Druid LogFilter输出可执行的SQL

配置

测试代码：

DruidDataSource dataSource = new DruidDataSource();
dataSource.setUrl("xxx");
dataSource.setUsername("xxx");
dataSource.setPassword("xxx");
dataSource.setFilters("slf4j");
dataSource.setValidationQuery("SELECT 1");
dataSource.setTestOnBorrow(true);
dataSource.setTestWhileIdle(true);
 
Connection connection = dataSource.getConnection();
 
PreparedStatement stmt = connection.prepareStatement("select * from tb_order where id=?");
stmt.setInt(1,1);
stmt.execute();

在配置输出可执行的SQL之前，看下Druid的日志输出：

SQL占位符和参数分开打印

配置输出可执行的SQL

Java启动参数配置方式

-Ddruid.log.stmt.executableSql=true

logFilter参数直接配置：

<bean id="log-filter" class="com.alibaba.druid.filter.logging.Log4jFilter">
        <property name="statementExecutableSqlLogEnable" value="true" />
  </bean>

在配置输出可执行的SQL之后，看下Druid的日志输出：

打印出了可执行的SQL

源码

调用栈是DruidPooledPreparedStatement.execute->StatementProxyImpl.execute-> FilterChainImpl.preparedStatement_execute->FilterEventAdapter.preparedStatement_execute（所有Filter都继承了FilterEventAdapter）-> LogFilter.statementExecuteAfter->LogFilter.logExecutableSql。源码如下：

// DruidPooledPreparedStatement.execute
public boolean execute() throws SQLException {
    checkOpen();
 
    incrementExecuteCount();
    transactionRecord(sql);
 
    oracleSetRowPrefetch();
 
    conn.beforeExecute();
    try {
        //这个stmt是一个StatementProxyImpl实例
        return stmt.execute();
    } catch (Throwable t) {
        errorCheck(t);
 
        throw checkException(t);
    } finally {
        conn.afterExecute();
    }
}

StatementProxyImpl.execute：

// StatementProxyImpl.execute
public boolean execute() throws SQLException {
    updateCount = null;
    lastExecuteSql = sql;
    lastExecuteType = StatementExecuteType.Execute;
    lastExecuteStartNano = -1L;
    lastExecuteTimeNano = -1L;
    //调用过滤器链的preparedStatement_execute
    firstResultSet = createChain().preparedStatement_execute(this);
    return firstResultSet;
}
 

FilterChainImpl.preparedStatement_execute：

public boolean preparedStatement_execute(PreparedStatementProxy statement) throws SQLException {
    if (this.pos < filterSize) {
        // 调用过滤器的preparedStatement_execute方法
        return nextFilter().preparedStatement_execute(this, statement);
    }
    return statement.getRawObject().execute();
}

所有过滤器都继承了FilterEventAdapter，看名字是个和事件有关的类，这个父类里面实现了preparedStatement_execute方法：

// FilterEventAdapter.preparedStatement_execute
public boolean preparedStatement_execute(FilterChain chain, PreparedStatementProxy statement) throws SQLException {
    try {
        statementExecuteBefore(statement, statement.getSql());
        // 递归调用配置的所有过滤器的preparedStatement_execute方法，直到真正执行完statement
        boolean firstResult = chain.preparedStatement_execute(statement);
        // 在执行完statement后执行，这是个空方法，具体的逻辑在子类里面实现，即LogFilter
        this.statementExecuteAfter(statement, statement.getSql(), firstResult);
 
        return firstResult;
 
    } catch (SQLException error) {
        statement_executeErrorAfter(statement, statement.getSql(), error);
        throw error;
    } catch (RuntimeException error) {
        statement_executeErrorAfter(statement, statement.getSql(), error);
        throw error;
    } catch (Error error) {
        statement_executeErrorAfter(statement, statement.getSql(), error);
        throw error;
    }
 
}

LogFilter.statementExecuteAfter方法：

protected void statementExecuteAfter(StatementProxy statement, String sql, boolean firstResult) {
    // 打印可执行的SQL
    logExecutableSql(statement, sql);
    // 统计SQL执行的时间
    if (statementExecuteAfterLogEnable && isStatementLogEnabled()) {
        statement.setLastExecuteTimeNano();
        double nanos = statement.getLastExecuteTimeNano();
        double millis = nanos / (1000 * 1000);
 
        statementLog("{conn-" + statement.getConnectionProxy().getId() + ", " + stmtId(statement) + "} executed. "
                     + millis + " millis. " + sql);
    }
}

LogFilter.logExecutableSql：

private void logExecutableSql(StatementProxy statement, String sql) {
    // 没有配置输出可执行SQL，直接返回
    if ((!isStatementExecutableSqlLogEnable()) || !isStatementLogEnabled()) {
        return;
    }
    // 获取SQL参数数量
    int parametersSize = statement.getParametersSize();
    if (parametersSize == 0) {
        statementLog("{conn-" + statement.getConnectionProxy().getId() + ", " + stmtId(statement) + "} executed. "
                     + sql);
        return;
    }
    // 获取SQL的参数，存到parameters中
    List<Object> parameters = new ArrayList<Object>(parametersSize);
    for (int i = 0; i < parametersSize; ++i) {
        JdbcParameter jdbcParam = statement.getParameter(i);
        parameters.add(jdbcParam != null
                ? jdbcParam.getValue()
                : null);
    }
 
    DbType dbType = DbType.of(statement.getConnectionProxy().getDirectDataSource().getDbType());
    // 最终由SQLUtils根据参数列表和预执行SQL，转换为可执行SQL
    String formattedSql = SQLUtils.format(sql, dbType, parameters, this.statementSqlFormatOption);
    // statementLog是个抽象方法，由具体的日志过滤器类实现（如Slf4jLogFilter）
    statementLog("{conn-" + statement.getConnectionProxy().getId() + ", " + stmtId(statement) + "} executed. "
                 + formattedSql);
}

总结

输出可执行SQL在日常排查SQL执行错误还是很实用的。其原理是在PreparedStatement.execute执行之后，调用SQLUtils.format打印出可执行的SQL。FilterEventAdapter这个类很关键，它会在SQL执行之前或者之后，调用扩展的处理，具体的处理逻辑又委派给子类实现。

Springboot Druid配置可执行sql配置_druid 执行sql_孙陆泉的博客-CSDN博客