June 16, 2014 Leave a comment
This is the first part of a 3 part blog/screencast on creating an end to end application that scrape data of a source and then re exposed these data as RESTful web services.
In the first part, we will look at pulling data into our web application.
In the second part, we will expose these data as RESTful web services
In the third and final part, we will write a mobile client to consume the web services.
In this first part, we will be reading RSS feed from BBC Football site (the RSS feed is right at the bottom of the page). The feed entries are saved to Mongo. The following diagram illustrates this
The feed is read at regular interval. This feature is implemented using JavaEE 7’s ManagedScheduledExecutorService which is a thread pool managed by the application server.
JavaEE application servers supports the concept of connection pooling, in particular database connection pool. However this support only extends to RDBMS. So if you’re using NoSQL databases, then you’re out of luck. One solution is to implement custom resources like this particular example for Glassfish.
Luckily for us, MongoClient, the Mongo’s Java driver has an internal pool (and is thread-safe!). So we can create one instance of MongoClient and get this instance to dish out connection as and when its required. To using this instance as a singleton, we use CDI to wrap it in an ApplicationScoped object. This ApplicationScoped object ensures that only a single instance of MongoClient is created and used throughout the entire application.
The following screencast shows how we create the a JavaEE web application, read RSS feed from BBC Football website and save that into a Mongo database.
The source code can be found here.
Till next time.