码迷,mamicode.com
首页 > 其他好文 > 详细

mybatis原理分析学习记录

时间:2019-03-31 10:02:31      阅读:121      评论:0      收藏:0      [点我收藏+]

标签:tokenizer   time   不用   expect   path   核心   driver   通过   load   

以下个人学习笔记,仅供参考,欢迎指正。

MyBatis 是支持定制化 SQL、存储过程以及高级映射的持久层框架,其主要就完成2件事情:

  • 封装JDBC操作

  • 利用反射打通Java类与SQL语句之间的相互转换

MyBatis的主要设计目的就是让我们对执行SQL语句时对输入输出的数据管理更加方便,所以方便地写出SQL和方便地获取SQL的执行结果才是MyBatis的核心竞争力。

MyBatis的配置

 spring整合mybatis(druid数据源)

技术图片
  1 (1)整合思路:把MyBatis框架中使用所涉及的核心组件配置到Spring容器中
  2 (2)步骤:
  3 -->1.添加pom依赖,mybatis-spring,spring-tx,spring-jdbc
  4 -->2.创建实体类(entity)
  5 -->3.创建数据访问接口(dao层的接口)
  6 -->4.配置SQL映射文件(resource下的mapper.xml文件)
  7 -->5.配置mybatis的配置文件(resource下的spring下的mybatis-config.xml)
  8 -->6.凡是使用了注解,都需要配置扫描注解定义的业务Bean: <context:component-scan base-package="com.one.ssm.dao"/>
  9 和<context:annotation-config/>(用于激活那些已经在spring容器里注册过的bean)
 10 
 11 
 12 <?xml version="1.0" encoding="UTF-8" ?>
 13 <!DOCTYPE configuration
 14         PUBLIC "-//mybatis.org//DTD Config 3.0//EN"
 15         "http://mybatis.org/dtd/mybatis-3-config.dtd">
 16 <configuration>
 17 <!--配置全局属性-->
 18 <settings>
 19     <!--使用jdbc的getGeneratedKeys获取数据库自增主键值-->
 20     <setting name="useGeneratedKeys" value="true"/>
 21     <!--使用列别名替换列名,默认true, eg:select name as title from table-->
 22     <setting name="useColumnLabel" value="true"/>
 23     <!--开启驼峰命名转换 table(create_time)-->entity(createTime)
 24     <setting name="mapUnderscoreToCamelCase" value="true"/>
 25 </settings>
 26 </configuration>
 27 
 28 (3)实现整合(spring-dao.xml操作)
 29 -->1.配置dataSource数据源
 30 jdbc.properties内容:
 31     jdbc.driver=com.mysql.jdbc.Driver
 32     jdbc.url=jdbc:mysql://localhost:3306/smbms?useUnicode=true&characterEncoding=UTF-8
 33     jdbc.username=root
 34     jdbc.password=123456
 35 
 36 <!--properties文件配置数据源-->
 37 <context:property-placeholder location="classpath:spring/jdbc.properties"/>
 38     <bean id="dataSource" class="com.alibaba.druid.pool.DruidDataSource" destroy-method="close">
 39         <!--配置连接池属性-->
 40         <property name="driverClassName" value="${jdbc.driver}"/>
 41         <property name="url" value="${jdbc.url}"/>
 42         <property name="username" value="${jdbc.username}"/>
 43         <property name="password" value="${jdbc.password}"/>
 44     </bean>
 45 
 46 -->2.配置SqlSessionFactoryBean
 47    <!--SqlSession 包含了所有执行数据库SQL语句的方法。能够直接地通过SqlSession实例执行映射SQL-->
 48     <!--缺少sqlSessionFactory:No bean named ‘sqlSessionFactory‘ available    完成对配置文件的读取-->
 49     <bean id="sqlSessionFactory" class="org.mybatis.spring.SqlSessionFactoryBean">
 50         <!--注入数据库连接池--><!--否则会出现java.lang.IllegalArgumentException: Property ‘dataSource‘ is required-->
 51         <property name="dataSource" ref="dataSource"/>
 52 
 53         <!--扫描entity包,使用别名,设置这个以后再Mapper配置文件中在parameterType
 54         的值就不用写成全路径名了-->
 55         <property name="typeAliasesPackage" value="com.one.ssm.entity"/>
 56 
 57         <!--扫描mapper需要的xml文件-->
 58         <property name="mapperLocations" value="classpath:mapper/*.xml"/>
 59     </bean>
 60 
 61     <!-- 配置扫描Dao接口包,动态实现Dao接口,注入到spring容器中 -->
 62     <bean class="org.mybatis.spring.mapper.MapperScannerConfigurer">
 63         <!-- 注入sqlSessionFactory -->
 64         <property name="sqlSessionFactoryBeanName" value="sqlSessionFactory"/>
 65         <!-- 给出需要扫描Dao接口包 -->
 66         <property name="basePackage" value="com.one.ssm.dao"/>
 67     </bean>
 68 
 69 (4)注入映射器的两种方式:使用了映射器注入就可以不用写dao层的接口的实现方法
 70 -->1.配置MapperFactoryBean生成映射器并注入到业务组件
 71  <bean id="userMapper" class="org.mybatis.spring.mapper.MapperFactoryBean">
 72         <!--mapperInterface属性指定映射器,只能是某一个接口类型-->
 73         <property name="mapperInterface" value="com.one.ssm.dao.UserDao"/>
 74         <!-- 注入sqlSessionFactory -->
 75         <property name="sqlSessionFactory" ref="sqlSessionFactory"/>
 76     </bean>
 77 
 78 -->2.配置MapperScannerConfiger生成映射器并注入到业务组件:优先使用MapperScannerConfiger,可以批量生成映射器的实现
 79 <!--mybatis-spring提供了MapperScannerConfigurer,
 80     可以扫描指定包中的接口并将它们直接注册为MapperFactoryBean,为了简化MapperFactoryBean映射器太多而导致多出的配置项-->
 81  <!--mybatis-spring提供了MapperScannerConfigurer,可以扫描指定包中的接口并将它们直接注册为MapperFactoryBean-->
 82     <bean class="org.mybatis.spring.mapper.MapperScannerConfigurer">
 83         <!-- 注入sqlSessionFactory -->
 84         <property name="sqlSessionFactoryBeanName" value="sqlSessionFactory"/>
 85         <!-- 给出需要扫描Dao接口包 -->
 86         <property name="basePackage" value="com.one.ssm.dao"/>
 87     </bean>
 88 
 89 (5)添加声明式事务(spring-service.xml操作)
 90 -->使用xml配置方法配置声明式事务
 91  <bean id="dataSource" class="com.alibaba.druid.pool.DruidDataSource" destroy-method="close">
 92     <property name="url" value="${jdbc.url}" />
 93     <property name="username" value="${jdbc.username}" />
 94     <property name="password" value="${jdbc.password}" />
 95     <property name="driverClassName" value="${jdbc.driver}" />
 96     </bean>
 97     <!--配置事务(事务就是对一系列的数据库操作进行统一的提交或回滚操作)管理器-->
 98 <bean id="transactionManager"
 99           class="org.springframework.jdbc.datasource.DataSourceTransactionManager">
100         <property name="dataSource" ref="dataSource"/>
101     </bean>
102 -->使用注解添加声明式事务
103  <!--声明式事务注解的添加方式-->
104     <tx:annotation-driven transaction-manager="transactionManager"/>
spring整合mybatis

 springboot整合mybatis,需要在yml文件中添加相应的配置信息。

MyBatis的主要成员

Configuration:MyBatis所有的配置信息都保存在Configuration对象之中,配置文件中的大部分配置都会存储到该类中。

SqlSession:作为MyBatis工作的主要顶层API,表示和数据库交互时的会话,完成必要数据库增删改查功能。

Executor:MyBatis执行器,是MyBatis 调度的核心,负责SQL语句的生成和查询缓存的维护。

StatementHandler:封装了JDBC Statement操作,负责对JDBC statement 的操作,如设置参数等。

ParameterHandler:负责对用户传递的参数转换成JDBC Statement所对应的数据类型。

ResultSetHandler:负责将JDBC返回的ResultSet结果集对象转换成List类型的集合。

TypeHandler:负责java数据类型和jdbc数据类型(也可以说是数据表列类型)之间的映射和转换,负责对statement对象设定特定的参数,对statement的返回结果result结果集取出特定的列

MappedStatement:MappedStatement维护一条<select|update|delete|insert>节点的封装。

SqlSource:负责根据用户传递的parameterObject,动态地生成SQL语句,将信息封装到BoundSql对象中,并返回。

BoundSql:表示动态生成的SQL语句以及相应的参数信息。

MyBatis的层次结构:

1、sqlSession -->excutor--> statementHander-->parameterHander--> typeHander-->(进入jdbc)statement(分为:preparedStatement、simpleStatement、callableStatement)-->(取出结果)resultSet--> typeHander-->resultSetHandler-->statementHandler--->excutor-->sqlSession

MyBatis的初始化(解析配置文件和初始化Configuration的过程)

String resource = "mybatis.xml";
// 加载mybatis的配置文件(它也加载关联的映射文件)
InputStream inputStream = null;
try {
    inputStream = Resources.getResourceAsStream(resource);
} catch (IOException e) {
    e.printStackTrace();
}

// 构建sqlSession的工厂
sessionFactory = new SqlSessionFactoryBuilder().build(inputStream);

首先会创建SqlSessionFactory建造者对象,然后由它进行创建SqlSessionFactory。这里用到的是建造者模式,建造者模式最简单的理解就是不手动new对象,而是由其他类来进行对象的创建。

// SqlSessionFactoryBuilder类
public SqlSessionFactory build(InputStream inputStream, String environment, Properties properties) {
    try {
//XMLConfigBuilder对象会进行XML配置文件的解析,实际为configuration节点的解析操作。
        XMLConfigBuilder parser = new XMLConfigBuilder(inputStream, environment, properties);
        return build(parser.parse()); // 开始进行解析了 :)
    } 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.
        }
    }
}
技术图片
 1 public Configuration parse() {
 2     if (parsed) {
 3         throw new BuilderException("Each XMLConfigBuilder can only be used once.");
 4     }
 5     parsed = true;
 6     parseConfiguration(parser.evalNode("/configuration"));
 7     return configuration;
 8 }
 9 
10 private void parseConfiguration(XNode root) {
11     try {
12         //issue #117 read properties first
13         propertiesElement(root.evalNode("properties"));
14         Properties settings = settingsAsProperties(root.evalNode("settings"));
15         loadCustomVfs(settings);
16         typeAliasesElement(root.evalNode("typeAliases"));
17         pluginElement(root.evalNode("plugins"));
18         objectFactoryElement(root.evalNode("objectFactory"));
19         objectWrapperFactoryElement(root.evalNode("objectWrapperFactory"));
20         reflectorFactoryElement(root.evalNode("reflectorFactory"));
21         settingsElement(settings);
22         // read it after objectFactory and objectWrapperFactory issue #631
23 
24         /* 处理environments节点数据 */
25         environmentsElement(root.evalNode("environments"));
26         databaseIdProviderElement(root.evalNode("databaseIdProvider"));
27         typeHandlerElement(root.evalNode("typeHandlers"));
28         mapperElement(root.evalNode("mappers"));
29     } catch (Exception e) {
30         throw new BuilderException("Error parsing SQL Mapper Configuration. Cause: " + e, e);
31     }
32 }
XMLConfigBuilder类

 在configuration节点下会依次解析properties/settings/.../mappers等节点配置。在解析environments节点时,会根据transactionManager的配置来创建事务管理器,根据dataSource的配置来创建DataSource对象,这里面包含了数据库登录的相关信息。在解析mappers节点时,会读取该节点下所有的mapper文件,然后进行解析,并将解析后的结果存到configuration对象中。

技术图片
 1 private void environmentsElement(XNode context) throws Exception {
 2     if (context != null) {
 3         if (environment == null) {
 4             environment = context.getStringAttribute("default");
 5         }
 6         for (XNode child : context.getChildren()) {
 7             String id = child.getStringAttribute("id");
 8             if (isSpecifiedEnvironment(id)) {
 9 
10                 /* 创建事务管理器 */
11                 TransactionFactory txFactory = transactionManagerElement(child.evalNode("transactionManager"));
12                 DataSourceFactory dsFactory = dataSourceElement(child.evalNode("dataSource"));
13                 DataSource dataSource = dsFactory.getDataSource();
14 
15                 /* 建造者模式 设计模式 */
16                 Environment.Builder environmentBuilder = new Environment.Builder(id)
17                         .transactionFactory(txFactory)
18                         .dataSource(dataSource);
19                 configuration.setEnvironment(environmentBuilder.build());
20             }
21         }
22     }
23 }
24 
25 // 解析单独的mapper文件
26 private void mapperElement(XNode parent) throws Exception {
27     if (parent != null) {
28       for (XNode child : parent.getChildren()) {
29         if ("package".equals(child.getName())) {
30           String mapperPackage = child.getStringAttribute("name");
31           configuration.addMappers(mapperPackage);
32         } else {
33           String resource = child.getStringAttribute("resource");
34           String url = child.getStringAttribute("url");
35           String mapperClass = child.getStringAttribute("class");
36           if (resource != null && url == null && mapperClass == null) {
37             ErrorContext.instance().resource(resource);
38             InputStream inputStream = Resources.getResourceAsStream(resource);
39             XMLMapperBuilder mapperParser = new XMLMapperBuilder(inputStream, configuration, resource, configuration.getSqlFragments());
40             mapperParser.parse(); // 开始解析mapper文件了 :)
41           } else if (resource == null && url != null && mapperClass == null) {
42             ErrorContext.instance().resource(url);
43             InputStream inputStream = Resources.getUrlAsStream(url);
44             XMLMapperBuilder mapperParser = new XMLMapperBuilder(inputStream, configuration, url, configuration.getSqlFragments());
45             mapperParser.parse();
46           } else if (resource == null && url == null && mapperClass != null) {
47             Class<?> mapperInterface = Resources.classForName(mapperClass);
48             configuration.addMapper(mapperInterface);
49           } else {
50             throw new BuilderException("A mapper element may only specify a url, resource or class, but not more than one.");
51           }
52         }
53       }
54     }
55   }
XMLConfigBuilder类

解析完MyBatis配置文件后,configuration就初始化完成了,然后根据configuration对象来创建SqlSession就初始化完成了

public SqlSessionFactory build(Configuration config) {
    return new DefaultSqlSessionFactory(config);
}

MyBatis的SQL查询流程

通过封装JDBC进行操作,然后使用Java反射技术完成JavaBean对象到数据库参数之间的相互转换,这种映射关系就是有TypeHandler对象来完成的,在获取数据表对应的元数据时,会保存该表所有列的数据库类型。

sqlSession = sessionFactory.openSession();

User user = sqlSession.selectOne("com.luo.dao.UserDao.getUserById", 1);
System.out.println(user);

调用selectOne方法进行SQL查询,selectOne方法最后调用的是selectList,在selectList中,会查询configuration中存储的MappedStatement对象,mapper文件中一个sql语句的配置对应一个MappedStatement对象,然后调用执行器进行查询操作。

public <T> T selectOne(String statement, Object parameter) {
    // Popular vote was to return null on 0 results and throw exception on too many.
    List<T> list = this.<T>selectList(statement, parameter);
    if (list.size() == 1) {
        return list.get(0);
    } else if (list.size() > 1) {
        throw new TooManyResultsException("Expected one result (or null) to be returned by selectOne(), but found: " + list.size());
    } else {
        return null;
    }
}

public <E> List<E> selectList(String statement, Object parameter, RowBounds rowBounds) {
    try {
        MappedStatement ms = configuration.getMappedStatement(statement);
        return executor.query(ms, wrapCollection(parameter), rowBounds, Executor.NO_RESULT_HANDLER);
    } catch (Exception e) {
        throw ExceptionFactory.wrapException("Error querying database.  Cause: " + e, e);
    } finally {
        ErrorContext.instance().reset();
    }
}

 执行器在query操作中,优先会查询缓存是否命中,命中则直接返回,否则从数据库中查询。

技术图片
 1 public <E> List<E> query(MappedStatement ms, Object parameterObject, RowBounds rowBounds, ResultHandler resultHandler) throws SQLException {
 2     /* 获取关联参数的sql,boundSql */
 3     BoundSql boundSql = ms.getBoundSql(parameterObject);
 4     /* 创建cache key值 */
 5     CacheKey key = createCacheKey(ms, parameterObject, rowBounds, boundSql);
 6     return query(ms, parameterObject, rowBounds, resultHandler, key, boundSql);
 7 }
 8 
 9 public <E> List<E> query(MappedStatement ms, Object parameterObject, RowBounds rowBounds, ResultHandler resultHandler, CacheKey key, BoundSql boundSql)
10       throws SQLException {
11     /* 获取二级缓存实例 */
12     Cache cache = ms.getCache();
13     if (cache != null) {
14         flushCacheIfRequired(ms);
15         if (ms.isUseCache() && resultHandler == null) {
16             ensureNoOutParams(ms, parameterObject, boundSql);
17             @SuppressWarnings("unchecked")
18             List<E> list = (List<E>) tcm.getObject(cache, key);
19             if (list == null) {
20                 list = delegate.<E> query(ms, parameterObject, rowBounds, resultHandler, key, boundSql);
21                 tcm.putObject(cache, key, list); // issue #578 and #116
22             }
23             return list;
24         }
25     }
26     return delegate.<E> query(ms, parameterObject, rowBounds, resultHandler, key, boundSql);
27 }
28 
29 private <E> List<E> queryFromDatabase(MappedStatement ms, Object parameter, RowBounds rowBounds, ResultHandler resultHandler, CacheKey key, BoundSql boundSql) throws SQLException {
30     List<E> list;
31     /**
32      * 先往localCache中插入一个占位对象,这个地方
33      */
34     localCache.putObject(key, EXECUTION_PLACEHOLDER);
35     try {
36         list = doQuery(ms, parameter, rowBounds, resultHandler, boundSql);
37     } finally {
38         localCache.removeObject(key);
39     }
40 
41     /* 往缓存中写入数据,也就是缓存查询结果 */
42     localCache.putObject(key, list);
43     if (ms.getStatementType() == StatementType.CALLABLE) {
44         localOutputParameterCache.putObject(key, parameter);
45     }
46     return list;
CachingExecutor类

真正的doQuery操作是由SimplyExecutor代理来完成的,该方法中有2个子流程,一个是SQL参数的设置,另一个是SQL查询操作和结果集的封装。

技术图片
 1 public <E> List<E> doQuery(MappedStatement ms, Object parameter, RowBounds rowBounds, ResultHandler resultHandler, BoundSql boundSql) throws SQLException {
 2     Statement stmt = null;
 3     try {
 4         Configuration configuration = ms.getConfiguration();
 5         StatementHandler handler = configuration.newStatementHandler(wrapper, ms, parameter, rowBounds, resultHandler, boundSql);
 6 
 7         /* 子流程1: SQL查询参数的设置 */
 8         stmt = prepareStatement(handler, ms.getStatementLog());
 9 
10         /* 子流程2: SQL查询操作和结果集封装 */
11         return handler.<E>query(stmt, resultHandler);
12     } finally {
13         closeStatement(stmt);
14     }
15 }
子流程

子流程1 SQL查询参数的设置:

技术图片
 1 private Statement prepareStatement(StatementHandler handler, Log statementLog) throws SQLException {
 2     Statement stmt;
 3     /* 获取Connection连接 */
 4     Connection connection = getConnection(statementLog);
 5 
 6     /* 准备Statement */
 7     stmt = handler.prepare(connection, transaction.getTimeout());
 8 
 9     /* 设置SQL查询中的参数值 */
10     handler.parameterize(stmt);
11     return stmt;
12 }
13 
14 // DefaultParameterHandler类
15 public void setParameters(PreparedStatement ps) {
16     /**
17      * 设置SQL参数值,从ParameterMapping中读取参数值和类型,然后设置到SQL语句中
18      */
19     ErrorContext.instance().activity("setting parameters").object(mappedStatement.getParameterMap().getId());
20     List<ParameterMapping> parameterMappings = boundSql.getParameterMappings();
21     if (parameterMappings != null) {
22         for (int i = 0; i < parameterMappings.size(); i++) {
23             ParameterMapping parameterMapping = parameterMappings.get(i);
24             if (parameterMapping.getMode() != ParameterMode.OUT) {
25                 Object value;
26                 String propertyName = parameterMapping.getProperty();
27                 if (boundSql.hasAdditionalParameter(propertyName)) { // issue #448 ask first for additional params
28                     value = boundSql.getAdditionalParameter(propertyName);
29                 } else if (parameterObject == null) {
30                     value = null;
31                 } else if (typeHandlerRegistry.hasTypeHandler(parameterObject.getClass())) {
32                     value = parameterObject;
33                 } else {
34                     MetaObject metaObject = configuration.newMetaObject(parameterObject);
35                     value = metaObject.getValue(propertyName);
36                 }
37                 TypeHandler typeHandler = parameterMapping.getTypeHandler();
38                 JdbcType jdbcType = parameterMapping.getJdbcType();
39                 if (value == null && jdbcType == null) {
40                     jdbcType = configuration.getJdbcTypeForNull();
41                 }
42                 try {
43                     typeHandler.setParameter(ps, i + 1, value, jdbcType);
44                 } catch (TypeException e) {
45                     throw new TypeException("Could not set parameters for mapping: " + parameterMapping + ". Cause: " + e, e);
46                 } catch (SQLException e) {
47                     throw new TypeException("Could not set parameters for mapping: " + parameterMapping + ". Cause: " + e, e);
48                 }
49             }
50         }
51     }
52 }
子流程1 SQL查询参数的设置:

子流程2 SQL查询结果集的封装:

技术图片
 1 public <E> List<E> query(Statement statement, ResultHandler resultHandler) throws SQLException {
 2     PreparedStatement ps = (PreparedStatement) statement;
 3     // 执行查询操作
 4     ps.execute();
 5     // 执行结果集封装
 6     return resultSetHandler.<E> handleResultSets(ps);
 7 }
 8 
 9 // DefaultReseltSetHandler类
10 public List<Object> handleResultSets(Statement stmt) throws SQLException {
11     ErrorContext.instance().activity("handling results").object(mappedStatement.getId());
12 
13     final List<Object> multipleResults = new ArrayList<Object>();
14 
15     int resultSetCount = 0;
16     /**
17      * 获取第一个ResultSet,同时获取数据库的MetaData数据,包括数据表列名、列的类型、类序号等。
18      * 这些信息都存储在了ResultSetWrapper中了
19      */
20     ResultSetWrapper rsw = getFirstResultSet(stmt);
21 
22     List<ResultMap> resultMaps = mappedStatement.getResultMaps();
23     int resultMapCount = resultMaps.size();
24     validateResultMapsCount(rsw, resultMapCount);
25     while (rsw != null && resultMapCount > resultSetCount) {
26       ResultMap resultMap = resultMaps.get(resultSetCount);
27       handleResultSet(rsw, resultMap, multipleResults, null);
28       rsw = getNextResultSet(stmt);
29       cleanUpAfterHandlingResultSet();
30       resultSetCount++;
31     }
32 
33     String[] resultSets = mappedStatement.getResultSets();
34     if (resultSets != null) {
35       while (rsw != null && resultSetCount < resultSets.length) {
36         ResultMapping parentMapping = nextResultMaps.get(resultSets[resultSetCount]);
37         if (parentMapping != null) {
38           String nestedResultMapId = parentMapping.getNestedResultMapId();
39           ResultMap resultMap = configuration.getResultMap(nestedResultMapId);
40           handleResultSet(rsw, resultMap, null, parentMapping);
41         }
42         rsw = getNextResultSet(stmt);
43         cleanUpAfterHandlingResultSet();
44         resultSetCount++;
45       }
46     }
47 
48     return collapseSingleResultList(multipleResults);
49   }
SQL查询结果集的封装

ResultSetWrapperResultSet的包装类,调用getFirstResultSet方法获取第一个ResultSet,同时获取数据库的MetaData数据,包括数据表列名、列的类型、类序号等,这些信息都存储在ResultSetWrapper类中了。然后调用handleResultSet方法来来进行结果集的封装。

技术图片
 1 private void handleResultSet(ResultSetWrapper rsw, ResultMap resultMap, List<Object> multipleResults, ResultMapping parentMapping) throws SQLException {
 2     try {
 3         if (parentMapping != null) {
 4             handleRowValues(rsw, resultMap, null, RowBounds.DEFAULT, parentMapping);
 5         } else {
 6             if (resultHandler == null) {
 7                 DefaultResultHandler defaultResultHandler = new DefaultResultHandler(objectFactory);
 8                 handleRowValues(rsw, resultMap, defaultResultHandler, rowBounds, null);
 9                 multipleResults.add(defaultResultHandler.getResultList());
10             } else {
11                 handleRowValues(rsw, resultMap, resultHandler, rowBounds, null);
12             }
13         }
14     } finally {
15         // issue #228 (close resultsets)
16         closeResultSet(rsw.getResultSet());
17     }
18 }
DefaultResultSetHandler类

调用handleRowValues方法进行结果值的设置

技术图片
 1 public void handleRowValues(ResultSetWrapper rsw, ResultMap resultMap, ResultHandler<?> resultHandler, RowBounds rowBounds, ResultMapping parentMapping) throws SQLException {
 2     if (resultMap.hasNestedResultMaps()) {
 3         ensureNoRowBounds();
 4         checkResultHandler();
 5         handleRowValuesForNestedResultMap(rsw, resultMap, resultHandler, rowBounds, parentMapping);
 6     } else {
 7         // 封装数据
 8         handleRowValuesForSimpleResultMap(rsw, resultMap, resultHandler, rowBounds, parentMapping);
 9     }
10 }
11 
12 private void handleRowValuesForSimpleResultMap(ResultSetWrapper rsw, ResultMap resultMap, ResultHandler<?> resultHandler, RowBounds rowBounds, ResultMapping parentMapping)
13         throws SQLException {
14     DefaultResultContext<Object> resultContext = new DefaultResultContext<Object>();
15     skipRows(rsw.getResultSet(), rowBounds);
16     while (shouldProcessMoreRows(resultContext, rowBounds) && rsw.getResultSet().next()) {
17         ResultMap discriminatedResultMap = resolveDiscriminatedResultMap(rsw.getResultSet(), resultMap, null);
18         Object rowValue = getRowValue(rsw, discriminatedResultMap);
19         storeObject(resultHandler, resultContext, rowValue, parentMapping, rsw.getResultSet());
20     }
21 }
22 
23 private Object getRowValue(ResultSetWrapper rsw, ResultMap resultMap) throws SQLException {
24     final ResultLoaderMap lazyLoader = new ResultLoaderMap();
25     // createResultObject为新创建的对象,数据表对应的类
26     Object rowValue = createResultObject(rsw, resultMap, lazyLoader, null);
27     if (rowValue != null && !hasTypeHandlerForResultObject(rsw, resultMap.getType())) {
28         final MetaObject metaObject = configuration.newMetaObject(rowValue);
29         boolean foundValues = this.useConstructorMappings;
30         if (shouldApplyAutomaticMappings(resultMap, false)) {
31             // 这里把数据填充进去,metaObject中包含了resultObject信息
32             foundValues = applyAutomaticMappings(rsw, resultMap, metaObject, null) || foundValues;
33         }
34         foundValues = applyPropertyMappings(rsw, resultMap, metaObject, lazyLoader, null) || foundValues;
35         foundValues = lazyLoader.size() > 0 || foundValues;
36         rowValue = (foundValues || configuration.isReturnInstanceForEmptyRow()) ? rowValue : null;
37     }
38     return rowValue;
39 }
40 
41 private boolean applyAutomaticMappings(ResultSetWrapper rsw, ResultMap resultMap, MetaObject metaObject, String columnPrefix) throws SQLException {
42     List<UnMappedColumnAutoMapping> autoMapping = createAutomaticMappings(rsw, resultMap, metaObject, columnPrefix);
43     boolean foundValues = false;
44     if (autoMapping.size() > 0) {
45         // 这里进行for循环调用,因为user表中总共有7列,所以也就调用7次
46         for (UnMappedColumnAutoMapping mapping : autoMapping) {
47             // 这里将esultSet中查询结果转换为对应的实际类型
48             final Object value = mapping.typeHandler.getResult(rsw.getResultSet(), mapping.column);
49             if (value != null) {
50                 foundValues = true;
51             }
52             if (value != null || (configuration.isCallSettersOnNulls() && !mapping.primitive)) {
53                 // gcode issue #377, call setter on nulls (value is not ‘found‘)
54                 metaObject.setValue(mapping.property, value);
55             }
56         }
57     }
58     return foundValues;
59 }
DefaultResultSetHandler类

mapping.typeHandler.getResult会获取查询结果值的实际类型,比如我们user表中id字段为int类型,那么它就对应Java中的Integer类型,然后通过调用statement.getInt("id")来获取其int值,其类型为IntegermetaObject.setValue方法会把获取到的Integer值设置到Java类中的对应字段。

metaValue.setValue方法最后会调用到Java类中对应数据域的set方法,这样也就完成了SQL查询结果集的Java类封装过程。

技术图片
 1 public void setValue(String name, Object value) {
 2     PropertyTokenizer prop = new PropertyTokenizer(name);
 3     if (prop.hasNext()) {
 4         MetaObject metaValue = metaObjectForProperty(prop.getIndexedName());
 5         if (metaValue == SystemMetaObject.NULL_META_OBJECT) {
 6             if (value == null && prop.getChildren() != null) {
 7                 // don‘t instantiate child path if value is null
 8                 return;
 9             } else {
10                 metaValue = objectWrapper.instantiatePropertyValue(name, prop, objectFactory);
11             }
12         }
13         metaValue.setValue(prop.getChildren(), value);
14     } else {
15         objectWrapper.set(prop, value);
16     }
17 }
MetaObject类

MyBatis缓存

MyBatis提供了一级缓存和二级缓存:

一级缓存是SqlSession级别的缓存,每个SqlSession对象都有一个哈希表用于缓存数据,不同SqlSession对象之间缓存不共享。同一个SqlSession对象对象执行2遍相同的SQL查询,在第一次查询执行完毕后将结果缓存起来,这样第二遍查询就不用向数据库查询了,直接返回缓存结果即可。MyBatis默认是开启一级缓存的。

二级缓存是mapper级别的缓存,二级缓存是跨SqlSession的,多个SqlSession对象可以共享同一个二级缓存。不同的SqlSession对象执行两次相同的SQL语句,第一次会将查询结果进行缓存,第二次查询直接返回二级缓存中的结果即可。MyBatis默认是不开启二级缓存的,可以在配置文件中使用如下配置来开启二级缓存:

<settings>
    <setting name="cacheEnabled" value="true"/>
</settings>

当SQL语句进行更新操作(删除/添加/更新)时,会清空对应的缓存,保证缓存中存储的都是最新的数据。

 

MyBatis动态sql

(1)多条件查询
<select id="getUserList" resultMap="userList">
        SELECT u.*,r.roleName FROM smbms_user u,smbms_role r 
        WHERE u.userName LIKE CONCAT(‘%‘,‘#{userName}‘,‘%‘)
        AND u.userRole=#{userRole} 
        AND u.userRole=r.id
    </select>

(2)if-where的用法
<select id="getUserList" resultType="User">
    SELECT * FROM smbms_user
    <where>
        <if test="userName!=null AND userName!=">
            AND userName LIKE CONCAT(‘%‘,‘#{userName}‘,‘%‘)
        </if>
        <if test="userRole!=null">
            AND userRole=#{userRole}
        </if>
    </where>
</select>
 <where>会自动去掉第一个and。

(3)if-trim
<select id="getUserList" resultType="User">
         SELECT * FROM smbms_user
         <trim prefix="where" prefixOverrides="and|or">
             <if test="userName!=null and userName!=">
                 AND userName LIKE CONCAT(‘%‘,‘#{userName}‘,‘%‘)
             </if>
             <if test="userRole!=null">
                 AND userRole=#{userRole}
             </if>
         </trim>
    </select>
<trim prefix="where" prefixOverrides="and|or">作用为自动添加where或者对and|or的自动忽略

(4)if-set 动态更新,假设没有涉及到的或者不需要更新的就可以不用更新,set标签可以自动剔除逗号(,)
<!--parameterType:属性名,如果是select就写resultMap,是其他的写对应实体对应的路径位置-->
    <update id="modifyXxx" parameterType="User">
        UPDATE smbms_user
        <set>
            <if test="userCode!=NULL">
                userCode=#{userCode},
            </if>
            <if test="userName!=null">
                userName=#{userName},
            </if>
            <if test="phone!=null">
                phone=#{phone},
            </if>
        </set>
        WHERE id=#{id}
    </update>

(5)if-set中的trim
<update id="modify" parameterType="User">
        UPDATE smbms_user
        <trim prefix="set" prefixOverrides="," suffix="where id=#{id}">
        </trim>
            <if test="userCode!=null">
                userCode=#{userCode},
            </if>
            <if test="userName!=null">
                userName=#{userName},
            </if>
            <if test="phone!=null">
                phone=#{phone},
            </if>
    </update>
<trim suffix="where id=#{id}">在trim内容的后面加上后缀

(6)foreach迭代collection数组类型的入参:对于sql语句中含有in语句,则需要foreach标签来实现sql条件的迭代
         eg:SELECT u.*   from smbms_user u WHERE userRole in(2,4)
<select id="getUserByRoleId_foreach_array" resultMap="userMapByRole">
        SELECT * FROM smbms_user WHERE userRole IN 
        <foreach collection="array" item="roleIds" open="(" separator="," close=")">
            #{roleIds}
        </foreach>
    </select>
    <resultMap id="userMapByRole" type="User">
        <id property="id" column="id"/>
        <result property="userCode" column="userCode"/>
        <result property="userName" column="userName"/>
    </resultMap>
-->Dao层接口方法为:LIst<User> getUserByRoleId_foreach_array(Integer[] roleIds)
-->item :集合中进行迭代时的别名,
-->index :指定一个名称,表示在迭代过程中每次迭代到的位置
-->separator:每次进行迭代用什么分隔符号,in条件语句用逗号(,)为分隔符
-->open:表示该语句以什么开始的,in语句以 “(”开始
-->close:表示该语句以什么符号结束 ,in语句以“)”结束
-->collection:如果是入参类型是参数是LIst,则collection属性值为list;是一个数组,则为array,如果为多参数,则需要封装成一个Map进行处理

(7)foreach迭代list类型的入参
-->Dao层接口方法为:LIst<User> getUserByRoleId_foreach_list(List<Integer> roleList);
<select id="getUserByRoleId_foreach_list" resultMap="userMapByRole">
        SELECT * FROM smbms_user WHERE userRole IN 
        <foreach collection="list" item="roleIds" open="(" separator="," close=")">
            #{roleIds}
        </foreach>
    </select>
    <resultMap id="userMapByRole" type="User">
        <id property="id" column="id"/>
        <result property="userCode" column="userCode"/>
        <result property="userName" column="userName"/>
    </resultMap>

(8)foreach迭代Map类型的入参
接口方法:public List<User> getUserByRoleId_foreach_map(Map<String,Object>  conditionMap);
@Test
    public void getUserListByUserName() throws Exception {
        Map<String,Object> conditionMap=new HashMap<String, Object>();
        List<Integer> roleList=new ArrayList<Integer>();
        roleList.add(2);
        //gender是一个限定条件
        conditionMap.put("gender",1);
        //roleIds 对应collection
        conditionMap.put("roleIds",roleList);
        System.out.println("----------------------------------");
        System.out.println(userDao.getUserByRoleId_foreach_map(conditionMap));
        System.out.println("-------------------------------------");

 <select id="getUserByRoleId_foreach_map" resultMap="userMapByRole">
        SELECT * FROM smbms_user WHERE gender=#{gender} and userRole in
        <foreach collection="roleIds" item="m" open="(" separator="," close=")">
            #{m}
        </foreach>
    </select>

(9)choose(when-otherwise)
接口方法:public List<User> getUserList_choose(@Param("userName") String userName, @Param("userRole")Integer userRole,
                                     @Param("userCode")String userCode, @Param("creationDate")Date creationDate);
测试类:
@Test
    public void getUserList_choose() throws Exception {
        List<User> userList =new ArrayList<>();
        String userName="张明";
        Integer userRole=2;
        String userCode="";
        Date creationDate=new SimpleDateFormat("yyy-MM-dd").parse("2030-10-26");
        userList=userDao.getUserList_choose(userName,userRole,userCode,creationDate);
        System.out.println(userList);
mapper:
<select id="getUserList_choose" resultMap="userMapByRole">
        SELECT * FROM smbms_user WHERE 1=1
        <choose>
            <when test="userName!=null and userName!=‘‘">
                AND userName LIKE CONCAT(‘%‘,#{userName},‘%‘)
            </when>
            <when test="userCode!=null and userCode!=‘‘">
                AND userCode LIKE CONCAT(‘%‘,#{userCode},‘%‘)
            </when>
            <when test="userRole!=null and userRole!=‘‘">
                AND userRole=#{userRole}
            </when>
            <otherwise>
                AND YEAR(creationDate)=YEAR(#{creationDate})
            </otherwise>
        </choose>
    </select>
-->when:当满足一个条件时跳出循环,
-->otherwise:当所有的when都不满足的时候,执行otherwise
-->choose:相当于switch
-->where 1=1:可以不需要处理多余的and

mybatis原理分析学习记录

标签:tokenizer   time   不用   expect   path   核心   driver   通过   load   

原文地址:https://www.cnblogs.com/zxh-xy/p/10630175.html

(0)
(0)
   
举报
评论 一句话评论(0
登录后才能评论!
© 2014 mamicode.com 版权所有  联系我们:gaon5@hotmail.com
迷上了代码!