iBatis3基于方言(Dialect)的分页

(注:以下代码是基于ibatis3 beta4的扩展，ibatis3正式版如果实现改变，将会继续跟进修改)

 

 

iBatis3默认使用的分页是基于游标的分页，而这种分页在不同的数据库上性能差异不一致，最好的办法当然是使用类似hibernate的基于方言(Dialect)的物理分页功能。

iBatis3现在提供插件功能，通过插件我们可以编写自己的拦截器来拦截iBatis3的主要执行方法来完成相关功能的扩展。

 

 

 

能够拦截的的类如下:

 

Executor
	(update,query,flushStatements,commit,rollback,getTransaction,close,isClosed)
ParameterHandler
	(getParameterObject,setParameters)
ResultSetHandler
	(handleResultSets,handleOutputParameters)
StatementHandler
	(prepare,parameterize,batch,update,query)

 

 但此次我们感兴趣的是Executor.query()方法，查询方法最终都是执行该方法。

该方法完整是:

 

Executor.query(MappedStatement ms, Object parameter, RowBounds rowBounds, ResultHandler resultHandler) throws SQLException;

 

 

分页方言的基本实现:

以Mysql数据库示例,即有一个方法 query(sql,offset,limit);

 

 

query("select * from user",5,10);

 

 

经过我们的拦截器拦截处理,变成

 

query("select * from user limit 5,10",0,0);

 

 

1.拦截类实现:

 

@Intercepts({@Signature(
		type= Executor.class,
		method = "query",
		args = {MappedStatement.class, Object.class, RowBounds.class, ResultHandler.class})})
public class OffsetLimitInterceptor implements Interceptor{
	static int MAPPED_STATEMENT_INDEX = 0;
	static int PARAMETER_INDEX = 1;
	static int ROWBOUNDS_INDEX = 2;
	static int RESULT_HANDLER_INDEX = 3;
	
	Dialect dialect;
	
	public Object intercept(Invocation invocation) throws Throwable {
		processIntercept(invocation.getArgs());
		return invocation.proceed();
	}

	void processIntercept(final Object[] queryArgs) {
		//queryArgs = query(MappedStatement ms, Object parameter, RowBounds rowBounds, ResultHandler resultHandler)
		MappedStatement ms = (MappedStatement)queryArgs[MAPPED_STATEMENT_INDEX];
		Object parameter = queryArgs[PARAMETER_INDEX];
		final RowBounds rowBounds = (RowBounds)queryArgs[ROWBOUNDS_INDEX];
		int offset = rowBounds.getOffset();
		int limit = rowBounds.getLimit();
		
		if(dialect.supportsLimit() && (offset != RowBounds.NO_ROW_OFFSET || limit != RowBounds.NO_ROW_LIMIT)) {
			BoundSql boundSql = ms.getBoundSql(parameter);
			String sql = boundSql.getSql().trim();
			if (dialect.supportsLimitOffset()) {
				sql = dialect.getLimitString(sql, offset, limit);
				offset = RowBounds.NO_ROW_OFFSET;
			} else {
				sql = dialect.getLimitString(sql, 0, limit);
			}
			limit = RowBounds.NO_ROW_LIMIT;
			
			queryArgs[ROWBOUNDS_INDEX] = new RowBounds(offset,limit);
			BoundSql newBoundSql = new BoundSql(sql, boundSql.getParameterMappings(), boundSql.getParameterObject());
			MappedStatement newMs = copyFromMappedStatement(ms, new BoundSqlSqlSource(newBoundSql));
			queryArgs[MAPPED_STATEMENT_INDEX] = newMs;
		}
	}
	
	private MappedStatement copyFromMappedStatement(MappedStatement ms,SqlSource newSqlSource) {
		Builder builder = new MappedStatement.Builder(ms.getConfiguration(),ms.getId(),newSqlSource,ms.getSqlCommandType());
		builder.resource(ms.getResource());
		builder.fetchSize(ms.getFetchSize());
		builder.statementType(ms.getStatementType());
		builder.keyGenerator(ms.getKeyGenerator());
		builder.keyProperty(ms.getKeyProperty());
		builder.timeout(ms.getTimeout());
		builder.parameterMap(ms.getParameterMap());
		builder.resultMaps(ms.getResultMaps());
		builder.cache(ms.getCache());
		MappedStatement newMs = builder.build();
		return newMs;
	}

	public Object plugin(Object target) {
		return Plugin.wrap(target, this);
	}

	public void setProperties(Properties properties) {
		String dialectClass = new PropertiesHelper(properties).getRequiredString("dialectClass");
		try {
			dialect = (Dialect)Class.forName(dialectClass).newInstance();
		} catch (Exception e) {
			throw new RuntimeException("cannot create dialect instance by dialectClass:"+dialectClass,e);
		} 
		System.out.println(OffsetLimitInterceptor.class.getSimpleName()+".dialect="+dialectClass);
	}
	
	public static class BoundSqlSqlSource implements SqlSource {
		BoundSql boundSql;
		public BoundSqlSqlSource(BoundSql boundSql) {
			this.boundSql = boundSql;
		}
		public BoundSql getBoundSql(Object parameterObject) {
			return boundSql;
		}
	}
	
}

 

 

2.ibatis3配置文件内容:

 

<configuration>
	<plugins>
		<!-- 指定数据库分页方言Dialect, 其它方言:OracleDialect,SQLServerDialect,SybaseDialect,DB2Dialect,PostgreSQLDialect,MySQLDialect,DerbyDialect-->
		<plugin interceptor="cn.org.rapid_framework.ibatis3.plugin.OffsetLimitInterceptor">
			<property name="dialectClass" value="cn.org.rapid_framework.jdbc.dialect.MySQLDialect"/>
		</plugin>
	</plugins>

	<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://localhost:3306/test" />
				<property name="username" value="root" />
				<property name="password" value="123456" />
			</dataSource>
		</environment>
	</environments>

	<mappers>
		<mapper resource="com/company/project/model/mapper/BlogMapper.xml" />
	</mappers>
</configuration>

 

3.MysqlDialect实现

 

public class MySQLDialect extends Dialect{

	public boolean supportsLimitOffset(){
		return true;
	}
	
	public boolean supportsLimit() {   
		return true;   
	}  
    
	public String getLimitString(String sql, int offset, int limit) {
    		return getLimitString(sql,offset,String.valueOf(offset),limit,String.valueOf(limit));
	}
    
	public String getLimitString(String sql, int offset,String offsetPlaceholder, int limit, String limitPlaceholder) {
		if (offset > 0) {   
			return sql + " limit "+offsetPlaceholder+","+limitPlaceholder; 
		} else {   
		    return sql + " limit "+limitPlaceholder;
		}  
	}   
  
}

 

其它数据库的Dialect实现请查看:

http://rapid-framework.googlecode.com/svn/trunk/rapid-framework/src/rapid_framework_common/cn/org/rapid_framework/jdbc/dialect/

 

 

4.分页使用

直接调用SqlSession.selectList(statement, parameter, new RowBounds(offset,limit))即可使用物理分页

 

 

5.存在的问题

现不是使用占位符的方式(即不是“limit ?,?”而是使用“limit 8,20”)使用分页

 

完整代码可以查看即将发布的rapid-framework 3.0的ibatis3插件