MyBatis源码阅读——MyBatis缓存的原理解析

约定不等于承诺〃 2022-05-23 10:51 225阅读 0赞

### 前言 ###

MyBatis 包含一个非常强大的查询缓存特性,它可以非常方便地配置和定制。除了局部的 session 缓存，默认情况下是没有二级开启缓存的。那么MyBatis的一级缓存跟二级缓存在它的框架内部是如何实现的？我们去一探究竟吧。

### 一级缓存 ###

一级缓存是默认开启的。当然你也可以关闭。在SQL 映射文件中，设置语句参数即可：

> flushCache 将其设置为 true，任何时候只要语句被调用，都会导致本地缓存和二级缓存都会被清空，默认值：false。

#### 使用 ####

因为，前面我们已经知道，一级缓存是基于SqlSession生命周期的所以，我们使用同一个例子，来看一级缓存的效果。记得开启日志。< setting name=”logImpl” value=”STDOUT\_LOGGING” />

public static void main(String[] args) throws IOException {
            String resource = "mybatis/conf/mybatis-config.xml";
            InputStream inputStream = Resources.getResourceAsStream(resource);
            SqlSessionFactory sqlSessionFactory = new SqlSessionFactoryBuilder().build(inputStream);
            SqlSession session = sqlSessionFactory.openSession();
            try {
                BlogMapper mapper = session.getMapper(BlogMapper.class);
                Blog blog = mapper.selectBlog(1L);
                System.out.println(blog);
                blog = mapper.selectBlog(1L);
                System.out.println(blog);
            } finally {
                session.close();
            }
        }

运行结果：

> Opening JDBC Connection  
> Created connection 2050462663.  
> Setting autocommit to false on JDBC Connection \[com.mysql.jdbc.JDBC4Connection@7a3793c7\]  
> ==> Preparing: select blog\_id,blog\_title,blog\_content,create\_time from blog where blog\_id = ?  
> ==> Parameters: 1(Long)  
> <== Columns: blog\_id, blog\_title, blog\_content, create\_time  
> <== Row: 1, 什么是mybatis, MyBatis是支持定制化SQL、存储过程以及高级映射的优秀的持久层框架。  
> , 2018-05-25 14:28:35.0  
> <== Total: 1  
> Blog\{blogId=1, blogTitle=’什么是mybatis’, blogContent=’MyBatis是支持定制化SQL、存储过程以及高级映射的优秀的持久层框架。  
> ‘, createTime=Fri May 25 14:28:35 CST 2018\}  
> Blog\{blogId=1, blogTitle=’什么是mybatis’, blogContent=’MyBatis是支持定制化SQL、存储过程以及高级映射的优秀的持久层框架。  
> ‘, createTime=Fri May 25 14:28:35 CST 2018\}  
> Resetting autocommit to true on JDBC Connection \[com.mysql.jdbc.JDBC4Connection@7a3793c7\]  
> Closing JDBC Connection \[com.mysql.jdbc.JDBC4Connection@7a3793c7\]  
> Returned connection 2050462663 to pool.  
> 在日志中，我们只看到了一次查询。

#### 源码 ####

在之前的文章中，我们知道Executor是最终执行者，在实现类BaseExecutor中，有个 PerpetualCache localCache;  
这是存放本地缓存（一级缓存的地方），我们看下查询的代码：

public <E> List<E> query(MappedStatement ms, Object parameter, RowBounds rowBounds, ResultHandler resultHandler, CacheKey key, BoundSql boundSql) throws SQLException {
        ErrorContext.instance().resource(ms.getResource()).activity("executing a query").object(ms.getId());
        if (closed) {
          throw new ExecutorException("Executor was closed.");
        }
        if (queryStack == 0 && ms.isFlushCacheRequired()) {
          clearLocalCache();
        }
        List<E> list;
        try {
          queryStack++;
          //首先从缓存中读取
          list = resultHandler == null ? (List<E>) localCache.getObject(key) : null;
          if (list != null) {
            handleLocallyCachedOutputParameters(ms, key, parameter, boundSql);
          } else {
            //缓存中没有，则从数据库中读取
            list = queryFromDatabase(ms, parameter, rowBounds, resultHandler, key, boundSql);
          }
        } finally {
          queryStack--;
        }
        if (queryStack == 0) {
          for (DeferredLoad deferredLoad : deferredLoads) {
            deferredLoad.load();
          }
          // issue #601
          deferredLoads.clear();
          if (configuration.getLocalCacheScope() == LocalCacheScope.STATEMENT) {
            // issue #482
            clearLocalCache();
          }
        }
        return list;
      }

localCache的作用域是BaseExecutor，而SqlSession的产生是独立的，且每次都是自己new了一个Executor  
![这里写图片描述][70]  
故，这个localCache缓存只在同一session下产生作用。  
一级缓存在update等操作时，会清空缓存。

public void clearLocalCache() {
        if (!closed) {
          localCache.clear();
          localOutputParameterCache.clear();
        }
      }

### 二级缓存 ###

#### 使用 ####

要开启二级缓存,在配置中开启的清空下，你还需要在你的 SQL 映射文件中添加一行:

现在，我们写个demo看下其作用：

public class Demo3CacheSecond {
        public static void main(String[] args) throws IOException {
            String resource = "mybatis/conf/mybatis-config.xml";
            InputStream inputStream = Resources.getResourceAsStream(resource);
            SqlSessionFactory sqlSessionFactory = new SqlSessionFactoryBuilder().build(inputStream);
    
            SqlSession session1 = sqlSessionFactory.openSession();
            try {
                BlogMapper mapper1 = session1.getMapper(BlogMapper.class);
                Blog blog1 = mapper1.selectBlog(1L);
                System.out.println("blog1:" + blog1);
                session1.commit();
            } finally {
                session1.close();
            }
    
            SqlSession session2 = sqlSessionFactory.openSession();
            try {
                BlogMapper mapper2 = session2.getMapper(BlogMapper.class);
                Blog blog2 = mapper2.selectBlog(1L);
                System.out.println("blog2:" + blog2);
                session2.commit();
            } finally {
                session2.close();
            }
        }
    }

运行结果：

> Cache Hit Ratio \[top.yuyufeng.learn.mybatis.mapper.BlogMapper\]: 0.0  
> Opening JDBC Connection  
> Created connection 1241569743.  
> Setting autocommit to false on JDBC Connection \[com.mysql.jdbc.JDBC4Connection@4a00d9cf\]  
> ==> Preparing: select blog\_id,blog\_title,blog\_content,create\_time from blog where blog\_id = ?  
> ==> Parameters: 1(Long)  
> <== Columns: blog\_id, blog\_title, blog\_content, create\_time  
> <== Row: 1, 什么是mybatis, MyBatis是支持定制化SQL、存储过程以及高级映射的优秀的持久层框架。  
> , 2018-05-25 14:28:35.0  
> <== Total: 1  
> blog1:Blog\{blogId=1, blogTitle=’什么是mybatis’, blogContent=’MyBatis是支持定制化SQL、存储过程以及高级映射的优秀的持久层框架。  
> ‘, createTime=Fri May 25 14:28:35 CST 2018\}  
> Resetting autocommit to true on JDBC Connection \[com.mysql.jdbc.JDBC4Connection@4a00d9cf\]  
> Closing JDBC Connection \[com.mysql.jdbc.JDBC4Connection@4a00d9cf\]  
> Returned connection 1241569743 to pool.  
> Cache Hit Ratio \[top.yuyufeng.learn.mybatis.mapper.BlogMapper\]: 0.5  
> blog2:Blog\{blogId=1, blogTitle=’什么是mybatis’, blogContent=’MyBatis是支持定制化SQL、存储过程以及高级映射的优秀的持久层框架。  
> ‘, createTime=Fri May 25 14:28:35 CST 2018\}

从日志中分析，在第二次查询过程中，是从缓存中读取的。现在是两个session，所以是从二级缓存中读取的。

#### 源码 ####

开启二级缓存后，执行器Executor会被CachingExecutor包装一层。（这里用到了装饰器模式）  
![这里写图片描述][70 1]  
那么，我们看下它是如何用 CachingExecutor来进行二级缓存的操作的：

public <E> List<E> query(MappedStatement ms, Object parameterObject, RowBounds rowBounds, ResultHandler resultHandler, CacheKey key, BoundSql boundSql)
          throws SQLException {
         //从MappedStatement 拿到Cache
        Cache cache = ms.getCache();
        if (cache != null) {
          flushCacheIfRequired(ms);
          if (ms.isUseCache() && resultHandler == null) {
            ensureNoOutParams(ms, parameterObject, boundSql);
            //首先从二级缓存中获取
            List<E> list = (List<E>) tcm.getObject(cache, key);
            if (list == null) {
              //二级缓存中没有，则跟原来一样获取。
              list = delegate.<E> query(ms, parameterObject, rowBounds, resultHandler, key, boundSql);
              //存入二级缓存中
              tcm.putObject(cache, key, list); // issue #578 and #116
            }
            return list;
          }
        }
        return delegate.<E> query(ms, parameterObject, rowBounds, resultHandler, key, boundSql);
      }

在代码中，tcm（TransactionalCacheManager ）是二级缓存的关键点。它是缓存存放的地方。二级缓存是Mapper域共享的，就会涉及到线程安全问题更多，所以需要更好的去维护。

public class CachingExecutor implements Executor {
          private Executor delegate;
          private TransactionalCacheManager tcm = new TransactionalCacheManager();
          }

而二级缓存是如何起作用的呢？看下如下源码：

public class TransactionalCache implements Cache {
      private Cache delegate;
      private boolean clearOnCommit;
      private Map<Object, Object> entriesToAddOnCommit;  
      private Set<Object> entriesMissedInCache;
    
      public TransactionalCache(Cache delegate) {
        this.delegate = delegate;
        this.clearOnCommit = false;
        this.entriesToAddOnCommit = new HashMap<Object, Object>();
        this.entriesMissedInCache = new HashSet<Object>();
      }
    
      @Override
      public Object getObject(Object key) {
        // issue #116 
        Object object = delegate.getObject(key);
        if (object == null) {
          //如果没拿到，那么加入失效缓存SET中
          entriesMissedInCache.add(key);
        }
        // issue #146
        if (clearOnCommit) {
          return null;
        } else {
          return object;
        }
      }
    
      @Override
      public ReadWriteLock getReadWriteLock() {
        return null;
      }
    
      @Override
      public void putObject(Object key, Object object) {
        entriesToAddOnCommit.put(key, object);
      }
    
      @Override
      public Object removeObject(Object key) {
        return null;
      }
    
      @Override
      public void clear() {
        clearOnCommit = true;
        entriesToAddOnCommit.clear();
      }
    
      public void commit() {
        if (clearOnCommit) {
          delegate.clear();
        }
        //刷新缓存
        flushPendingEntries();
        reset();
      }
    
      public void rollback() {
        unlockMissedEntries();
        reset();
      }
    
      private void reset() {
        clearOnCommit = false;
        entriesToAddOnCommit.clear();
        entriesMissedInCache.clear();
      }
    
        //刷新缓存的具体操作
      private void flushPendingEntries() {
        for (Map.Entry<Object, Object> entry : entriesToAddOnCommit.entrySet()) {
          delegate.putObject(entry.getKey(), entry.getValue());
        }
        for (Object entry : entriesMissedInCache) {
          if (!entriesToAddOnCommit.containsKey(entry)) {
            delegate.putObject(entry, null);
          }
        }
      }
    
        //移除缓存
      private void unlockMissedEntries() {
        for (Object entry : entriesMissedInCache) {
          try {
            delegate.removeObject(entry);
          } catch (Exception e) {
            log.warn("Unexpected exception while notifiying a rollback to the cache adapter."
                + "Consider upgrading your cache adapter to the latest version.  Cause: " + e);
          }
        }
      }
    
    }

> 因为Cache是从MappedStatement 拿到的，而这个是在初始化的时候从xml或者注解中加载的，是全局共享的，所以二级缓存可以在不同的session的生效。

![这里写图片描述][70 2]

### Cache是如何包装的 ###

查看org.apache.ibatis.mapping.CacheBuilder中源码 即可知道。

public Cache build() {
        setDefaultImplementations();
        Cache cache = newBaseCacheInstance(implementation, id);
        setCacheProperties(cache);
        // issue #352, do not apply decorators to custom caches
        //加上一些装饰
        if (PerpetualCache.class.equals(cache.getClass())) {
          for (Class<? extends Cache> decorator : decorators) {
            cache = newCacheDecoratorInstance(decorator, cache);
            setCacheProperties(cache);
          }
          cache = setStandardDecorators(cache);
        } else if (!LoggingCache.class.isAssignableFrom(cache.getClass())) {
          cache = new LoggingCache(cache);
        }
        return cache;
      }

### 引用 ###

[http://www.mybatis.org/mybatis-3/zh/index.html][http_www.mybatis.org_mybatis-3_zh_index.html]

[70]: /images/20220523/3d23b111aedb4568ab9fcce26b8229ca.png
[70 1]: /images/20220523/7abe63f390874d61ae307404549b34a6.png
[70 2]: /images/20220523/e48376017e854484836171b6d9047e22.png
[http_www.mybatis.org_mybatis-3_zh_index.html]: http://www.mybatis.org/mybatis-3/zh/index.html