java-design-patterns/localization/zh/dao/README.md

layout, title, folder, permalink, categories, language, tags

layout	title	folder	permalink	categories	language	tags
pattern	Data Access Object	dao	/patterns/dao/	Architectural	zh	Data access

目的

对象为某种类型的数据库或其他持久性机制提供了抽象接口。

解释

真实世界例子

有一组客户数据需要持久化到数据库中。 我们需要整个额外的增删改查操作以便操作客户数据。

通俗的说

DAO是我们通过基本持久性机制提供的接口。

维基百科说

在计算机软件中，数据访问对象（DAO）是一种模式，可为某种类型的数据库或其他持久性机制提供抽象接口。

程序示例

通过我们的客户示例，下面是基本的客户实体。

public class Customer {

  private int id;
  private String firstName;
  private String lastName;

  public Customer(int id, String firstName, String lastName) {
    this.id = id;
    this.firstName = firstName;
    this.lastName = lastName;
  }
  // getters and setters ->
  ...
}

这是CustomerDao接口及其两个不同的实现。

Here's the CustomerDao interface and two different implementations for it. InMemoryCustomerDao 将简单的客户数据映射保存在内存中 而DBCustomerDao是真正的RDBMS实现。

public interface CustomerDao {

  Stream<Customer> getAll() throws Exception;

  Optional<Customer> getById(int id) throws Exception;

  boolean add(Customer customer) throws Exception;

  boolean update(Customer customer) throws Exception;

  boolean delete(Customer customer) throws Exception;
}

public class InMemoryCustomerDao implements CustomerDao {

  private final Map<Integer, Customer> idToCustomer = new HashMap<>();

  // implement the interface using the map
  ...
}

public class DbCustomerDao implements CustomerDao {

  private static final Logger LOGGER = LoggerFactory.getLogger(DbCustomerDao.class);

  private final DataSource dataSource;

  public DbCustomerDao(DataSource dataSource) {
    this.dataSource = dataSource;
  }

  // implement the interface using the data source
  ...

最后，这是我们使用DAO管理客户数据的方式。

    final var dataSource = createDataSource();
    createSchema(dataSource);
    final var customerDao = new DbCustomerDao(dataSource);
    
    addCustomers(customerDao);
    log.info(ALL_CUSTOMERS);
    try (var customerStream = customerDao.getAll()) {
      customerStream.forEach((customer) -> log.info(customer.toString()));
    }
    log.info("customerDao.getCustomerById(2): " + customerDao.getById(2));
    final var customer = new Customer(4, "Dan", "Danson");
    customerDao.add(customer);
    log.info(ALL_CUSTOMERS + customerDao.getAll());
    customer.setFirstName("Daniel");
    customer.setLastName("Danielson");
    customerDao.update(customer);
    log.info(ALL_CUSTOMERS);
    try (var customerStream = customerDao.getAll()) {
      customerStream.forEach((cust) -> log.info(cust.toString()));
    }
    customerDao.delete(customer);
    log.info(ALL_CUSTOMERS + customerDao.getAll());
    
    deleteSchema(dataSource);

程序输出:

customerDao.getAllCustomers(): 
Customer{id=1, firstName='Adam', lastName='Adamson'}
Customer{id=2, firstName='Bob', lastName='Bobson'}
Customer{id=3, firstName='Carl', lastName='Carlson'}
customerDao.getCustomerById(2): Optional[Customer{id=2, firstName='Bob', lastName='Bobson'}]
customerDao.getAllCustomers(): java.util.stream.ReferencePipeline$Head@7cef4e59
customerDao.getAllCustomers(): 
Customer{id=1, firstName='Adam', lastName='Adamson'}
Customer{id=2, firstName='Bob', lastName='Bobson'}
Customer{id=3, firstName='Carl', lastName='Carlson'}
Customer{id=4, firstName='Daniel', lastName='Danielson'}
customerDao.getAllCustomers(): java.util.stream.ReferencePipeline$Head@2db0f6b2
customerDao.getAllCustomers(): 
Customer{id=1, firstName='Adam', lastName='Adamson'}
Customer{id=2, firstName='Bob', lastName='Bobson'}
Customer{id=3, firstName='Carl', lastName='Carlson'}
customerDao.getCustomerById(2): Optional[Customer{id=2, firstName='Bob', lastName='Bobson'}]
customerDao.getAllCustomers(): java.util.stream.ReferencePipeline$Head@12c8a2c0
customerDao.getAllCustomers(): 
Customer{id=1, firstName='Adam', lastName='Adamson'}
Customer{id=2, firstName='Bob', lastName='Bobson'}
Customer{id=3, firstName='Carl', lastName='Carlson'}
Customer{id=4, firstName='Daniel', lastName='Danielson'}
customerDao.getAllCustomers(): java.util.stream.ReferencePipeline$Head@6ec8211c

类图

适用性

在以下情况下，请使用数据访问对象：:

• 当您要巩固如何访问数据层时。
• 当您要避免编写多个数据检索/持久层时。

鸣谢

• J2EE Design Patterns