8
头图

作者:小傅哥
博客:https://bugstack.cn

沉淀、分享、成长,让自己和他人都能有所收获!😄

一、前言

Mybatis 最核心的原理也是它最便于使用的体现,为什么这说?

因为我们在使用 Mybatis 的时候,只需要定义一个不需要写实现类的接口,就能通过注解或者配置SQL语句的方式,对数据库进行 CRUD 操作。

那么这是怎么做到的呢,其中有一点非常重要,就是在 Spring 中可以把你的代理对象交给 Spring 容器,这个代理对象就是可以当做是 DAO 接口的具体实现类,而这个被代理的实现类就可以完成对数据库的一个操作,也就是这个封装过程被称为 ORM 框架。

说了基本的流程,我们来做点测试,让大家可以动手操作起来!学知识,一定是上手,才能得到!你可以通过以下源码仓库进行练习

源码https://github.com/fuzhengwei/CodeGuide/wiki


二、把Bean塞到Spring容器,分几步

Bean注册

  • 关于Bean注册的技术场景,在我们日常用到的技术框架中,MyBatis 是最为常见的。通过在使用 MyBatis 时都只是定义一个接口不需要写实现类,但是这个接口却可以和配置的 SQL 语句关联,执行相应的数据库操作时可以返回对应的结果。那么这个接口与数据库的操作就用到的 Bean 的代理和注册。
  • 我们都知道类的调用是不能直接调用没有实现的接口的,所以需要通过代理的方式给接口生成对应的实现类。接下来再通过把代理类放到 Spring 的 FactoryBean 的实现中,最后再把这个 FactoryBean 实现类注册到 Spring 容器。那么现在你的代理类就已经被注册到 Spring 容器了,接下来就可以通过注解的方式注入到属性中。

按照这个实现方式,我们来操作一下,看看一个 Bean 的注册过程在代码中是如何实现的。

1. 定义接口

public interface IUserDao {

    String queryUserInfo();

}
  • 先定义一个类似 DAO 的接口,基本这样的接口在使用 MyBatis 时还是非常常见的。后面我们会对这个接口做代理和注册。

2. 类代理实现

ClassLoader classLoader = Thread.currentThread().getContextClassLoader();
Class<?>[] classes = {IUserDao.class};    

InvocationHandler handler = (proxy, method, args) -> "你被代理了 " + method.getName();
IUserDao userDao = (IUserDao) Proxy.newProxyInstance(classLoader, classes, handler); 

String res = userDao.queryUserInfo();
logger.info("测试结果:{}", res);
  • Java 本身的代理方式使用起来还是比较简单的,用法也很固定。
  • InvocationHandler 是个接口类,它对应的实现内容就是代理对象的具体实现。
  • 最后就是把代理交给 Proxy 创建代理对象,Proxy.newProxyInstance

3. 实现Bean工厂

public class ProxyBeanFactory implements FactoryBean {

    @Override
    public Object getObject() throws Exception {

        ClassLoader classLoader = Thread.currentThread().getContextClassLoader();
        Class[] classes = {IUserDao.class};
        InvocationHandler handler = (proxy, method, args) -> "你被代理了 " + method.getName();

        return Proxy.newProxyInstance(classLoader, classes, handler);
    }

    @Override
    public Class<?> getObjectType() {
        return IUserDao.class;
    } 

}
  • FactoryBean 在 spring 起到着二当家的地位,它将近有70多个小弟(实现它的接口定义),那么它有三个方法;

    • T getObject() throws Exception; 返回bean实例对象
    • Class<?> getObjectType(); 返回实例类类型
    • boolean isSingleton(); 判断是否单例,单例会放到Spring容器中单实例缓存池中
  • 在这里我们把上面使用Java代理的对象放到了 getObject() 方法中,那么现在再从 Spring 中获取到的对象,就是我们的代理对象了。

4. Bean 注册

public class RegisterBeanFactory implements BeanDefinitionRegistryPostProcessor {

    @Override
    public void postProcessBeanDefinitionRegistry(BeanDefinitionRegistry registry) throws BeansException {

        GenericBeanDefinition beanDefinition = new GenericBeanDefinition();
        beanDefinition.setBeanClass(ProxyBeanFactory.class);

        BeanDefinitionHolder definitionHolder = new BeanDefinitionHolder(beanDefinition, "userDao");
        BeanDefinitionReaderUtils.registerBeanDefinition(definitionHolder, registry);
    }

}

在 Spring 的 Bean 管理中,所有的 Bean 最终都会被注册到类 DefaultListableBeanFactory 中,以上这部分代码主要的内容包括:

  • 实现 BeanDefinitionRegistryPostProcessor.postProcessBeanDefinitionRegistry方法,获取 Bean 注册对象。
  • 定义 Bean,GenericBeanDefinition,这里主要设置了我们的代理类工厂。
  • 创建 Bean 定义处理类,BeanDefinitionHolder,这里需要的主要参数;定义 Bean 和名称 setBeanClass(ProxyBeanFactory.class)
  • 最后将我们自己的bean注册到spring容器中去,registry.registerBeanDefinition()

5. 测试验证

在上面我们已经把自定义代理的 Bean 注册到了 Spring 容器中,接下来我们来测试下这个代理的 Bean 被如何调用。

1. 定义 spring-config.xml

<bean id="userDao" class="org.itstack.interview.bean.RegisterBeanFactory"/>
  • 这里我们把 RegisterBeanFactory 配置到 spring 的 xml 配置中,便于启动时加载。

2. 单元测试

@Test
public void test_IUserDao() {
    BeanFactory beanFactory = new ClassPathXmlApplicationContext("spring-config.xml");
    IUserDao userDao = beanFactory.getBean("userDao", IUserDao.class);
    String res = userDao.queryUserInfo();
    logger.info("测试结果:{}", res);
}

测试结果

22:53:14.759 [main] DEBUG o.s.c.e.PropertySourcesPropertyResolver - Could not find key 'spring.liveBeansView.mbeanDomain' in any property source
22:53:14.760 [main] DEBUG o.s.b.f.s.DefaultListableBeanFactory - Returning cached instance of singleton bean 'userDao'
22:53:14.796 [main] INFO  org.itstack.interview.test.ApiTest - 测试结果:你被代理了 queryUserInfo

Process finished with exit code 0
  • 从测试结果可以看到,我们已经可以通过注入到Spring的代理Bean对象,实现我们的预期结果。
  • 其实这个过程也是很多框架中用到的方式,尤其是在一些中间件开发,类似的 ORM 框架都需要使用到。

三、手写个Mybatis

扩展上一篇源码分析工程;itstack-demo-mybatis,增加 like 包,模仿 Mybatis 工程。完整规程下载 https://github.com/fuzhengwei...

itstack-demo-mybatis
└── src
    ├── main
    │   ├── java
    │   │   └── org.itstack.demo
    │   │       ├── dao
    │   │       │    ├── ISchool.java        
    │   │       │    └── IUserDao.java    
    │   │       ├── like
    │   │       │    ├── Configuration.java
    │   │       │    ├── DefaultSqlSession.java
    │   │       │    ├── DefaultSqlSessionFactory.java
    │   │       │    ├── Resources.java
    │   │       │    ├── SqlSession.java
    │   │       │    ├── SqlSessionFactory.java
    │   │       │    ├── SqlSessionFactoryBuilder.java    
    │   │       │    └── SqlSessionFactoryBuilder.java    
    │   │       └── interfaces     
    │   │             ├── School.java    
    │   │            └── User.java
    │   ├── resources    
    │   │   ├── mapper
    │   │   │   ├── School_Mapper.xml
    │   │   │   └── User_Mapper.xml
    │   │   ├── props    
    │   │   │   └── jdbc.properties
    │   │   ├── spring
    │   │   │   ├── mybatis-config-datasource.xml
    │   │   │   └── spring-config-datasource.xml
    │   │   ├── logback.xml
    │   │   ├── mybatis-config.xml
    │   │   └── spring-config.xml
    │   └── webapp
    │       └── WEB-INF
    └── test
         └── java
             └── org.itstack.demo.test
                 ├── ApiLikeTest.java
                 ├── MybatisApiTest.java
                 └── SpringApiTest.java

关于整个 Demo 版本,并不是把所有 Mybatis 全部实现一遍,而是拨丝抽茧将最核心的内容展示给你,从使用上你会感受一模一样,但是实现类已经全部被替换,核心类包括;

  • Configuration
  • DefaultSqlSession
  • DefaultSqlSessionFactory
  • Resources
  • SqlSession
  • SqlSessionFactory
  • SqlSessionFactoryBuilder
  • XNode

1. 先测试下整个DemoJdbc框架

ApiLikeTest.test_queryUserInfoById()
@Test
public void test_queryUserInfoById() {
    String resource = "spring/mybatis-config-datasource.xml";
    Reader reader;
    try {
        reader = Resources.getResourceAsReader(resource);
        SqlSessionFactory sqlMapper = new SqlSessionFactoryBuilder().build(reader);
        SqlSession session = sqlMapper.openSession();
        
        try {
            User user = session.selectOne("org.itstack.demo.dao.IUserDao.queryUserInfoById", 1L);
            System.out.println(JSON.toJSONString(user));
        } finally {
            session.close();
            reader.close();
        }
    } catch (Exception e) {
        e.printStackTrace();
    }
}

一切顺利结果如下(新人往往会遇到各种问题);

{"age":18,"createTime":1576944000000,"id":1,"name":"水水","updateTime":1576944000000}

Process finished with exit code 0

可能乍一看这测试类完全和 MybatisApiTest.java 测试的代码一模一样呀,也看不出区别。其实他们的引入的包是不一样;

MybatisApiTest.java 里面引入的包
import org.apache.ibatis.io.Resources;
import org.apache.ibatis.session.SqlSession;
import org.apache.ibatis.session.SqlSessionFactory;
import org.apache.ibatis.session.SqlSessionFactoryBuilder;
ApiLikeTest.java 里面引入的包
import org.itstack.demo.like.Resources;
import org.itstack.demo.like.SqlSession;
import org.itstack.demo.like.SqlSessionFactory;
import org.itstack.demo.like.SqlSessionFactoryBuilder;

好!接下来我们开始分析这部分核心代码。

2. 加载XML配置文件

这里我们采用 mybatis 的配置文件结构进行解析,在不破坏原有结构的情况下,最大可能的贴近源码。mybatis 单独使用的使用的时候使用了两个配置文件;数据源配置、Mapper 映射配置,如下;

mybatis-config-datasource.xml & 数据源配置
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE configuration PUBLIC "-//mybatis.org//DTD Config 3.0//EN"
        "http://mybatis.org/dtd/mybatis-3-config.dtd">

<configuration>
    <environments default="development">
        <environment id="development">
            <transactionManager type="JDBC"/>
            <dataSource type="POOLED">
                <property name="driver" value="com.mysql.jdbc.Driver"/>
                <property name="url" value="jdbc:mysql://127.0.0.1:3306/itstack?useUnicode=true"/>
                <property name="username" value="root"/>
                <property name="password" value="123456"/>
            </dataSource>
        </environment>
    </environments>

    <mappers>
        <mapper resource="mapper/User_Mapper.xml"/>
        <mapper resource="mapper/School_Mapper.xml"/>
    </mappers>

</configuration>
User_Mapper.xml & Mapper 映射配置
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE mapper PUBLIC "-//mybatis.org//DTD Mapper 3.0//EN" "http://mybatis.org/dtd/mybatis-3-mapper.dtd">
<mapper namespace="org.itstack.demo.dao.IUserDao">

    <select id="queryUserInfoById" parameterType="java.lang.Long" resultType="org.itstack.demo.po.User">
        SELECT id, name, age, createTime, updateTime
        FROM user
        where id = #{id}
    </select>

    <select id="queryUserList" parameterType="org.itstack.demo.po.User" resultType="org.itstack.demo.po.User">
        SELECT id, name, age, createTime, updateTime
        FROM user
        where age = #{age}
    </select>

</mapper>

这里的加载过程与 mybaits 不同,我们采用 dom4j 方式。在案例中会看到最开始获取资源,如下;

ApiLikeTest.test_queryUserInfoById() & 部分截取
String resource = "spring/mybatis-config-datasource.xml";
    Reader reader;
    try {
        reader = Resources.getResourceAsReader(resource);
    ...

从上可以看到这是通过配置文件地址获取到了读取流的过程,从而为后面解析做基础。首先我们先看 Resources 类,整个是我们的资源类。

Resources.java & 资源类
/**
 * 博 客 | https://bugstack.cn
 * Create by 小傅哥 @2020
 */
public class Resources {

    public static Reader getResourceAsReader(String resource) throws IOException {
        return new InputStreamReader(getResourceAsStream(resource));
    }

    private static InputStream getResourceAsStream(String resource) throws IOException {
        ClassLoader[] classLoaders = getClassLoaders();
        for (ClassLoader classLoader : classLoaders) {
            InputStream inputStream = classLoader.getResourceAsStream(resource);
            if (null != inputStream) {
                return inputStream;
            }
        }
        throw new IOException("Could not find resource " + resource);
    }

    private static ClassLoader[] getClassLoaders() {
        return new ClassLoader[]{
                ClassLoader.getSystemClassLoader(),
                Thread.currentThread().getContextClassLoader()};
    }

}

这段代码方法的入口是getResourceAsReader,直到往下以此做了;

  1. 获取 ClassLoader 集合,最大限度搜索配置文件
  2. 通过 classLoader.getResourceAsStream 读取配置资源,找到后立即返回,否则抛出异常

3. 解析XML配置文件

配置文件加载后开始进行解析操作,这里我们也仿照 mybatis 但进行简化,如下;

SqlSessionFactory sqlMapper = new SqlSessionFactoryBuilder().build(reader);
SqlSessionFactoryBuilder.build() & 入口构建类
public DefaultSqlSessionFactory build(Reader reader) {
    SAXReader saxReader = new SAXReader();
    try {
        Document document = saxReader.read(new InputSource(reader));
        Configuration configuration = parseConfiguration(document.getRootElement());
        return new DefaultSqlSessionFactory(configuration);
    } catch (DocumentException e) {
        e.printStackTrace();
    }
    return null;
}
  • 通过读取流创建 xml 解析的 Document 类
  • parseConfiguration 进行解析 xml 文件,并将结果设置到配置类中,包括;连接池、数据源、mapper关系
SqlSessionFactoryBuilder.parseConfiguration() & 解析过程
private Configuration parseConfiguration(Element root) {
    Configuration configuration = new Configuration();
    configuration.setDataSource(dataSource(root.selectNodes("//dataSource")));
    configuration.setConnection(connection(configuration.dataSource));
    configuration.setMapperElement(mapperElement(root.selectNodes("mappers")));
    return configuration;
}
  • 在前面的 xml 内容中可以看到,我们需要解析出数据库连接池信息 datasource,还有数据库语句映射关系 mappers
SqlSessionFactoryBuilder.dataSource() & 解析出数据源
private Map<String, String> dataSource(List<Element> list) {
    Map<String, String> dataSource = new HashMap<>(4);
    Element element = list.get(0);
    List content = element.content();
    for (Object o : content) {
        Element e = (Element) o;
        String name = e.attributeValue("name");
        String value = e.attributeValue("value");
        dataSource.put(name, value);
    }
    return dataSource;
}
  • 这个过程比较简单,只需要将数据源信息获取即可
SqlSessionFactoryBuilder.connection() & 获取数据库连接
private Connection connection(Map<String, String> dataSource) {
    try {
        Class.forName(dataSource.get("driver"));
        return DriverManager.getConnection(dataSource.get("url"), dataSource.get("username"), dataSource.get("password"));
    } catch (ClassNotFoundException | SQLException e) {
        e.printStackTrace();
    }
    return null;
}
  • 这个就是jdbc最原始的代码,获取了数据库连接池
SqlSessionFactoryBuilder.mapperElement() & 解析SQL语句
private Map<String, XNode> mapperElement(List<Element> list) {
    Map<String, XNode> map = new HashMap<>();
    Element element = list.get(0);
    List content = element.content();
    for (Object o : content) {
        Element e = (Element) o;
        String resource = e.attributeValue("resource");
        try {
            Reader reader = Resources.getResourceAsReader(resource);
            SAXReader saxReader = new SAXReader();
            Document document = saxReader.read(new InputSource(reader));
            Element root = document.getRootElement();
            //命名空间
            String namespace = root.attributeValue("namespace");
            // SELECT
            List<Element> selectNodes = root.selectNodes("select");
            for (Element node : selectNodes) {
                String id = node.attributeValue("id");
                String parameterType = node.attributeValue("parameterType");
                String resultType = node.attributeValue("resultType");
                String sql = node.getText();
                // ? 匹配
                Map<Integer, String> parameter = new HashMap<>();
                Pattern pattern = Pattern.compile("(#\\{(.*?)})");
                Matcher matcher = pattern.matcher(sql);
                for (int i = 1; matcher.find(); i++) {
                    String g1 = matcher.group(1);
                    String g2 = matcher.group(2);
                    parameter.put(i, g2);
                    sql = sql.replace(g1, "?");
                }
                XNode xNode = new XNode();
                xNode.setNamespace(namespace);
                xNode.setId(id);
                xNode.setParameterType(parameterType);
                xNode.setResultType(resultType);
                xNode.setSql(sql);
                xNode.setParameter(parameter);
                
                map.put(namespace + "." + id, xNode);
            }
        } catch (Exception ex) {
            ex.printStackTrace();
        }
    }
    return map;
}
  • 这个过程首先包括是解析所有的sql语句,目前为了测试只解析 select 相关
  • 所有的 sql 语句为了确认唯一,都是使用;namespace + select中的id进行拼接,作为 key,之后与sql一起存放到 map 中。
  • 在 mybaits 的 sql 语句配置中,都有占位符,用于传参。where id = #{id} 所以我们需要将占位符设置为问号,另外需要将占位符的顺序信息与名称存放到 map 结构,方便后续设置查询时候的入参。

4. 创建DefaultSqlSessionFactory

最后将初始化后的配置类 Configuration,作为参数进行创建 DefaultSqlSessionFactory,如下;

public DefaultSqlSessionFactory build(Reader reader) {
    SAXReader saxReader = new SAXReader();
    try {
        Document document = saxReader.read(new InputSource(reader));
        Configuration configuration = parseConfiguration(document.getRootElement());
        return new DefaultSqlSessionFactory(configuration);
    } catch (DocumentException e) {
        e.printStackTrace();
    }
    return null;
}
DefaultSqlSessionFactory.java & SqlSessionFactory的实现类
public class DefaultSqlSessionFactory implements SqlSessionFactory {
    
    private final Configuration configuration;
    
    public DefaultSqlSessionFactory(Configuration configuration) {
        this.configuration = configuration;
    }
    
    @Override
    public SqlSession openSession() {
        return new DefaultSqlSession(configuration.connection, configuration.mapperElement);
    }
    
}
  • 这个过程比较简单,构造函数只提供了配置类入参
  • 实现 SqlSessionFactory 的 openSession(),用于创建 DefaultSqlSession,也就可以执行 sql 操作

5. 开启SqlSession

SqlSession session = sqlMapper.openSession();

上面这一步就是创建了DefaultSqlSession,比较简单。如下;

@Override
public SqlSession openSession() {
    return new DefaultSqlSession(configuration.connection, configuration.mapperElement);
}

6. 执行SQL语句

User user = session.selectOne("org.itstack.demo.dao.IUserDao.queryUserInfoById", 1L);

在 DefaultSqlSession 中通过实现 SqlSession,提供数据库语句查询和关闭连接池,如下;

SqlSession.java & 定义
public interface SqlSession {

    <T> T selectOne(String statement);

    <T> T selectOne(String statement, Object parameter);

    <T> List<T> selectList(String statement);

    <T> List<T> selectList(String statement, Object parameter);

    void close();
}

接下来看具体的执行过程,session.selectOne

DefaultSqlSession.selectOne() & 执行查询
public <T> T selectOne(String statement, Object parameter) {
    XNode xNode = mapperElement.get(statement);
    Map<Integer, String> parameterMap = xNode.getParameter();
    try {
        PreparedStatement preparedStatement = connection.prepareStatement(xNode.getSql());
        buildParameter(preparedStatement, parameter, parameterMap);
        ResultSet resultSet = preparedStatement.executeQuery();
        List<T> objects = resultSet2Obj(resultSet, Class.forName(xNode.getResultType()));
        return objects.get(0);
    } catch (Exception e) {
        e.printStackTrace();
    }
    return null;
}
  • selectOne 就objects.get(0);,selectList 就全部返回
  • 通过 statement 获取最初解析 xml 时候的存储的 select 标签信息;

    <select id="queryUserInfoById" parameterType="java.lang.Long" resultType="org.itstack.demo.po.User">
        SELECT id, name, age, createTime, updateTime
        FROM user
        where id = #{id}
    </select>
  • 获取 sql 语句后交给 jdbc 的 PreparedStatement 类进行执行
  • 这里还需要设置入参,我们将入参设置进行抽取,如下;

    private void buildParameter(PreparedStatement preparedStatement, Object parameter, Map<Integer, String> parameterMap) throws SQLException, IllegalAccessException {
    
          int size = parameterMap.size();
          // 单个参数
          if (parameter instanceof Long) {
              for (int i = 1; i <= size; i++) {
                  preparedStatement.setLong(i, Long.parseLong(parameter.toString()));
              }
              return;
          }
    
          if (parameter instanceof Integer) {
              for (int i = 1; i <= size; i++) {
                  preparedStatement.setInt(i, Integer.parseInt(parameter.toString()));
              }
              return;
          }
    
          if (parameter instanceof String) {
              for (int i = 1; i <= size; i++) {
                  preparedStatement.setString(i, parameter.toString());
              }
              return;
          }
    
          Map<String, Object> fieldMap = new HashMap<>();
          // 对象参数
          Field[] declaredFields = parameter.getClass().getDeclaredFields();
          for (Field field : declaredFields) {
              String name = field.getName();
              field.setAccessible(true);
              Object obj = field.get(parameter);
              field.setAccessible(false);
              fieldMap.put(name, obj);
          }
    
          for (int i = 1; i <= size; i++) {
              String parameterDefine = parameterMap.get(i);
              Object obj = fieldMap.get(parameterDefine);
    
              if (obj instanceof Short) {
                  preparedStatement.setShort(i, Short.parseShort(obj.toString()));
                  continue;
              }
    
              if (obj instanceof Integer) {
                  preparedStatement.setInt(i, Integer.parseInt(obj.toString()));
                  continue;
              }
    
              if (obj instanceof Long) {
                  preparedStatement.setLong(i, Long.parseLong(obj.toString()));
                  continue;
              }
    
              if (obj instanceof String) {
                  preparedStatement.setString(i, obj.toString());
                  continue;
              }
    
              if (obj instanceof Date) {
                  preparedStatement.setDate(i, (java.sql.Date) obj);
              }
    
          }
    
      }
    • 单个参数比较简单直接设置值即可,Long、Integer、String ...
    • 如果是一个类对象,需要通过获取 Field 属性,与参数 Map 进行匹配设置
  • 设置参数后执行查询 preparedStatement.executeQuery()
  • 接下来需要将查询结果转换为我们的类(主要是反射类的操作),resultSet2Obj(resultSet, Class.forName(xNode.getResultType()));

    private <T> List<T> resultSet2Obj(ResultSet resultSet, Class<?> clazz) {
        List<T> list = new ArrayList<>();
        try {
            ResultSetMetaData metaData = resultSet.getMetaData();
            int columnCount = metaData.getColumnCount();
            // 每次遍历行值
            while (resultSet.next()) {
                T obj = (T) clazz.newInstance();
                for (int i = 1; i <= columnCount; i++) {
                    Object value = resultSet.getObject(i);
                    String columnName = metaData.getColumnName(i);
                    String setMethod = "set" + columnName.substring(0, 1).toUpperCase() + columnName.substring(1);
                    Method method;
                    if (value instanceof Timestamp) {
                        method = clazz.getMethod(setMethod, Date.class);
                    } else {
                        method = clazz.getMethod(setMethod, value.getClass());
                    }
                    method.invoke(obj, value);
                }
                list.add(obj);
            }
        } catch (Exception e) {
            e.printStackTrace();
        }
        return list;
    }
    • 主要通过反射生成我们的类对象,这个类的类型定义在 sql 标签上
    • 时间类型需要判断后处理,Timestamp,与 java 不是一个类型

7. Sql查询补充说明

sql 查询有入参、有不需要入参、有查询一个、有查询集合,只需要合理包装即可,例如下面的查询集合,入参是对象类型;

ApiLikeTest.test_queryUserList()
@Test
public void test_queryUserList() {
    String resource = "spring/mybatis-config-datasource.xml";
    Reader reader;
    try {
        reader = Resources.getResourceAsReader(resource);
        SqlSessionFactory sqlMapper = new SqlSessionFactoryBuilder().build(reader);
        SqlSession session = sqlMapper.openSession();
        
        try {
            User req = new User();
            req.setAge(18);
            List<User> userList = session.selectList("org.itstack.demo.dao.IUserDao.queryUserList", req);
            System.out.println(JSON.toJSONString(userList));
        } finally {
            session.close();
            reader.close();
        }
    } catch (Exception e) {
        e.printStackTrace();
    }
    
}

*测试结果:*

[{"age":18,"createTime":1576944000000,"id":1,"name":"水水","updateTime":1576944000000},{"age":18,"createTime":1576944000000,"id":2,"name":"豆豆","updateTime":1576944000000}]

Process finished with exit code 0

四、源码分析(mybatis)

<dependency>
    <groupId>org.mybatis</groupId>
    <artifactId>mybatis</artifactId>
    <version>3.4.6</version>
</dependency>

Mybatis的整个源码还是很大的,以下主要将部分核心内容进行整理分析,以便于后续分析Mybatis与Spring整合的源码部分。简要包括;容器初始化、配置文件解析、Mapper加载与动态代理。

1. 从一个简单的案例开始

要学习Mybatis源码,最好的方式一定是从一个简单的点进入,而不是从Spring整合开始分析。SqlSessionFactory是整个Mybatis的核心实例对象,SqlSessionFactory对象的实例又通过SqlSessionFactoryBuilder对象来获得。SqlSessionFactoryBuilder对象可以从XML配置文件加载配置信息,然后创建SqlSessionFactory。如下例子:

MybatisApiTest.java
public class MybatisApiTest {

    @Test
    public void test_queryUserInfoById() {
        String resource = "spring/mybatis-config-datasource.xml";
        Reader reader;
        try {
            reader = Resources.getResourceAsReader(resource);
            SqlSessionFactory sqlMapper = new SqlSessionFactoryBuilder().build(reader);

            SqlSession session = sqlMapper.openSession();
            try {
                User user = session.selectOne("org.itstack.demo.dao.IUserDao.queryUserInfoById", 1L);
                System.out.println(JSON.toJSONString(user));
            } finally {
                session.close();
                reader.close();
            }
        } catch (IOException e) {
            e.printStackTrace();
        }
    }

}
dao/IUserDao.java
public interface IUserDao {

     User queryUserInfoById(Long id);

}
spring/mybatis-config-datasource.xml
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE configuration PUBLIC "-//mybatis.org//DTD Config 3.0//EN"
        "http://mybatis.org/dtd/mybatis-3-config.dtd">

<configuration>
    <environments default="development">
        <environment id="development">
            <transactionManager type="JDBC"/>
            <dataSource type="POOLED">
                <property name="driver" value="com.mysql.jdbc.Driver"/>
                <property name="url" value="jdbc:mysql://127.0.0.1:3306/itstack?useUnicode=true"/>
                <property name="username" value="root"/>
                <property name="password" value="123456"/>
            </dataSource>
        </environment>
    </environments>

    <mappers>
        <mapper resource="mapper/User_Mapper.xml"/>
    </mappers>

</configuration>

如果一切顺利,那么会有如下结果:

{"age":18,"createTime":1571376957000,"id":1,"name":"花花","updateTime":1571376957000}

从上面的代码块可以看到,核心代码;SqlSessionFactoryBuilder().build(reader),负责Mybatis配置文件的加载、解析、构建等职责,直到最终可以通过SqlSession来执行并返回结果。

2. 容器初始化

从上面代码可以看到,SqlSessionFactory是通过SqlSessionFactoryBuilder工厂类创建的,而不是直接使用构造器。容器的配置文件加载和初始化流程如下:

微信公众号:bugstack虫洞栈 & 初始化流程

  • 流程核心类

    • SqlSessionFactoryBuilder
    • XMLConfigBuilder
    • XPathParser
    • Configuration
SqlSessionFactoryBuilder.java
public class SqlSessionFactoryBuilder {

  public SqlSessionFactory build(Reader reader) {
    return build(reader, null, null);
  }

  public SqlSessionFactory build(Reader reader, String environment) {
    return build(reader, environment, null);
  }

  public SqlSessionFactory build(Reader reader, Properties properties) {
    return build(reader, null, properties);
  }

  public SqlSessionFactory build(Reader reader, String environment, Properties properties) {
    try {
      XMLConfigBuilder parser = new XMLConfigBuilder(reader, environment, properties);
      return build(parser.parse());
    } catch (Exception e) {
      throw ExceptionFactory.wrapException("Error building SqlSession.", e);
    } finally {
      ErrorContext.instance().reset();
      try {
        reader.close();
      } catch (IOException e) {
        // Intentionally ignore. Prefer previous error.
      }
    }
  }

  public SqlSessionFactory build(InputStream inputStream) {
    return build(inputStream, null, null);
  }

  public SqlSessionFactory build(InputStream inputStream, String environment) {
    return build(inputStream, environment, null);
  }

  public SqlSessionFactory build(InputStream inputStream, Properties properties) {
    return build(inputStream, null, properties);
  }

  public SqlSessionFactory build(InputStream inputStream, String environment, Properties properties) {
    try {
      XMLConfigBuilder parser = new XMLConfigBuilder(inputStream, environment, properties);
      return build(parser.parse());
    } catch (Exception e) {
      throw ExceptionFactory.wrapException("Error building SqlSession.", e);
    } finally {
      ErrorContext.instance().reset();
      try {
        inputStream.close();
      } catch (IOException e) {
        // Intentionally ignore. Prefer previous error.
      }
    }
  }
    
  public SqlSessionFactory build(Configuration config) {
    return new DefaultSqlSessionFactory(config);
  }

}

从上面的源码可以看到,SqlSessionFactory提供三种方式build构建对象;

  • 字节流:java.io.InputStream
  • 字符流:java.io.Reader
  • 配置类:org.apache.ibatis.session.Configuration

那么,字节流、字符流都会创建配置文件解析类:XMLConfigBuilder,并通过parser.parse()生成Configuration,最后调用配置类构建方法生成SqlSessionFactory。

XMLConfigBuilder.java
public class XMLConfigBuilder extends BaseBuilder {

  private boolean parsed;
  private final XPathParser parser;
  private String environment;
  private final ReflectorFactory localReflectorFactory = new DefaultReflectorFactory();

  ...
  public XMLConfigBuilder(Reader reader, String environment, Properties props) {
    this(new XPathParser(reader, true, props, new XMLMapperEntityResolver()), environment, props);
  }
  ...
}  
  1. XMLConfigBuilder对于XML文件的加载和解析都委托于XPathParser,最终使用JDK自带的javax.xml进行XML解析(XPath)
  2. XPathParser(Reader reader, boolean validation, Properties variables, EntityResolver entityResolver)

    1. reader:使用字符流创建新的输入源,用于对XML文件的读取
    2. validation:是否进行DTD校验
    3. variables:属性配置信息
    4. entityResolver:Mybatis硬编码了new XMLMapperEntityResolver()提供XML默认解析器
XMLMapperEntityResolver.java
public class XMLMapperEntityResolver implements EntityResolver {

  private static final String IBATIS_CONFIG_SYSTEM = "ibatis-3-config.dtd";
  private static final String IBATIS_MAPPER_SYSTEM = "ibatis-3-mapper.dtd";
  private static final String MYBATIS_CONFIG_SYSTEM = "mybatis-3-config.dtd";
  private static final String MYBATIS_MAPPER_SYSTEM = "mybatis-3-mapper.dtd";

  private static final String MYBATIS_CONFIG_DTD = "org/apache/ibatis/builder/xml/mybatis-3-config.dtd";
  private static final String MYBATIS_MAPPER_DTD = "org/apache/ibatis/builder/xml/mybatis-3-mapper.dtd";

  /*
   * Converts a public DTD into a local one
   * 
   * @param publicId The public id that is what comes after "PUBLIC"
   * @param systemId The system id that is what comes after the public id.
   * @return The InputSource for the DTD
   * 
   * @throws org.xml.sax.SAXException If anything goes wrong
   */
  @Override
  public InputSource resolveEntity(String publicId, String systemId) throws SAXException {
    try {
      if (systemId != null) {
        String lowerCaseSystemId = systemId.toLowerCase(Locale.ENGLISH);
        if (lowerCaseSystemId.contains(MYBATIS_CONFIG_SYSTEM) || lowerCaseSystemId.contains(IBATIS_CONFIG_SYSTEM)) {
          return getInputSource(MYBATIS_CONFIG_DTD, publicId, systemId);
        } else if (lowerCaseSystemId.contains(MYBATIS_MAPPER_SYSTEM) || lowerCaseSystemId.contains(IBATIS_MAPPER_SYSTEM)) {
          return getInputSource(MYBATIS_MAPPER_DTD, publicId, systemId);
        }
      }
      return null;
    } catch (Exception e) {
      throw new SAXException(e.toString());
    }
  }

  private InputSource getInputSource(String path, String publicId, String systemId) {
    InputSource source = null;
    if (path != null) {
      try {
        InputStream in = Resources.getResourceAsStream(path);
        source = new InputSource(in);
        source.setPublicId(publicId);
        source.setSystemId(systemId);        
      } catch (IOException e) {
        // ignore, null is ok
      }
    }
    return source;
  }

}
  1. Mybatis依赖于dtd文件进行进行解析,其中的ibatis-3-config.dtd主要是用于兼容用途
  2. getInputSource(String path, String publicId, String systemId)的调用里面有两个参数publicId(公共标识符)和systemId(系统标示符)
XPathParser.java
public XPathParser(Reader reader, boolean validation, Properties variables, EntityResolver entityResolver) {
  commonConstructor(validation, variables, entityResolver);
  this.document = createDocument(new InputSource(reader));
}

private void commonConstructor(boolean validation, Properties variables, EntityResolver entityResolver) {
  this.validation = validation;
  this.entityResolver = entityResolver;
  this.variables = variables;
  XPathFactory factory = XPathFactory.newInstance();
  this.xpath = factory.newXPath();
}

private Document createDocument(InputSource inputSource) {
  // important: this must only be called AFTER common constructor
  try {
    DocumentBuilderFactory factory = DocumentBuilderFactory.newInstance();
    factory.setValidating(validation);
    factory.setNamespaceAware(false);
    factory.setIgnoringComments(true);
    factory.setIgnoringElementContentWhitespace(false);
    factory.setCoalescing(false);
    factory.setExpandEntityReferences(true);
    DocumentBuilder builder = factory.newDocumentBuilder();
    builder.setEntityResolver(entityResolver);
    builder.setErrorHandler(new ErrorHandler() {
      @Override
      public void error(SAXParseException exception) throws SAXException {
        throw exception;
      }
      @Override
      public void fatalError(SAXParseException exception) throws SAXException {
        throw exception;
      }
      @Override
      public void warning(SAXParseException exception) throws SAXException {
      }
    });
    return builder.parse(inputSource);
  } catch (Exception e) {
    throw new BuilderException("Error creating document instance.  Cause: " + e, e);
  }
  
}    
  1. 从上到下可以看到主要是为了创建一个Mybatis的文档解析器,最后根据builder.parse(inputSource)返回Document
  2. 得到XPathParser实例后,接下来在调用方法:this(new XPathParser(reader, true, props, new XMLMapperEntityResolver()), environment, props);

     XMLConfigBuilder.this(new XPathParser(reader, true, props, new XMLMapperEntityResolver()), environment, props);
    
     private XMLConfigBuilder(XPathParser parser, String environment, Properties props) {
       super(new Configuration());
       ErrorContext.instance().resource("SQL Mapper Configuration");
       this.configuration.setVariables(props);
       this.parsed = false;
       this.environment = environment;
       this.parser = parser;
     }
  3. 其中调用了父类的构造函数

    public abstract class BaseBuilder {
      protected final Configuration configuration;
      protected final TypeAliasRegistry typeAliasRegistry;
      protected final TypeHandlerRegistry typeHandlerRegistry;
    
      public BaseBuilder(Configuration configuration) {
        this.configuration = configuration;
        this.typeAliasRegistry = this.configuration.getTypeAliasRegistry();
        this.typeHandlerRegistry = this.configuration.getTypeHandlerRegistry();
      }
    }
  4. XMLConfigBuilder创建完成后,sqlSessionFactoryBuild调用parser.parse()创建Configuration

    public class XMLConfigBuilder extends BaseBuilder {    
         public Configuration parse() {
           if (parsed) {
             throw new BuilderException("Each XMLConfigBuilder can only be used once.");
           }
           parsed = true;
           parseConfiguration(parser.evalNode("/configuration"));
           return configuration;
         }
    }

3. 配置文件解析

这一部分是整个XML文件解析和装载的核心内容,其中包括;

  1. 属性解析propertiesElement
  2. 加载settings节点settingsAsProperties
  3. 载自定义VFS loadCustomVfs
  4. 解析类型别名typeAliasesElement
  5. 加载插件pluginElement
  6. 加载对象工厂objectFactoryElement
  7. 创建对象包装器工厂objectWrapperFactoryElement
  8. 加载反射工厂reflectorFactoryElement
  9. 元素设置settingsElement
  10. 加载环境配置environmentsElement
  11. 数据库厂商标识加载databaseIdProviderElement
  12. 加载类型处理器typeHandlerElement
  13. (核心)加载mapper文件mapperElement
parseConfiguration(parser.evalNode("/configuration"));

private void parseConfiguration(XNode root) {
    try {
      //issue #117 read properties first
      //属性解析propertiesElement
      propertiesElement(root.evalNode("properties"));
      //加载settings节点settingsAsProperties
      Properties settings = settingsAsProperties(root.evalNode("settings"));
      //加载自定义VFS loadCustomVfs
      loadCustomVfs(settings);
      //解析类型别名typeAliasesElement
      typeAliasesElement(root.evalNode("typeAliases"));
      //加载插件pluginElement
      pluginElement(root.evalNode("plugins"));
      //加载对象工厂objectFactoryElement
      objectFactoryElement(root.evalNode("objectFactory"));
      //创建对象包装器工厂objectWrapperFactoryElement
      objectWrapperFactoryElement(root.evalNode("objectWrapperFactory"));
      //加载反射工厂reflectorFactoryElement
      reflectorFactoryElement(root.evalNode("reflectorFactory"));
      //元素设置
      settingsElement(settings);
      // read it after objectFactory and objectWrapperFactory issue #631
      //加载环境配置environmentsElement
      environmentsElement(root.evalNode("environments"));
      //数据库厂商标识加载databaseIdProviderElement
      databaseIdProviderElement(root.evalNode("databaseIdProvider"));
      //加载类型处理器typeHandlerElement
      typeHandlerElement(root.evalNode("typeHandlers"));
      //加载mapper文件mapperElement
      mapperElement(root.evalNode("mappers"));
    } catch (Exception e) {
      throw new BuilderException("Error parsing SQL Mapper Configuration. Cause: " + e, e);
    }
} 

所有的root.evalNode()底层都是调用XML DOM方法:Object evaluate(String expression, Object item, QName returnType),表达式参数expression,通过XObject resultObject = eval( expression, item )返回最终节点内容,可以参考http://mybatis.org/dtd/mybati...,如下;

<!ELEMENT configuration (properties?, settings?, typeAliases?, typeHandlers?, objectFactory?, objectWrapperFactory?, reflectorFactory?, plugins?, environments?, databaseIdProvider?, mappers?)>
 
<!ELEMENT databaseIdProvider (property*)>
<!ATTLIST databaseIdProvider
type CDATA #REQUIRED
>
 
<!ELEMENT properties (property*)>
<!ATTLIST properties
resource CDATA #IMPLIED
url CDATA #IMPLIED
>
 
<!ELEMENT property EMPTY>
<!ATTLIST property
name CDATA #REQUIRED
value CDATA #REQUIRED
>
 
<!ELEMENT settings (setting+)>
 
<!ELEMENT setting EMPTY>
<!ATTLIST setting
name CDATA #REQUIRED
value CDATA #REQUIRED
>
 
<!ELEMENT typeAliases (typeAlias*,package*)>
 
<!ELEMENT typeAlias EMPTY>
<!ATTLIST typeAlias
type CDATA #REQUIRED
alias CDATA #IMPLIED
>
 
<!ELEMENT typeHandlers (typeHandler*,package*)>
 
<!ELEMENT typeHandler EMPTY>
<!ATTLIST typeHandler
javaType CDATA #IMPLIED
jdbcType CDATA #IMPLIED
handler CDATA #REQUIRED
>
 
<!ELEMENT objectFactory (property*)>
<!ATTLIST objectFactory
type CDATA #REQUIRED
>
 
<!ELEMENT objectWrapperFactory EMPTY>
<!ATTLIST objectWrapperFactory
type CDATA #REQUIRED
>
 
<!ELEMENT reflectorFactory EMPTY>
<!ATTLIST reflectorFactory
type CDATA #REQUIRED
>
 
<!ELEMENT plugins (plugin+)>
 
<!ELEMENT plugin (property*)>
<!ATTLIST plugin
interceptor CDATA #REQUIRED
>
 
<!ELEMENT environments (environment+)>
<!ATTLIST environments
default CDATA #REQUIRED
>
 
<!ELEMENT environment (transactionManager,dataSource)>
<!ATTLIST environment
id CDATA #REQUIRED
>
 
<!ELEMENT transactionManager (property*)>
<!ATTLIST transactionManager
type CDATA #REQUIRED
>
 
<!ELEMENT dataSource (property*)>
<!ATTLIST dataSource
type CDATA #REQUIRED
>
 
<!ELEMENT mappers (mapper*,package*)>
 
<!ELEMENT mapper EMPTY>
<!ATTLIST mapper
resource CDATA #IMPLIED
url CDATA #IMPLIED
class CDATA #IMPLIED
>
 
<!ELEMENT package EMPTY>
<!ATTLIST package
name CDATA #REQUIRED
>

mybatis-3-config.dtd 定义文件中有11个配置文件,如下;

  1. properties?,
  2. settings?,
  3. typeAliases?,
  4. typeHandlers?,
  5. objectFactory?,
  6. objectWrapperFactory?,
  7. reflectorFactory?,
  8. plugins?,
  9. environments?,
  10. databaseIdProvider?,
  11. mappers?

以上每个配置都是可选。最终配置内容会保存到org.apache.ibatis.session.Configuration,如下;

public class Configuration {

  protected Environment environment;
  // 允许在嵌套语句中使用分页(RowBounds)。如果允许使用则设置为false。默认为false
  protected boolean safeRowBoundsEnabled;
  // 允许在嵌套语句中使用分页(ResultHandler)。如果允许使用则设置为false。
  protected boolean safeResultHandlerEnabled = true;
  // 是否开启自动驼峰命名规则(camel case)映射,即从经典数据库列名 A_COLUMN 到经典 Java 属性名 aColumn 的类似映射。默认false
  protected boolean mapUnderscoreToCamelCase;
  // 当开启时,任何方法的调用都会加载该对象的所有属性。否则,每个属性会按需加载。默认值false (true in ≤3.4.1)
  protected boolean aggressiveLazyLoading;
  // 是否允许单一语句返回多结果集(需要兼容驱动)。
  protected boolean multipleResultSetsEnabled = true;
  // 允许 JDBC 支持自动生成主键,需要驱动兼容。这就是insert时获取mysql自增主键/oracle sequence的开关。注:一般来说,这是希望的结果,应该默认值为true比较合适。
  protected boolean useGeneratedKeys;
  // 使用列标签代替列名,一般来说,这是希望的结果
  protected boolean useColumnLabel = true;
  // 是否启用缓存 {默认是开启的,可能这也是你的面试题}
  protected boolean cacheEnabled = true;
  // 指定当结果集中值为 null 的时候是否调用映射对象的 setter(map 对象时为 put)方法,这对于有 Map.keySet() 依赖或 null 值初始化的时候是有用的。
  protected boolean callSettersOnNulls;
  // 允许使用方法签名中的名称作为语句参数名称。 为了使用该特性,你的工程必须采用Java 8编译,并且加上-parameters选项。(从3.4.1开始)
  protected boolean useActualParamName = true;
  //当返回行的所有列都是空时,MyBatis默认返回null。 当开启这个设置时,MyBatis会返回一个空实例。 请注意,它也适用于嵌套的结果集 (i.e. collectioin and association)。(从3.4.2开始) 注:这里应该拆分为两个参数比较合适, 一个用于结果集,一个用于单记录。通常来说,我们会希望结果集不是null,单记录仍然是null
  protected boolean returnInstanceForEmptyRow;
  // 指定 MyBatis 增加到日志名称的前缀。
  protected String logPrefix;
  // 指定 MyBatis 所用日志的具体实现,未指定时将自动查找。一般建议指定为slf4j或log4j
  protected Class <? extends Log> logImpl;
   // 指定VFS的实现, VFS是mybatis提供的用于访问AS内资源的一个简便接口
  protected Class <? extends VFS> vfsImpl;
  // MyBatis 利用本地缓存机制(Local Cache)防止循环引用(circular references)和加速重复嵌套查询。 默认值为 SESSION,这种情况下会缓存一个会话中执行的所有查询。 若设置值为 STATEMENT,本地会话仅用在语句执行上,对相同 SqlSession 的不同调用将不会共享数据。
  protected LocalCacheScope localCacheScope = LocalCacheScope.SESSION;
  // 当没有为参数提供特定的 JDBC 类型时,为空值指定 JDBC 类型。 某些驱动需要指定列的 JDBC 类型,多数情况直接用一般类型即可,比如 NULL、VARCHAR 或 OTHER。
  protected JdbcType jdbcTypeForNull = JdbcType.OTHER;
  // 指定对象的哪个方法触发一次延迟加载。
  protected Set<String> lazyLoadTriggerMethods = new HashSet<String>(Arrays.asList(new String[] { "equals", "clone", "hashCode", "toString" }));
  // 设置超时时间,它决定驱动等待数据库响应的秒数。默认不超时
  protected Integer defaultStatementTimeout;
  // 为驱动的结果集设置默认获取数量。
  protected Integer defaultFetchSize;
  // SIMPLE 就是普通的执行器;REUSE 执行器会重用预处理语句(prepared statements); BATCH 执行器将重用语句并执行批量更新。
  protected ExecutorType defaultExecutorType = ExecutorType.SIMPLE;
  // 指定 MyBatis 应如何自动映射列到字段或属性。 NONE 表示取消自动映射;PARTIAL 只会自动映射没有定义嵌套结果集映射的结果集。 FULL 会自动映射任意复杂的结果集(无论是否嵌套)。
  protected AutoMappingBehavior autoMappingBehavior = AutoMappingBehavior.PARTIAL;
  // 指定发现自动映射目标未知列(或者未知属性类型)的行为。这个值应该设置为WARNING比较合适
  protected AutoMappingUnknownColumnBehavior autoMappingUnknownColumnBehavior = AutoMappingUnknownColumnBehavior.NONE;
  // settings下的properties属性
  protected Properties variables = new Properties();
  // 默认的反射器工厂,用于操作属性、构造器方便
  protected ReflectorFactory reflectorFactory = new DefaultReflectorFactory();
  // 对象工厂, 所有的类resultMap类都需要依赖于对象工厂来实例化
  protected ObjectFactory objectFactory = new DefaultObjectFactory();
  // 对象包装器工厂,主要用来在创建非原生对象,比如增加了某些监控或者特殊属性的代理类
  protected ObjectWrapperFactory objectWrapperFactory = new DefaultObjectWrapperFactory();
  // 延迟加载的全局开关。当开启时,所有关联对象都会延迟加载。特定关联关系中可通过设置fetchType属性来覆盖该项的开关状态。
  protected boolean lazyLoadingEnabled = false;
  // 指定 Mybatis 创建具有延迟加载能力的对象所用到的代理工具。MyBatis 3.3+使用JAVASSIST
  protected ProxyFactory proxyFactory = new JavassistProxyFactory(); // #224 Using internal Javassist instead of OGNL
  // MyBatis 可以根据不同的数据库厂商执行不同的语句,这种多厂商的支持是基于映射语句中的 databaseId 属性。
  protected String databaseId;
  ...
}

以上可以看到,Mybatis把所有的配置;resultMap、Sql语句、插件、缓存等都维护在Configuration中。这里还有一个小技巧,在Configuration还有一个StrictMap内部类,它继承于HashMap完善了put时防重、get时取不到值的异常处理,如下;

protected static class StrictMap<V> extends HashMap<String, V> {

    private static final long serialVersionUID = -4950446264854982944L;
    private final String name;

    public StrictMap(String name, int initialCapacity, float loadFactor) {
      super(initialCapacity, loadFactor);
      this.name = name;
    }

    public StrictMap(String name, int initialCapacity) {
      super(initialCapacity);
      this.name = name;
    }

    public StrictMap(String name) {
      super();
      this.name = name;
    }

    public StrictMap(String name, Map<String, ? extends V> m) {
      super(m);
      this.name = name;
    }
}    

(核心)加载mapper文件mapperElement

Mapper文件处理是Mybatis框架的核心服务,所有的SQL语句都编写在Mapper中,这块也是我们分析的重点,其他模块可以后续讲解。

XMLConfigBuilder.parseConfiguration()->mapperElement(root.evalNode("mappers"));
private void mapperElement(XNode parent) throws Exception {
   if (parent != null) {
     for (XNode child : parent.getChildren()) {
       // 如果要同时使用package自动扫描和通过mapper明确指定要加载的mapper,一定要确保package自动扫描的范围不包含明确指定的mapper,否则在通过package扫描的interface的时候,尝试加载对应xml文件的loadXmlResource()的逻辑中出现判重出错,报org.apache.ibatis.binding.BindingException异常,即使xml文件中包含的内容和mapper接口中包含的语句不重复也会出错,包括加载mapper接口时自动加载的xml mapper也一样会出错。
       if ("package".equals(child.getName())) {
         String mapperPackage = child.getStringAttribute("name");
         configuration.addMappers(mapperPackage);
       } else {
         String resource = child.getStringAttribute("resource");
         String url = child.getStringAttribute("url");
         String mapperClass = child.getStringAttribute("class");
         if (resource != null && url == null && mapperClass == null) {
           ErrorContext.instance().resource(resource);
           InputStream inputStream = Resources.getResourceAsStream(resource);
           XMLMapperBuilder mapperParser = new XMLMapperBuilder(inputStream, configuration, resource, configuration.getSqlFragments());
           mapperParser.parse();
         } else if (resource == null && url != null && mapperClass == null) {
           ErrorContext.instance().resource(url);
           InputStream inputStream = Resources.getUrlAsStream(url);
           XMLMapperBuilder mapperParser = new XMLMapperBuilder(inputStream, configuration, url, configuration.getSqlFragments());
           mapperParser.parse();
         } else if (resource == null && url == null && mapperClass != null) {
           Class<?> mapperInterface = Resources.classForName(mapperClass);
           configuration.addMapper(mapperInterface);
         } else {
           throw new BuilderException("A mapper element may only specify a url, resource or class, but not more than one.");
         }
       }
     }
   }
}
  • Mybatis提供了两类配置Mapper的方法,第一类是使用package自动搜索的模式,这样指定package下所有接口都会被注册为mapper,也是在Spring中比较常用的方式,例如:

    <mappers>
      <package name="org.itstack.demo"/>
    </mappers>
  • 另外一类是明确指定Mapper,这又可以通过resource、url或者class进行细分,例如;

    <mappers>
        <mapper resource="mapper/User_Mapper.xml"/>
        <mapper class=""/>
        <mapper url=""/>
    </mappers>

4. Mapper加载与动态代理

通过package方式自动搜索加载,生成对应的mapper代理类,代码块和流程,如下;

private void mapperElement(XNode parent) throws Exception {
  if (parent != null) {
    for (XNode child : parent.getChildren()) {
      if ("package".equals(child.getName())) {
        String mapperPackage = child.getStringAttribute("name");
        configuration.addMappers(mapperPackage);
      } else {
        ...
      }
    }
  }
}

微信公众号:bugstack虫洞栈 & 动态代理过程

Mapper加载到生成代理对象的流程中,主要的核心类包括;

  1. XMLConfigBuilder
  2. Configuration
  3. MapperRegistry
  4. MapperAnnotationBuilder
  5. MapperProxyFactory
MapperRegistry.java

解析加载Mapper

public void addMappers(String packageName, Class<?> superType) {
  // mybatis框架提供的搜索classpath下指定package以及子package中符合条件(注解或者继承于某个类/接口)的类,默认使用Thread.currentThread().getContextClassLoader()返回的加载器,和spring的工具类殊途同归。
  ResolverUtil<Class<?>> resolverUtil = new ResolverUtil<Class<?>>();   
  // 无条件的加载所有的类,因为调用方传递了Object.class作为父类,这也给以后的指定mapper接口预留了余地
  resolverUtil.find(new ResolverUtil.IsA(superType), packageName); 
 // 所有匹配的calss都被存储在ResolverUtil.matches字段中
  Set<Class<? extends Class<?>>> mapperSet = resolverUtil.getClasses();
  for (Class<?> mapperClass : mapperSet) {   
    //调用addMapper方法进行具体的mapper类/接口解析
    addMapper(mapperClass);
  }
}

生成代理类:MapperProxyFactory

public <T> void addMapper(Class<T> type) {    
  // 对于mybatis mapper接口文件,必须是interface,不能是class
  if (type.isInterface()) {
    if (hasMapper(type)) {
      throw new BindingException("Type " + type + " is already known to the MapperRegistry.");
    }
    boolean loadCompleted = false;
    try {      
      // 为mapper接口创建一个MapperProxyFactory代理
      knownMappers.put(type, new MapperProxyFactory<T>(type));
      // It's important that the type is added before the parser is run
      // otherwise the binding may automatically be attempted by the
      // mapper parser. If the type is already known, it won't try.
      MapperAnnotationBuilder parser = new MapperAnnotationBuilder(config, type);
      parser.parse();
      loadCompleted = true;
    } finally {
      if (!loadCompleted) {
        knownMappers.remove(type);
      }
    }
  }
}

在MapperRegistry中维护了接口类与代理工程的映射关系,knownMappers;

private final Map<Class<?>, MapperProxyFactory<?>> knownMappers = new HashMap<Class<?>, MapperProxyFactory<?>>();
MapperProxyFactory.java
public class MapperProxyFactory<T> {
  private final Class<T> mapperInterface;
  private final Map<Method, MapperMethod> methodCache = new ConcurrentHashMap<Method, MapperMethod>();
  public MapperProxyFactory(Class<T> mapperInterface) {
    this.mapperInterface = mapperInterface;
  }
  public Class<T> getMapperInterface() {
    return mapperInterface;
  }
  public Map<Method, MapperMethod> getMethodCache() {
    return methodCache;
  }
  @SuppressWarnings("unchecked")
  protected T newInstance(MapperProxy<T> mapperProxy) {
    return (T) Proxy.newProxyInstance(mapperInterface.getClassLoader(), new Class[] { mapperInterface }, mapperProxy);
  }
  public T newInstance(SqlSession sqlSession) {
    final MapperProxy<T> mapperProxy = new MapperProxy<T>(sqlSession, mapperInterface, methodCache);
    return newInstance(mapperProxy);
  }
}

如上是Mapper的代理类工程,构造函数中的mapperInterface就是对应的接口类,当实例化时候会获得具体的MapperProxy代理,里面主要包含了SqlSession。

五、源码分析(mybatis-spring)

<dependency>
    <groupId>org.mybatis</groupId>
    <artifactId>mybatis-spring</artifactId>
    <version>1.3.2</version>
</dependency>

作为一款好用的ORM框架,一定是萝莉脸(单纯)、御姐心(强大),铺的了床(屏蔽与JDBC直接打交道)、暖的了房(速度性能好)!鉴于这些优点几乎在国内互联网大部分开发框架都会使用到Mybatis,尤其在一些需要高性能的场景下需要优化sql那么一定需要手写sql在xml中。那么,准备好了吗!开始分析分析它的源码;

1. 从一个简单的案例开始

与分析mybatis源码一样,先做一个简单的案例;定义dao、编写配置文件、junit单元测试;

SpringApiTest.java
@RunWith(SpringJUnit4ClassRunner.class)
@ContextConfiguration("classpath:spring-config.xml")
public class SpringApiTest {

    private Logger logger = LoggerFactory.getLogger(SpringApiTest.class);

    @Resource
    private ISchoolDao schoolDao;
    @Resource
    private IUserDao userDao;

    @Test
    public void test_queryRuleTreeByTreeId(){
        School ruleTree = schoolDao.querySchoolInfoById(1L);
        logger.info(JSON.toJSONString(ruleTree));

        User user = userDao.queryUserInfoById(1L);
        logger.info(JSON.toJSONString(user));
    }

}
spring-config-datasource.xml
<?xml version="1.0" encoding="UTF-8"?>
<beans xmlns="http://www.springframework.org/schema/beans"
       xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
       xsi:schemaLocation="http://www.springframework.org/schema/beans
        http://www.springframework.org/schema/beans/spring-beans.xsd">

    <!-- 1.数据库连接池: DriverManagerDataSource 也可以使用DBCP2-->
    <bean id="dataSource" class="org.springframework.jdbc.datasource.DriverManagerDataSource">
        <property name="driverClassName" value="${db.jdbc.driverClassName}"/>
        <property name="url" value="${db.jdbc.url}"/>
        <property name="username" value="${db.jdbc.username}"/>
        <property name="password" value="${db.jdbc.password}"/>
    </bean>

    <!-- 2.配置SqlSessionFactory对象 -->
    <bean id="sqlSessionFactory" class="org.mybatis.spring.SqlSessionFactoryBean">
        <!-- 注入数据库连接池 -->
        <property name="dataSource" ref="dataSource"/>
        <!-- 配置MyBaties全局配置文件:mybatis-config.xml -->
        <property name="configLocation" value="classpath:mybatis-config.xml"/>
        <!-- 扫描entity包 使用别名 -->
        <property name="typeAliasesPackage" value="org.itstack.demo.po"/>
        <!-- 扫描sql配置文件:mapper需要的xml文件 -->
        <property name="mapperLocations" value="classpath:mapper/*.xml"/>
    </bean>

    <!-- 3.配置扫描Dao接口包,动态实现Dao接口,注入到spring容器中 -->
    <bean class="org.mybatis.spring.mapper.MapperScannerConfigurer">
        <!-- 注入sqlSessionFactory -->
        <property name="sqlSessionFactoryBeanName" value="sqlSessionFactory"/>
        <!-- 给出需要扫描Dao接口包,多个逗号隔开 -->
        <property name="basePackage" value="org.itstack.demo.dao"/>
    </bean>
              
</beans>

如果一切顺利,那么会有如下结果:

{"address":"北京市海淀区颐和园路5号","createTime":1571376957000,"id":1,"name":"北京大学","updateTime":1571376957000}
{"age":18,"createTime":1571376957000,"id":1,"name":"花花","updateTime":1571376957000}

从上面单元测试的代码可以看到,两个没有方法体的注解就这么神奇的执行了我们的xml中的配置语句并输出了结果。其实主要得益于以下两个类;

  • org.mybatis.spring.SqlSessionFactoryBean
  • org.mybatis.spring.mapper.MapperScannerConfigurer

2. 扫描装配注册(MapperScannerConfigurer)

MapperScannerConfigurer为整个Dao接口层生成动态代理类注册,启动到了核心作用。这个类实现了如下接口,用来对扫描的Mapper进行处理:

  • BeanDefinitionRegistryPostProcessor
  • InitializingBean
  • ApplicationContextAware
  • BeanNameAware

整体类图如下;

微信公众号:bugstack虫洞栈 & MapperScannerConfigurer类图

执行流程如下;

微信公众号:bugstack虫洞栈 & 执行流程图

上面的类图+流程图,其实已经很清楚的描述了MapperScannerConfigurer初始化过程,但对于头一次看的新人来说依旧是我太难了,好继续!

MapperScannerConfigurer.java & 部分截取
@Override
public void postProcessBeanDefinitionRegistry(BeanDefinitionRegistry registry) {
  if (this.processPropertyPlaceHolders) {
    processPropertyPlaceHolders();
  }
  ClassPathMapperScanner scanner = new ClassPathMapperScanner(registry);
  scanner.setAddToConfig(this.addToConfig);
  scanner.setAnnotationClass(this.annotationClass);
  scanner.setMarkerInterface(this.markerInterface);
  scanner.setSqlSessionFactory(this.sqlSessionFactory);
  scanner.setSqlSessionTemplate(this.sqlSessionTemplate);
  scanner.setSqlSessionFactoryBeanName(this.sqlSessionFactoryBeanName);
  scanner.setSqlSessionTemplateBeanName(this.sqlSessionTemplateBeanName);
  scanner.setResourceLoader(this.applicationContext);
  scanner.setBeanNameGenerator(this.nameGenerator);
  scanner.registerFilters();
  scanner.scan(StringUtils.tokenizeToStringArray(this.basePackage, ConfigurableApplicationContext.CONFIG_LOCATION_DELIMITERS));
}
  • 实现了BeanDefinitionRegistryPostProcessor.postProcessBeanDefinitionRegistry用于注册Bean到Spring容器中
  • 306行:new ClassPathMapperScanner(registry); 硬编码类路径扫描器,用于解析Mybatis的Mapper文件
  • 317行:scanner.scan 对Mapper进行扫描。这里包含了一个继承类实现关系的调用,也就是本文开头的测试题。
ClassPathMapperScanner.java & 部分截取
@Override
public Set<BeanDefinitionHolder> doScan(String... basePackages) {
  Set<BeanDefinitionHolder> beanDefinitions = super.doScan(basePackages);
  if (beanDefinitions.isEmpty()) {
    logger.warn("No MyBatis mapper was found in '" + Arrays.toString(basePackages) + "' package. Please check your configuration.");
  } else {
    processBeanDefinitions(beanDefinitions);
  }
  return beanDefinitions;
}
  • 优先调用父类的super.doScan(basePackages);进行注册Bean信息
ClassPathBeanDefinitionScanner.java & 部分截取
protected Set<BeanDefinitionHolder> doScan(String... basePackages) {
    Assert.notEmpty(basePackages, "At least one base package must be specified");
    Set<BeanDefinitionHolder> beanDefinitions = new LinkedHashSet<BeanDefinitionHolder>();
    for (String basePackage : basePackages) {
        Set<BeanDefinition> candidates = findCandidateComponents(basePackage);
        for (BeanDefinition candidate : candidates) {
            ScopeMetadata scopeMetadata = this.scopeMetadataResolver.resolveScopeMetadata(candidate);
            candidate.setScope(scopeMetadata.getScopeName());
            String beanName = this.beanNameGenerator.generateBeanName(candidate, this.registry);
            if (candidate instanceof AbstractBeanDefinition) {
                postProcessBeanDefinition((AbstractBeanDefinition) candidate, beanName);
            }
            if (candidate instanceof AnnotatedBeanDefinition) {
                AnnotationConfigUtils.processCommonDefinitionAnnotations((AnnotatedBeanDefinition) candidate)
            }
            if (checkCandidate(beanName, candidate)) {
                BeanDefinitionHolder definitionHolder = new BeanDefinitionHolder(candidate, beanName);
                definitionHolder =
                        AnnotationConfigUtils.applyScopedProxyMode(scopeMetadata, definitionHolder, this.regi
                beanDefinitions.add(definitionHolder);
                registerBeanDefinition(definitionHolder, this.registry);
            }
        }
    }
    return beanDefinitions;
}
  • 优先调用了父类的doScan方法,用于Mapper扫描和Bean的定义以及注册到DefaultListableBeanFactory。{DefaultListableBeanFactory是Spring中IOC容器的始祖,所有需要实例化的类都需要注册进来,之后在初始化}
  • 272行:findCandidateComponents(basePackage),扫描package包路径,对于注解类的有另外的方式,大同小异
  • 288行:registerBeanDefinition(definitionHolder, this.registry);注册Bean信息的过程,最终会调用到:org.springframework.beans.factory.support.DefaultListableBeanFactory
ClassPathMapperScanner.java & 部分截取
**processBeanDefinitions(beanDefinitions);**

private void processBeanDefinitions(Set<BeanDefinitionHolder> beanDefinitions) {
  GenericBeanDefinition definition;
  for (BeanDefinitionHolder holder : beanDefinitions) {
    definition = (GenericBeanDefinition) holder.getBeanDefinition();
    if (logger.isDebugEnabled()) {
      logger.debug("Creating MapperFactoryBean with name '" + holder.getBeanName() 
        + "' and '" + definition.getBeanClassName() + "' mapperInterface");
    }
    // the mapper interface is the original class of the bean
    // but, the actual class of the bean is MapperFactoryBean
    definition.getConstructorArgumentValues().addGenericArgumentValue(definition.getBeanClassName()); // issue #59
    definition.setBeanClass(this.mapperFactoryBean.getClass());
    definition.getPropertyValues().add("addToConfig", this.addToConfig);
    boolean explicitFactoryUsed = false;
    if (StringUtils.hasText(this.sqlSessionFactoryBeanName)) {
      definition.getPropertyValues().add("sqlSessionFactory", new RuntimeBeanReference(this.sqlSessionFactoryBeanName));
      explicitFactoryUsed = true;
    } else if (this.sqlSessionFactory != null) {
      definition.getPropertyValues().add("sqlSessionFactory", this.sqlSessionFactory);
      explicitFactoryUsed = true;
    }
    if (StringUtils.hasText(this.sqlSessionTemplateBeanName)) {
      if (explicitFactoryUsed) {
        logger.warn("Cannot use both: sqlSessionTemplate and sqlSessionFactory together. sqlSessionFactory is ignored.");
      }
      definition.getPropertyValues().add("sqlSessionTemplate", new RuntimeBeanReference(this.sqlSessionTemplateBeanName));
      explicitFactoryUsed = true;
    } else if (this.sqlSessionTemplate != null) {
      if (explicitFactoryUsed) {
        logger.warn("Cannot use both: sqlSessionTemplate and sqlSessionFactory together. sqlSessionFactory is ignored.");
      }
      definition.getPropertyValues().add("sqlSessionTemplate", this.sqlSessionTemplate);
      explicitFactoryUsed = true;
    }
    if (!explicitFactoryUsed) {
      if (logger.isDebugEnabled()) {
        logger.debug("Enabling autowire by type for MapperFactoryBean with name '" + holder.getBeanName() + "'.");
      }
      definition.setAutowireMode(AbstractBeanDefinition.AUTOWIRE_BY_TYPE);
    }
  }
}
  • 163行:super.doScan(basePackages);,调用完父类方法后开始执行内部方法:processBeanDefinitions(beanDefinitions)
  • 186行:definition.getConstructorArgumentValues().addGenericArgumentValue(definition.getBeanClassName()); 设置BeanName参数,也就是我们的:ISchoolDao、IUserDao
  • 187行:definition.setBeanClass(this.mapperFactoryBean.getClass());,设置BeanClass,接口本身是没有类的,那么这里将MapperFactoryBean类设置进来,最终所有的dao层接口类都是这个MapperFactoryBean
MapperFactoryBean.java & 部分截取

这个类有继承也有接口实现,最好先了解下整体类图,如下;

微信公众号:bugstack虫洞栈 & MapperFactoryBean类图

这个类就非常重要了,最终所有的sql信息执行都会通过这个类获取getObject(),也就是SqlSession获取mapper的代理类:MapperProxyFactory->MapperProxy

public class MapperFactoryBean<T> extends SqlSessionDaoSupport implements FactoryBean<T> {

  private Class<T> mapperInterface;

  private boolean addToConfig = true;

  public MapperFactoryBean() {
    //intentionally empty 
  }
  
  public MapperFactoryBean(Class<T> mapperInterface) {
    this.mapperInterface = mapperInterface;
  }

  /**  
   * 当SpringBean容器初始化时候会调用到checkDaoConfig(),他是继承类中的抽象方法
   * {@inheritDoc}
   */
  @Override
  protected void checkDaoConfig() {
    super.checkDaoConfig();

    notNull(this.mapperInterface, "Property 'mapperInterface' is required");

    Configuration configuration = getSqlSession().getConfiguration();
    if (this.addToConfig && !configuration.hasMapper(this.mapperInterface)) {
      try {
        configuration.addMapper(this.mapperInterface);
      } catch (Exception e) {
        logger.error("Error while adding the mapper '" + this.mapperInterface + "' to configuration.", e);
        throw new IllegalArgumentException(e);
      } finally {
        ErrorContext.instance().reset();
      }
    }
  }

  /**
   * {@inheritDoc}
   */
  @Override
  public T getObject() throws Exception {
    return getSqlSession().getMapper(this.mapperInterface);
  }

  ...
}
  • 72行:checkDaoConfig(),当SpringBean容器初始化时候会调用到checkDaoConfig(),他是继承类中的抽象方法
  • 95行:getSqlSession().getMapper(this.mapperInterface);,通过接口获取Mapper(代理类),调用过程如下;

    • DefaultSqlSession.getMapper(Class<T> type),获取Mapper
    • Configuration.getMapper(Class<T> type, SqlSession sqlSession),从配置中获取
    • MapperRegistry.getMapper(Class<T> type, SqlSession sqlSession),从注册中心获取到实例化生成

      public <T> T getMapper(Class<T> type, SqlSession sqlSession) {
        final MapperProxyFactory<T> mapperProxyFactory = (MapperProxyFactory<T>) knownMappers.get(type);
        if (mapperProxyFactory == null) {
          throw new BindingException("Type " + type + " is not known to the MapperRegistry.");
        }
        try {
          return mapperProxyFactory.newInstance(sqlSession);
        } catch (Exception e) {
          throw new BindingException("Error getting mapper instance. Cause: " + e, e);
        }
      }
      
    • mapperProxyFactory.newInstance(sqlSession);,通过反射工程生成MapperProxy

      @SuppressWarnings("unchecked")
      protected T newInstance(MapperProxy<T> mapperProxy) {
        return (T) Proxy.newProxyInstance(mapperInterface.getClassLoader(), new Class[] { mapperInterface }, mapperProxy);
      }
      public T newInstance(SqlSession sqlSession) {
        final MapperProxy<T> mapperProxy = new MapperProxy<T>(sqlSession, mapperInterface, methodCache);
        return newInstance(mapperProxy);
      }
MapperProxy.java & 部分截取
public class MapperProxy<T> implements InvocationHandler, Serializable {

  private static final long serialVersionUID = -6424540398559729838L;
  private final SqlSession sqlSession;
  private final Class<T> mapperInterface;
  private final Map<Method, MapperMethod> methodCache;

  public MapperProxy(SqlSession sqlSession, Class<T> mapperInterface, Map<Method, MapperMethod> methodCache) {
    this.sqlSession = sqlSession;
    this.mapperInterface = mapperInterface;
    this.methodCache = methodCache;
  }

  @Override
  public Object invoke(Object proxy, Method method, Object[] args) throws Throwable {
    try {
      if (Object.class.equals(method.getDeclaringClass())) {
        return method.invoke(this, args);
      } else if (isDefaultMethod(method)) {
        return invokeDefaultMethod(proxy, method, args);
      }
    } catch (Throwable t) {
      throw ExceptionUtil.unwrapThrowable(t);
    }
    final MapperMethod mapperMethod = cachedMapperMethod(method);
    return mapperMethod.execute(sqlSession, args);
  }

  private MapperMethod cachedMapperMethod(Method method) {
    MapperMethod mapperMethod = methodCache.get(method);
    if (mapperMethod == null) {
      mapperMethod = new MapperMethod(mapperInterface, method, sqlSession.getConfiguration());
      methodCache.put(method, mapperMethod);
    }
    return mapperMethod;
  }

  @UsesJava7
  private Object invokeDefaultMethod(Object proxy, Method method, Object[] args)
      throws Throwable {
    final Constructor<MethodHandles.Lookup> constructor = MethodHandles.Lookup.class
        .getDeclaredConstructor(Class.class, int.class);
    if (!constructor.isAccessible()) {
      constructor.setAccessible(true);
    }
    final Class<?> declaringClass = method.getDeclaringClass();
    return constructor
        .newInstance(declaringClass,
            MethodHandles.Lookup.PRIVATE | MethodHandles.Lookup.PROTECTED
                | MethodHandles.Lookup.PACKAGE | MethodHandles.Lookup.PUBLIC)
        .unreflectSpecial(method, declaringClass).bindTo(proxy).invokeWithArguments(args);
  }

  ...
}
  • 58行:final MapperMethod mapperMethod = cachedMapperMethod(method);,从缓存中获取MapperMethod
  • 59行:mapperMethod.execute(sqlSession, args);,执行SQL语句,并返回结果(到这关于查询获取结果就到骨头(干)层了);INSERT、UPDATE、DELETE、SELECT

    public Object execute(SqlSession sqlSession, Object[] args) {
      Object result;
      switch (command.getType()) {
        case INSERT: {
        Object param = method.convertArgsToSqlCommandParam(args);
          result = rowCountResult(sqlSession.insert(command.getName(), param));
          break;
        }
        case UPDATE: {
          Object param = method.convertArgsToSqlCommandParam(args);
          result = rowCountResult(sqlSession.update(command.getName(), param));
          break;
        }
        case DELETE: {
          Object param = method.convertArgsToSqlCommandParam(args);
          result = rowCountResult(sqlSession.delete(command.getName(), param));
          break;
        }
        case SELECT:
          if (method.returnsVoid() && method.hasResultHandler()) {
            executeWithResultHandler(sqlSession, args);
            result = null;
          } else if (method.returnsMany()) {
            result = executeForMany(sqlSession, args);
          } else if (method.returnsMap()) {
            result = executeForMap(sqlSession, args);
          } else if (method.returnsCursor()) {
            result = executeForCursor(sqlSession, args);
          } else {
            Object param = method.convertArgsToSqlCommandParam(args);
            result = sqlSession.selectOne(command.getName(), param);
          }
          break;
        case FLUSH:
          result = sqlSession.flushStatements();
          break;
        default:
          throw new BindingException("Unknown execution method for: " + command.getName());
      }
      if (result == null && method.getReturnType().isPrimitive() && !method.returnsVoid()) {
        throw new BindingException("Mapper method '" + command.getName() 
            + " attempted to return null from a method with a primitive return type (" + method.getReturnType() + ").");
      }
      return result;
    }

以上对于MapperScannerConfigurer这一层就分析完了,从扫描定义注入到为Spring容器准备Bean的信息,代理、反射、SQL执行,基本就包括全部核心内容了,接下来在分析下SqlSessionFactoryBean

3. SqlSession容器工厂初始化(SqlSessionFactoryBean)

SqlSessionFactoryBean初始化过程中需要对一些自身内容进行处理,因此也需要实现如下接口;

  • FactoryBean<SqlSessionFactory>
  • InitializingBean -> void afterPropertiesSet() throws Exception
  • ApplicationListener<ApplicationEvent>

微信公众号:bugstack虫洞栈 & SqlSessionFactoryBean初始化流程

以上的流程其实已经很清晰的描述整个核心流程,但同样对于新手上路会有障碍,那么!好,继续!

SqlSessionFactoryBean.java & 部分截取
public void afterPropertiesSet() throws Exception {
  notNull(dataSource, "Property 'dataSource' is required");
  notNull(sqlSessionFactoryBuilder, "Property 'sqlSessionFactoryBuilder' is required");
  state((configuration == null && configLocation == null) || !(configuration != null && configLocation != null),
            "Property 'configuration' and 'configLocation' can not specified with together");
  this.sqlSessionFactory = buildSqlSessionFactory();
}
  • afterPropertiesSet(),InitializingBean接口为bean提供了初始化方法的方式,它只包括afterPropertiesSet方法,凡是继承该接口的类,在初始化bean的时候都会执行该方法。
  • 380行:buildSqlSessionFactory();内部方法构建,核心功能继续往下看。
SqlSessionFactoryBean.java & 部分截取
protected SqlSessionFactory buildSqlSessionFactory() throws IOException {
  Configuration configuration;
  XMLConfigBuilder xmlConfigBuilder = null;
  
  ...

  if (!isEmpty(this.mapperLocations)) {
    for (Resource mapperLocation : this.mapperLocations) {
      if (mapperLocation == null) {
        continue;
      }
      try {
        XMLMapperBuilder xmlMapperBuilder = new XMLMapperBuilder(mapperLocation.getInputStream(),
            configuration, mapperLocation.toString(), configuration.getSqlFragments());
        xmlMapperBuilder.parse();
      } catch (Exception e) {
        throw new NestedIOException("Failed to parse mapping resource: '" + mapperLocation + "'", e);
      } finally {
        ErrorContext.instance().reset();
      }
      if (LOGGER.isDebugEnabled()) {
        LOGGER.debug("Parsed mapper file: '" + mapperLocation + "'");
      }
    }
  } else {
    if (LOGGER.isDebugEnabled()) {
      LOGGER.debug("Property 'mapperLocations' was not specified or no matching resources found");
    }
  }
  return this.sqlSessionFactoryBuilder.build(configuration);
}
  • 513行:for (Resource mapperLocation : this.mapperLocations) 循环解析Mapper内容
  • 519行:XMLMapperBuilder xmlMapperBuilder = new XMLMapperBuilder(...) 解析XMLMapperBuilder
  • 521行:xmlMapperBuilder.parse() 执行解析,具体如下;
XMLMapperBuilder.java & 部分截取
public class XMLMapperBuilder extends BaseBuilder {
   private final XPathParser parser;
   private final MapperBuilderAssistant builderAssistant;
   private final Map<String, XNode> sqlFragments;
   private final String resource;

   private void bindMapperForNamespace() {
     String namespace = builderAssistant.getCurrentNamespace();
     if (namespace != null) {
       Class<?> boundType = null;
       try {
         boundType = Resources.classForName(namespace);
       } catch (ClassNotFoundException e) {
         //ignore, bound type is not required
       }
       if (boundType != null) {
         if (!configuration.hasMapper(boundType)) {
           // Spring may not know the real resource name so we set a flag
           // to prevent loading again this resource from the mapper interface
           // look at MapperAnnotationBuilder#loadXmlResource
           configuration.addLoadedResource("namespace:" + namespace);
           configuration.addMapper(boundType);
         }
       }
     }
   }
}
  • 这里413行非常重要,configuration.addMapper(boundType);,真正到了添加Mapper到配置中心
MapperRegistry.java & 部分截取
public class MapperRegistry {

  public <T> void addMapper(Class<T> type) {
    if (type.isInterface()) {
      if (hasMapper(type)) {
        throw new BindingException("Type " + type + " is already known to the MapperRegistry.");
      }
      boolean loadCompleted = false;
      try {
        knownMappers.put(type, new MapperProxyFactory<T>(type));
        // It's important that the type is added before the parser is run
        // otherwise the binding may automatically be attempted by the
        // mapper parser. If the type is already known, it won't try.
        MapperAnnotationBuilder parser = new MapperAnnotationBuilder(config, type);
        parser.parse();
        loadCompleted = true;
      } finally {
        if (!loadCompleted) {
          knownMappers.remove(type);
        }
      }
    }
  }
  
}
  • 67行:创建代理工程knownMappers.put(type, new MapperProxyFactory<T>(type));

截至到这,MapperScannerConfigurer、SqlSessionFactoryBean,两个类干的事情就相融合了;

  • 第一个用于扫描Dao接口设置代理类注册到IOC中,用于后续生成Bean实体类,MapperFactoryBean,并可以通过mapperInterface从Configuration获取Mapper
  • 另一个用于生成SqlSession工厂初始化,解析Mapper里的XML配置进行动态代理MapperProxyFactory->MapperProxy注入到Configuration的Mapper
  • 最终在注解类的帮助下进行方法注入,等执行操作时候即可获得动态代理对象,从而执行相应的CRUD操作

    @Resource
    private ISchoolDao schoolDao;
    
    schoolDao.querySchoolInfoById(1L);

六、综上总结

  • 分析过程较长篇幅也很大,不一定一天就能看懂整个流程,但当耐下心来一点点研究,还是可以获得很多的收获的。以后在遇到这类的异常就可以迎刃而解了,同时也有助于面试、招聘!
  • 之所以分析Mybatis最开始是想在Dao上加自定义注解,发现切面拦截不到。想到这是被动态代理的类,之后层层往往下扒直到MapperProxy.invoke!当然,Mybatis提供了自定义插件开发。
  • 以上的源码分析只是对部分核心内容进行分析,如果希望了解全部可以参考资料;MyBatis 3源码深度解析,并调试代码。IDEA中还是很方便看源码的,包括可以查看类图、调用顺序等。
  • mybatis、mybatis-spring中其实最重要的是将Mapper配置文件解析与接口类组装成代理类进行映射,以此来方便对数据库的CRUD操作。从源码分析后,可以获得更多的编程经验(套路)。
  • Mybatis相关链接;


小傅哥
4.7k 声望28.4k 粉丝

CodeGuide | 程序员编码指南 - 原创文章、案例源码、资料书籍、简历模版等下载。