MySQL JDBC PreparedStatement

JDBC作为JAVA访问数据库的一套规范与标准，统一了数据库操作的API，大大简化了程序开发工作。不过由于历史原因，MySQL对JDBC默认的实现与规范定义或者说其它数据库如Oracle并不一致，为了更完整记录这些差异，计划抽空写个系列，整理下这些可能会误解的常用功能。第一篇首先聊聊PreparedStatement。

MySQL JDBC PreparedStatement

Prepare SQL的产生原因与实现原理

数据库SQL执行过程包括以下阶段：词法分析->语法分析->语义分析->执行计划优化->执行。【词法分析->语法分析】这两个阶段称之为硬解析。词法分析识别SQL中每个词，语法分析解析SQL语句是否符合（SQL92、99、方言等）语法，并得到一棵语法树。

其实基于SQL的架构设计，基本都有这样一个处理过程，TDDL、ShardingJDBC、MyCAT都如此，当然这些产品都相比于数据库，支持的关键词、语法都只是其子集。另外SQL解析器可基于Yacc、Lex、Antlr、Javacc等构建，当然如果对解析性能要更好要求，则需要进行一个纯手工编写的解析器，例如阿里的Druid中的SQL解析器，应用可基于Vistor模式进行使用。

Prepare SQL也叫预编译SQL、Prepared Statements或者Parameterized Statements，就是将这类SQL中的值用占位符?替代，可以视为将SQL语句模板化或者说参数化。预编译语句的优势在于归纳为：一次编译、多次运行，省去了解析优化等过程。

Prepare的出现就是为了优化硬解析的问题，Prepare在服务器端的执行过程如下：

【Prepare】接收客户端带?的SQL, 硬解析得到语法树(stmt->Lex), 缓存在线程所在的PS cache中。此cache是一个HASH MAP. Key为stmt->id. 然后返回客户端stmt->id等信息。
【Execute】接收客户端stmt->id和参数等信息（客户端不需要再发SQL过来）。服务器根据stmt->id在PS cache中查找得到硬解析后的stmt, 并设置参数，就可以继续后面的优化和执行。
Prepare在execute阶段可以节省硬解析的时间。因此prepare适用于频繁执行的SQL。

Prepare的另一个作用是防止SQL注入，这个是纯客户端JDBC通过转义实现的。这也是一般更推荐使用PreparedStatement而不是Statement的主要理由。防SQL注入的具体实现可以参见MySQL驱动中com.mysql.jdbc.PreparedStatement.setString代码。

MySQL驱动中PrepareStament的实现逻辑

看完Prepare的功能原理后，我们看下JDBC操作MySQL时的PreparaStatement，
在com.mysql.jdbc.ConnectionImpl类中

public java.sql.PreparedStatement prepareStatement(String sql,
            int resultSetType, int resultSetConcurrency) throws SQLException {
        synchronized (getConnectionMutex()) {
            checkClosed();

            //
            // FIXME: Create warnings if can't create results of the given
            // type or concurrency
            //
            PreparedStatement pStmt = null;

            boolean canServerPrepare = true;

            String nativeSql = getProcessEscapeCodesForPrepStmts() ? nativeSQL(sql): sql;

            if (this.useServerPreparedStmts && getEmulateUnsupportedPstmts()) {
                canServerPrepare = canHandleAsServerPreparedStatement(nativeSql);
            }

            if (this.useServerPreparedStmts && canServerPrepare) {
                if (this.getCachePreparedStatements()) {
                    synchronized (this.serverSideStatementCache) {
                        pStmt = (com.mysql.jdbc.ServerPreparedStatement)this.serverSideStatementCache.remove(sql);

                        if (pStmt != null) {
                            ((com.mysql.jdbc.ServerPreparedStatement)pStmt).setClosed(false);
                            pStmt.clearParameters();
                        }

                        if (pStmt == null) {
                            try {
                                pStmt = ServerPreparedStatement.getInstance(getLoadBalanceSafeProxy(), nativeSql,
                                        this.database, resultSetType, resultSetConcurrency);
                                if (sql.length() < getPreparedStatementCacheSqlLimit()) {
                                    ((com.mysql.jdbc.ServerPreparedStatement)pStmt).isCached = true;
                                }

                                pStmt.setResultSetType(resultSetType);
                                pStmt.setResultSetConcurrency(resultSetConcurrency);
                            } catch (SQLException sqlEx) {
                                // Punt, if necessary
                                if (getEmulateUnsupportedPstmts()) {
                                    pStmt = (PreparedStatement) clientPrepareStatement(nativeSql, resultSetType, resultSetConcurrency, false);

                                    if (sql.length() < getPreparedStatementCacheSqlLimit()) {
                                        this.serverSideStatementCheckCache.put(sql, Boolean.FALSE);
                                    }
                                } else {
                                    throw sqlEx;
                                }
                            }
                        }
                    }
                } else {
                    try {
                        pStmt = ServerPreparedStatement.getInstance(getLoadBalanceSafeProxy(), nativeSql,
                                this.database, resultSetType, resultSetConcurrency);

                        pStmt.setResultSetType(resultSetType);
                        pStmt.setResultSetConcurrency(resultSetConcurrency);
                    } catch (SQLException sqlEx) {
                        // Punt, if necessary
                        if (getEmulateUnsupportedPstmts()) {
                            pStmt = (PreparedStatement) clientPrepareStatement(nativeSql, resultSetType, resultSetConcurrency, false);
                        } else {
                            throw sqlEx;
                        }
                    }
                }
            } else {
                pStmt = (PreparedStatement) clientPrepareStatement(nativeSql, resultSetType, resultSetConcurrency, false);
            }

            return pStmt;
        }
    }

从以上源代码中，我们看到其实MySQL的Prepare竟然有两种，分为是客户端（JDBC4PreparedStatement）与服务器端（ServerPrepareStatement），根据应用连接参数设置（useServerPrepStmts），选择不同的PreparedStatement。另外还会根据缓存参数设置（cachePrepStmts），选择是否从缓存重获取解析对象，该缓存是针对连接的，这对于应用端使用连接池的场景是比较适用的。

不同参数对应的Prepare区别

分别设置不同参数，查看服务器端操作日志。

使用客户端PreparedStatement，不开启缓存

public static void selectWithClientPs(int count) throws SQLException{
        Connection connection = DriverManager.getConnection("jdbc:mysql://localhost:3306/test", "root", "123456");

        long begin = System.currentTimeMillis();
        for(int i=0;i<count;i++){
            PreparedStatement statement = connection.prepareStatement("select * from test where id= ?");
            statement.setInt(1, i);
            ResultSet resultSet = statement.executeQuery();
            resultSet.close();
            statement.close();
        }
        System.out.println("selectWithClientPs span time="+(System.currentTimeMillis()-begin) + "ms");
        connection.close();
    }

MySQL服务器执行日志：

Time                 Id Command    Argument
181225 13:23:43     1 Connect   root@localhost on test
            1 Query /* mysql-connector-java-5.1.46 ( Revision: 9cc87a48e75c2d2e87c1a293b2862ce651cb256e ) */SELECT  @@session.auto_increment_increment AS auto_increment_increment, @@character_set_client AS character_set_client, @@character_set_connection AS character_set_connection, @@character_set_results AS character_set_results, @@character_set_server AS character_set_server, @@collation_server AS collation_server, @@init_connect AS init_connect, @@interactive_timeout AS interactive_timeout, @@license AS license, @@lower_case_table_names AS lower_case_table_names, @@max_allowed_packet AS max_allowed_packet, @@net_buffer_length AS net_buffer_length, @@net_write_timeout AS net_write_timeout, @@query_cache_size AS query_cache_size, @@query_cache_type AS query_cache_type, @@sql_mode AS sql_mode, @@system_time_zone AS system_time_zone, @@time_zone AS time_zone, @@tx_isolation AS transaction_isolation, @@wait_timeout AS wait_timeout
            1 Query SET NAMES latin1
            1 Query SET character_set_results = NULL
            1 Query SET autocommit=1
            1 Query select * from test where id= 0
            1 Query select * from test where id= 1
            1 Query select * from test where id= 2
            1 Quit

使用客户端PreparedStatement，开启缓存

    public static void selectWithClientPsAndCache(int count) throws SQLException{
        Connection connection = DriverManager.getConnection("jdbc:mysql://localhost:3306/test?cachePrepStmts=true", "root", "123456");

        long begin = System.currentTimeMillis();
        for(int i=0;i<count;i++){
            PreparedStatement statement = connection.prepareStatement("select * from test where id= ?");
            statement.setInt(1, i);
            ResultSet resultSet = statement.executeQuery();
            resultSet.close();
            statement.close();
        }

        System.out.println("selectWithClientPsAndCache span time="+(System.currentTimeMillis()-begin) + "ms");
        connection.close();
    }

MySQL服务器执行日志：

            2 Connect   root@localhost on test
            2 Query /* mysql-connector-java-5.1.46 ( Revision: 9cc87a48e75c2d2e87c1a293b2862ce651cb256e ) */SELECT  @@session.auto_increment_increment AS auto_increment_increment, @@character_set_client AS character_set_client, @@character_set_connection AS character_set_connection, @@character_set_results AS character_set_results, @@character_set_server AS character_set_server, @@collation_server AS collation_server, @@init_connect AS init_connect, @@interactive_timeout AS interactive_timeout, @@license AS license, @@lower_case_table_names AS lower_case_table_names, @@max_allowed_packet AS max_allowed_packet, @@net_buffer_length AS net_buffer_length, @@net_write_timeout AS net_write_timeout, @@query_cache_size AS query_cache_size, @@query_cache_type AS query_cache_type, @@sql_mode AS sql_mode, @@system_time_zone AS system_time_zone, @@time_zone AS time_zone, @@tx_isolation AS transaction_isolation, @@wait_timeout AS wait_timeout
            2 Query SET NAMES latin1
            2 Query SET character_set_results = NULL
            2 Query SET autocommit=1
            2 Query select * from test where id= 0
            2 Query select * from test where id= 1
            2 Query select * from test where id= 2
            2 Quit

使用服务器端PreparedStatement，不开启缓存

public static void selectWithServerPs(int count) throws SQLException{
        Connection connection = DriverManager.getConnection("jdbc:mysql://localhost:3306/test?useServerPrepStmts=true", "root", "123456");

        long begin = System.currentTimeMillis();
        for(int i=0;i<count;i++){
            PreparedStatement statement = connection.prepareStatement("select * from test where id= ?");
            statement.setInt(1, i);
            ResultSet resultSet = statement.executeQuery();
            resultSet.close();
            statement.close();
        }
        System.out.println("selectWithServerPs span time="+(System.currentTimeMillis()-begin) + "ms");
        connection.close();
    }

MySQL服务器执行日志：

            3 Connect   root@localhost on test
            3 Query /* mysql-connector-java-5.1.46 ( Revision: 9cc87a48e75c2d2e87c1a293b2862ce651cb256e ) */SELECT  @@session.auto_increment_increment AS auto_increment_increment, @@character_set_client AS character_set_client, @@character_set_connection AS character_set_connection, @@character_set_results AS character_set_results, @@character_set_server AS character_set_server, @@collation_server AS collation_server, @@init_connect AS init_connect, @@interactive_timeout AS interactive_timeout, @@license AS license, @@lower_case_table_names AS lower_case_table_names, @@max_allowed_packet AS max_allowed_packet, @@net_buffer_length AS net_buffer_length, @@net_write_timeout AS net_write_timeout, @@query_cache_size AS query_cache_size, @@query_cache_type AS query_cache_type, @@sql_mode AS sql_mode, @@system_time_zone AS system_time_zone, @@time_zone AS time_zone, @@tx_isolation AS transaction_isolation, @@wait_timeout AS wait_timeout
            3 Query SET NAMES latin1
            3 Query SET character_set_results = NULL
            3 Query SET autocommit=1
            3 Prepare   select * from test where id= ?
            3 Execute   select * from test where id= 0
            3 Close stmt    
            3 Prepare   select * from test where id= ?
            3 Execute   select * from test where id= 1
            3 Close stmt    
            3 Prepare   select * from test where id= ?
            3 Execute   select * from test where id= 2
            3 Close stmt    
            3 Quit

使用服务器端PreparedStatement，开启缓存

   public static void selectWithServerPsAndCache(int count) throws SQLException{
       Connection connection = DriverManager.getConnection("jdbc:mysql://localhost:3306/test?useServerPrepStmts=true&cachePrepStmts=true", "root", "123456");

       long begin = System.currentTimeMillis();
       for(int i=0;i<count;i++){

           PreparedStatement statement = connection.prepareStatement("select * from test where id= ?");
           statement.setInt(1, i);
           ResultSet resultSet = statement.executeQuery();
           resultSet.close();
           statement.close();
       }
       System.out.println("selectWithServerPsAndCache span time="+(System.currentTimeMillis()-begin) + "ms");
       connection.close();
   }

MySQL服务器执行日志：

            4 Connect   root@localhost on test
            4 Query /* mysql-connector-java-5.1.46 ( Revision: 9cc87a48e75c2d2e87c1a293b2862ce651cb256e ) */SELECT  @@session.auto_increment_increment AS auto_increment_increment, @@character_set_client AS character_set_client, @@character_set_connection AS character_set_connection, @@character_set_results AS character_set_results, @@character_set_server AS character_set_server, @@collation_server AS collation_server, @@init_connect AS init_connect, @@interactive_timeout AS interactive_timeout, @@license AS license, @@lower_case_table_names AS lower_case_table_names, @@max_allowed_packet AS max_allowed_packet, @@net_buffer_length AS net_buffer_length, @@net_write_timeout AS net_write_timeout, @@query_cache_size AS query_cache_size, @@query_cache_type AS query_cache_type, @@sql_mode AS sql_mode, @@system_time_zone AS system_time_zone, @@time_zone AS time_zone, @@tx_isolation AS transaction_isolation, @@wait_timeout AS wait_timeout
            4 Query SET NAMES latin1
            4 Query SET character_set_results = NULL
            4 Query SET autocommit=1
            4 Prepare   select * from test where id= ?
            4 Execute   select * from test where id= 0
            4 Execute   select * from test where id= 1
            4 Execute   select * from test where id= 2
            4 Quit

另外对这四种情况分别进行5000次查询，执行时间对比

selectWithClientPs span time=232601ms
selectWithClientPsAndCache span time=231493ms
selectWithServerPs span time=233999ms
selectWithServerPsAndCache span time=231262ms

结论：

使用客户端PreparedStatement，无论是否开启缓存，服务器端都不使用prepare，即硬解析时间不会减少，而且客户端缓存执行时间差别并不大。
使用服务器端PreparedStatement，如果开启缓存，则会使用prepare，硬解析仅为一次；如果不开启缓存，每次PreparedStatement进行close后，都需要重新进行prepare。
从测试效果来看，如果SQL本身比较简单，服务器端prepare并没有太大优势，使用客户端prepare即可，如果SQL较为复杂，则可尝试开启服务器端prepare，网上也有文章做过测试称可提高7%性能，不过笔者本地测试提升不到1%，测试数据实际跟SQL的复杂程度有关。

另外还有两个参数

prepStmtCacheSize参数，控制缓存的条数，MySQL驱动默认是25，实际使用时一般会根据需要调整大些；
prepStmtCacheSqlLimit参数，控制长度多大的SQL可以被缓存，MySQL驱动默认是256，实际使用时如果SQL较大，可调整大些。

关于PrepareStament，MySQL还有一个参数max_prepared_stmt_count，默认值为16382。

mysql> show variables like "max_prepare%"; 
+----------------------------+----------------------+ 
| Variable_name              | Value                | 
+----------------------------+----------------------+ 
| max_prepared_stmt_count    | 16382                |

如果创建的PS数量超过这个数值，则会报以下错误：