++] = 4; } else { if (v <= 69) { answers[upto++] = 4; } } } } } Finally, using the ASM library, I compiled the tree directly to specialized Java bytecode, and this proved to be fastest (up to 2.5X faster in some cases). As a baseline, I also added the simple linear search method, LinearLongRangeMultiSet; as long as you don't have too many ranges (say 10 or less), its performance is often better than the Java segment tree. The implementation also allows you to specify the allowed range of input values (for example, maybe all values are >=0 in your usage), which can save an if statement or two in the lookup method. Counting All Matched Ranges While the segment tree allows you to quickly look up all matching ranges for a given value, after a nice tip from fellow Lucene committee Robert Muir, we realized Lucene's range faceting does not need to know the ranges for each value; instead, it only requires the aggregated counts for each range in the end, after seeing many values. This leads to an optimization: compute counts for each elementary interval and then in the end, roll up those counts to get the count for each range. This will only work for single-valued fields, since for a multi-valued field you'd need to carefully never increment the same range more than once per hit. So based on that approach, I created a new LongRangeCounter abstract base class, and the SimpleLongRangeCounter Java implementation, and also the ASM specialized version, and the results are indeed faster (~20 to 50%) than using the lookup method to count; I'll use this approach with Lucene. Segment trees are normally always "perfectly" balanced but one final twist I explored was to use a training set of values to bias the order of the if statements. For example, if your ranges cover a tiny portion of the search space, as is the case for the Updated drill-down, then it should be faster to use a slightly unbalanced tree, by first checking if the value is less than the maximum range. However, in testing, while there are some cases where this "training" is a bit faster, often it's slower; I'm not sure why. Lucene I haven't folded this into Lucene yet, but I plan to; all the exploratory code lives in the segment-trees Google code project for now. Results on the micro-benchmarks can be entirely different once the implementations are folded into a "real" search application. While ASM is a powerful way to generate specialized code, and it gives sizable performance gains at least in the micro-benchmarks, it is an added dependency and complexity for ongoing development and many more developers know Java than ASM. It may also confuse hotspot, causing deoptimizations when there are multiple implementations for an abstract base class. Furthermore, if there are many ranges, the resulting specialized bytecode can be become quite large (but, still O(N*log(N)) in size), which may cause other problems. On balance I'm not sure the sizable performance gains (on a micro-benchmark) warrant using ASM in Lucene's range faceting.

Optimization is a very common technique in computer science and machine learning to search for the best (or good enough) solution. Optimization itself is a big topic and involves a wide range of mathematical techniques in different scenarios. In this post, I will be focusing on local search, which is a very popular technique in searching for an optimal solution based on a series of greedy local moves. The general setting of local search is as follows ... 1. Define an objective function 2. Pick an initial starting point 3. Repeat 3.1 Find a neighborhood 3.2 Locate a subset of neighbors that is a candidate move 3.3 Select a candidate from the candidate set 3.4 Move to the candidate One requirement is that the optimal solution need to be reachable by a sequence of moves. Usually this requires a proof that any arbitrary state is reachable by any arbitrary state through a sequence of moves. Notice that there are many possible strategies for each steps in 3.1, 3.2, 3.3. For example, one strategy can examine all members within the neighborhood, pick the best one (in terms of the objective function) and move to that. Another strategy can randomly pick a member within the neighborhood, and move to the member if it is better than the current state. Regardless of the strategies, the general theme is to move towards the members which is improving the objective function, hence the greedy nature of this algorithm. One downside of this algorithm is that it is possible to be trapped in a local optimum, whose is the best candidate within its neighborhood, but not the best candidate from a global sense. In the following, we'll explore a couple meta-heuristic techniques that can mitigate the local optimum trap. Multiple rounds We basically conduct k rounds of local search, each round getting a local optimum and then pick the best one out of these k local optimum. Simulated Anealing This strategy involves a dynamic combination of exploitation (better neighbor) and exploration (random walk to worse neighbor). The algorithm works in the following way ... 1. Pick an initial starting point 2. Repeat until terminate condition 2.1 Within neighborhood, pick a random member 2.2 If neighbor is better than me move to the neighbor else With chance exp(-(myObj - neighborObj)/Temp) move to the worse neighbor 2.3 Temp = alpha * Temp Tabu Search This strategy maintains a list of previously visited states (called Tabu list) and make sure these states will not be re-visited in subsequent exploration. The search will keep exploring the best move but skipping the previously visited nodes. This way the algorithm will explore the path that hasn't be visited before. The search also remember the best state obtained so far. 1. Initialization 1.1 Pick an initial starting point S 1.2 Initial an empty Tabu list 1.3 Set the best state to S 1.4 Put S into the Tabu list 2. Repeat until terminate condition 2.1 Find a neighborhood 2.2 Locate a smaller subset that is a candidate move 2.3 Remove elements that is already in Tabu list 2.4 Select the best candidate and move there 2.5 Add the new state to the Tabu list 2.6 If the new state is better that best state 2.6.1 Set the best state to this state 2.7 If the Tabu list is too large 2.7.1 Trim Tabu list by removing old items

To serialize a Business object marked with [DataContract] and properties marked with [DataMember] attributes… public static string DataContractSerializeObject(T objectToSerialize) { using (var output = new StringWriter()) { using (var writer = new XmlTextWriter(output) { Formatting = Formatting.Indented }) { var dataContractSerializer = new DataContractSerializer(typeof(T), EntityUtilities.GetAllKnownTypes(), int.MaxValue, true, true, null); dataContractSerializer.WriteObject(writer, objectToSerialize); return output.GetStringBuilder().ToString(); } } } Then to deserialize back to your DataContract type, use this logic… public static T Deserialize(string xml) { using (Stream stream = new MemoryStream()) { byte[] data = System.Text.Encoding.UTF8.GetBytes(xml); stream.Write(data, 0, data.Length); stream.Position = 0; var dataContractSerializer = new DataContractSerializer(typeof(T), EntityUtilities.GetAllKnownTypes(), int.MaxValue, true, true, null); return (T)dataContractSerializer.ReadObject(stream); } }

A typical Maven and Spring web application has a fair amount of XML and verbosity to it. Add in Jersey and Spring Security and you can have hundreds of lines of XML before you even start to write your Java code. As part of a recent project, I was tasked with upgrading a webapp like this to use Spring 4 and Spring Boot. I also figured I'd try to minimize the XML. This is my story on how I upgraded to Spring 4, Jersey 2, Java 8 and Spring Boot 0.5.0 M6. When I started, the app was using Spring 3.2.5, Spring Security 3.1.4 and Jersey 1.18. The pom.xml had four Jersey dependencies, three Spring dependencies and three Spring Security dependencies, along with a number of exclusions for "jersey-spring". Upgrading to Spring 4 Upgrading to Spring 4 was easy, I changed the version property to 4.0.0.RC2 and added the new Spring bill of materials to my pom.xml. I also add the Spring milestone repo since Spring 4 won't be released to Maven central until tomorrow. org.springframework spring-framework-bom ${spring.framework.version} pom import spring-milestones http://repo.spring.io/milestone true Next, I removed all the references to ${spring.framework.version} in dependencies since it'd be controlled by Maven's dependency management feature. org.springframework spring-web - ${spring.framework.version} I also changed to use Maven 3's wildcard syntax to exclude multiple dependencies. com.sun.jersey.contribs jersey-spring org.springframework - spring - - - org.springframework - spring-core - - - org.springframework - spring-web - - - org.springframework - spring-beans - - - org.springframework - spring-context + * I confirmed the upgrade worked by running "mvn dependency:tree | grep spring", followed by "mvn jetty:run" and viewing the app in my browser. Upgrading to Jersey 2 The next item I tackled was upgrading to Jersey 2.4.1. I changed the version number in my pom.xml, then added the Jersey BOM. org.glassfish.jersey jersey-bom ${jersey.version} pom import You might ask "why Jersey?" if we already have Spring MVC and its REST support? You might also ask why not Play or Grails instead of a Java + Spring stack? For this particular project, I recommended technology options, and these were certainly among them. However, the team chose differently and I support their decision. The project is creating an iOS app, as well as a responsive HTML5 mobile/desktop app. We figured we had enough risk with new technologies on the front-end that we should play it a bit safer on the backend. To make the backend work a bit sexier, we've decided to allow Spring 4, Java 8 and possibly some reactive principles. Next, I changed from the old com.sun.jersey dependencies to org.glassfish.jersey and removed jersey-spring. org.glassfish.jersey.containers jersey-container-servlet org.glassfish.jersey.media jersey-media-json-jackson The last thing I needed to do was change the servlet-class and param-name in web.xml: jersey-servlet org.glassfish.jersey.servlet.ServletContainer jersey.config.server.provider.packages com.raibledesigns.boot.service 1 Requiring Java 8 Requiring Java 8 to compile was easy enough. I added the maven-compiler-plugin to enforce a minimum version. maven-compiler-plugin 3.1 1.8 1.8 I downloaded the latest Java 8 SDK and installed it. Then I set my JAVA_HOME to use it. export JAVA_HOME=`/usr/libexec/java_home -v 1.8` Integrating Spring Boot I learned about Spring Boot a few weeks ago at Devoxx. Josh Long gave me a 3-minute demo at the speaker's dinner and showed me enough to peak my interest. To integrate it into my project, I started with the Quick Start. I added the boot-parent, dependencies for web, security and actuator (logging, metrics, etc.) and the Maven plugin. I removed all the Spring and Spring Security dependencies. org.springframework.boot spring-boot-starter-parent 0.5.0.M6 ... spring-milestones http://repo.spring.io/milestone ... org.springframework.boot spring-boot-starter-web org.springframework.boot spring-boot-starter-security org.springframework.boot spring-boot-starter-actuator ... org.springframework.boot spring-boot-maven-plugin Upon restarting my app, I got an error about spring-security.xml using a 3.1 XSD. I fixed it by changing to 3.2. Next, I wanted to eliminate web.xml. First of all, I created an ApplicationInitializer so the WAR could be started from the command line. package com.raibledesigns.boot.config; import org.springframework.boot.SpringApplication; import org.springframework.boot.autoconfigure.EnableAutoConfiguration; import org.springframework.boot.builder.SpringApplicationBuilder; import org.springframework.boot.web.SpringBootServletInitializer; import org.springframework.context.annotation.ComponentScan; import org.springframework.context.annotation.Configuration; @Configuration @EnableAutoConfiguration @ComponentScan public class ApplicationInitializer extends SpringBootServletInitializer { @Override protected SpringApplicationBuilder configure(SpringApplicationBuilder application) { return application.sources(ApplicationInitializer.class); } public static void main(String[] args) { SpringApplication.run(ApplicationInitializer.class, args); } } However, after adding this, I received the following error on startup: org.springframework.beans.factory.BeanCreationException: Error creating bean with name 'org.springframework.boot.context.properties.ConfigurationPropertiesBindingPostProcessor': Invocation of init method failed; nested exception is java.lang.AbstractMethodError: org.hibernate.validator.internal.engine.ConfigurationImpl .getDefaultParameterNameProvider()Ljavax/validation/ParameterNameProvider; Adding hibernate-validator as a dependency solved this problem: org.hibernate hibernate-validator To configure Spring Security without web.xml and spring-security.xml, I created WebSecurityConfig.java: package com.raibledesigns.boot.config; import org.springframework.context.annotation.Configuration; import org.springframework.core.Ordered; import org.springframework.core.annotation.Order; import org.springframework.security.config.annotation.authentication.builders.AuthenticationManagerBuilder; import org.springframework.security.config.annotation.web.builders.HttpSecurity; import org.springframework.security.config.annotation.web.configuration.EnableWebSecurity; import org.springframework.security.config.annotation.web.configuration.WebSecurityConfigurerAdapter; @Configuration @EnableWebSecurity @Order(Ordered.LOWEST_PRECEDENCE - 6) public class WebSecurityConfig extends WebSecurityConfigurerAdapter { @Override protected void configure(HttpSecurity http) throws Exception { http.authorizeRequests() .antMatchers("/", "/home").permitAll() .antMatchers("/v1.0/**").hasRole("USER") .anyRequest().authenticated(); http.httpBasic().realmName("My API"); } @Override protected void configure(AuthenticationManagerBuilder authManagerBuilder) throws Exception { authManagerBuilder.inMemoryAuthentication() .withUser("test").password("test123").roles("USER"); } } To configure Jersey without web.xml, I created a JerseyConfig class: package com.raibledesigns.boot.config; import org.glassfish.jersey.filter.LoggingFilter; import org.glassfish.jersey.jackson.JacksonFeature; import org.glassfish.jersey.server.ResourceConfig; import org.glassfish.jersey.server.ServerProperties; import javax.ws.rs.ApplicationPath; @ApplicationPath("/v1.0") public class JerseyConfig extends ResourceConfig { public JerseyConfig() { packages("com.raibledesigns.boot.service"); property(ServerProperties.BV_SEND_ERROR_IN_RESPONSE, true); property(ServerProperties.JSON_PROCESSING_FEATURE_DISABLE, false); property(ServerProperties.MOXY_JSON_FEATURE_DISABLE, true); property(ServerProperties.WADL_FEATURE_DISABLE, true); register(LoggingFilter.class); register(JacksonFeature.class); } } Finally, I created MvcConfig.java to set the welcome page. package com.raibledesigns.boot.config; import org.springframework.context.annotation.Configuration; import org.springframework.web.servlet.config.annotation.ViewControllerRegistry; import org.springframework.web.servlet.config.annotation.WebMvcConfigurerAdapter; @Configuration public class MvcConfig extends WebMvcConfigurerAdapter { @Override public void addViewControllers(ViewControllerRegistry registry) { registry.addViewController("/").setViewName("index"); } } To cleanup, I deleted src/main/webapp/WEB-INF and created src/main/resources/logback.xml: Since Boot doesn't support JSP out-of-the-box, I renamed my index.jsp file to index.html and changed the URL in it to point to "/v1.0/hello". I was pleased to see that everything worked nicely. I learned shortly after that I could remove the Spring BOM since Spring Boot uses a property to control its Spring version. The only issue I found is when started the app with "mvn package && java -jar target/app.war", it failed to initialize Jersey. I tried adding a @Bean for the servlet: @Bean public ServletRegistrationBean jerseyServlet() { ServletRegistrationBean registration = new ServletRegistrationBean(new ServletContainer(), "/v1.0/*"); registration.addInitParameter(ServletProperties.JAXRS_APPLICATION_CLASS, JerseyConfig.class.getName()); return registration; } Unfortunately, when running it using "java -jar", I get the following error: org.glassfish.hk2.api.MultiException: A MultiException has 1 exceptions. They are: 1. org.glassfish.jersey.server.internal.scanning.ResourceFinderException: java.io.FileNotFoundException: /.../target/app.war!/WEB-INF/classes (No such file or directory) at org.jvnet.hk2.internal.Utilities.justCreate(Utilities.java:869) at org.jvnet.hk2.internal.ServiceLocatorImpl.create(ServiceLocatorImpl.java:814) at org.jvnet.hk2.internal.ServiceLocatorImpl.createAndInitialize(ServiceLocatorImpl.java:906) at org.jvnet.hk2.internal.ServiceLocatorImpl.createAndInitialize(ServiceLocatorImpl.java:898) at org.glassfish.jersey.server.ApplicationHandler.createApplication(ApplicationHandler.java:300) at org.glassfish.jersey.server.ApplicationHandler.(ApplicationHandler.java:279) at org.glassfish.jersey.servlet.WebComponent.(WebComponent.java:302) This seems strange since there is a WEB-INF/classes in my WAR. Regardless, this is not a Boot problem per se, but more of a Jersey issue. From one of the Boot developers: The whole idea with Boot is that servlets are just a transport - they are a means to an end, and hopefully not the only one - the "container" is Spring, not the servlet container. We probably could add some form of support for SCI but only by hacking the containers since the spec really doesn't allow for much control of their lifecycle. It hasn't been a priority so far. Summary I hope this article is useful to see how you to upgrade your Java webapps to use Spring 4 and Spring Boot. I've created a boot-makeover project on GitHub with all the code mentioned. You can also view the commits for each step.