Chapter12.md

Chapter 12: Starting with Persistance

Data Sources

1. What is a data source

• A data source is a component between the data base and the app that manages connections to the database

• The data source's goal is to reuse connections to the database throught the application and to only create connections when necessary

• This is used because establishing a connection to the database is an expensive operation and so we try to reuse the connections as much as possible

• The technique we described above is also known as Connection Pooling and componenets providing such functionality are sometimes called Connection Pools

2. Default data source

• By default, spring boot configures HikariCP as the default data source for our application

• We don't need to request connections for the data source ourselves if we are not writing data access directly using the JDK JDBC inteface (such as when using jdbcTempalte)

3. Defining custom data sources

• we might want to define multiple or custom data sources due to a couple of reasons such as:

  • having multiple data bases
  • wanting to initlize the source based on information only available at runtime
  • etc..

• we do that through the configuraiton class

  @Configuration
  public class ProjectConfig {
      @Value("${custom.datasource.url}") //these values are retrieved from the project properties files
      private String datasourceUrl;
      @Value("${custom.datasource.username}")
      private String datasourceUsername;
      @Value("${custom.datasource.password}")
      private String datasourcePassword;
  
      @Bean
      public DataSource dataSource() {
          HikariDataSource dataSource = new HikariDataSource();
          dataSource.setJdbcUrl(datasourceUrl);
          dataSource.setUsername(datasourceUsername);
          dataSource.setPassword(datasourcePassword);
          dataSource.setConnectionTimeout(1000);
          return dataSource;
      }
  }

• DataSource is an interfae and HikariDataSource is an implementation to that interface

• By default spring uses the data found in spring.datasource. properties to initlize the data source without us explicity using these properties

Using jdbcTemplate for persistance

• jdbcTempalte is a class that provides an abstraction over the JDK JDBC API to reduce boilerplate code required for interacting with a db

• The special thing about jdbcTempalte is that it requires an object called a RowMapper to put each row in a model object, the RowMapper is just a lamda specifying the mapping

• Example showing the execution of a SELECT and mapping of rows

  public list<Purchase> findPurchase(int purchaseID){
      String stmt = "SELECT * FROM purchase WHERE id = ?";
  
      RowMapper<Purchase> rowMapper = (r,i) -> { // r is the result set and i is the row number
          Purchase rowObject = new Purchase();
          rowObject.setId(r.getInt("id"));
          rowObject.setProduct(r.getString("product"));
          rowObject.setPrice(r.getBigDecimal("price"));
          return rowObject;
      }
  
      jdbc.query(stmt, new Object[]{purchaseID}, rowMapper)
  }

• There are "variations" on jdbcTemplate such as namedParameterJdbcTempalte that makes certain tasks like passing parameters even easier.