Java IoBuffer.skip方法代码示例

import org.apache.mina.core.buffer.IoBuffer; //导入方法依赖的package包/类
private int messageLength ( final IoBuffer data )
{
    final int len = data.getUnsignedShort ( data.position () + 3 );
    if ( data.remaining () < 5 + len )
    {
        return -1;
    }

    data.skip ( 5 );

    return len;
}
 

import org.apache.mina.core.buffer.IoBuffer; //导入方法依赖的package包/类
@Override
public MessageDecoderResult decodable(IoSession session, IoBuffer in)
{
	
	Context context = (Context) session.getAttribute(CONTEXT);
	if(context!=null){
		if(context.limit_data){
			in.buf().position(0);
		}
	}
	//表示数据不够，需要读到新的数据后，再次调用decode()方法。
	if (in.remaining() < 2){
		return MessageDecoderResult.NEED_DATA;
	}
	else{
			context = new Context();
			//获取一个字符表示新的数据开始 用于判断异常数据
			in.skip(1);//这个数据是十六进制的 01 也就是1
			//获取第一个字符用于判断是否可以被当前解码器解码
			context.dataType = in.getInt();
			
			if(context.dataType == BeanUtil.UPLOAD_STR){
				//读取标题长度
				context.strLength = in.getInt();
				//声明数组长度
				context.byteStr = new byte[context.strLength];
				//System.out.println("我收到2了");
				session.setAttribute(CONTEXT, context);
				//表示可以解码
				return MessageDecoderResult.OK;
			}else{
				//System.out.println("服务端收到意外数据");
				return MessageDecoderResult.NOT_OK;
			}
		}
}
 

import org.apache.mina.core.buffer.IoBuffer; //导入方法依赖的package包/类
@Override
protected boolean doDecode ( final IoSession session, final IoBuffer data, final ProtocolDecoderOutput out ) throws Exception
{
    byte marker1;
    byte marker2;

    do
    {
        if ( data.remaining () < 2 ) // we may only eat the start when there could be a packet after it
        {
            return false;
        }

        // peek marker
        marker1 = data.get ( data.position () + 0 );
        marker2 = data.get ( data.position () + 1 );

        // TODO: re-think the idea of just skipping

        if ( marker1 != 0x12 || marker2 != 0x02 )
        {
            data.skip ( 2 ); // eat garbage
        }
    } while ( marker1 != 0x12 || marker2 != 0x02 );

    if ( data.remaining () < 3 )
    {
        return false;
    }

    data.order ( Sessions.isLittleEndian ( session ) ? ByteOrder.LITTLE_ENDIAN : ByteOrder.BIG_ENDIAN );

    final byte command = data.get ( data.position () + 2 );
    switch ( command )
    {
        case Messages.MC_HELLO:
            return processHello ( session, data, out );
        case Messages.MC_WELCOME:
            return processWelcome ( session, data, out );
        case Messages.MC_READ_ALL:
            out.write ( new ReadAll () );
            return true;
        case Messages.MC_DATA_UPDATE:
            return processDataUpdate ( session, data, out );
        case Messages.MC_START_BROWSE:
            out.write ( new SubscribeBrowse () );
            return true;
        case Messages.MC_STOP_BROWSE:
            out.write ( new UnsubscribeBrowse () );
            return true;
        case Messages.MC_NS_ADDED:
            return processBrowseAdded ( session, data, out );
        case Messages.MC_WRITE_COMMAND:
            return processWriteCommand ( session, data, out );
        case Messages.MC_WRITE_RESULT:
            return processWriteResult ( session, data, out );
    }

    throw new ProtocolCodecException ( String.format ( "Message code %02x is unkown", command ) );
}
 

import org.apache.mina.core.buffer.IoBuffer; //导入方法依赖的package包/类
@Override
public MessageDecoderResult decode(IoSession session, IoBuffer in,ProtocolDecoderOutput outPut) throws Exception {
	//将客户端发来的对象进行封装
	//TODO  等待测试超长数据是否能正常解码 
	Context context = (Context) session.getAttribute(CONTEXT);
	//跳过第一个字节
	if(!context.init){
		in.buf().position(0);//由于分包之后mina不能让没有改变的缓冲数据返回正常,于是先移动了下游标,这里给归0
		context.init = true;
		in.skip(1);
		in.getInt();
		in.getInt();
	}
 
	int count = context.count;
	//System.out.println("这里是第一次COnut："+context.count);
	//System.out.println("一共有"+in.remaining()+"数据");
	while (in.hasRemaining()) {
		//System.out.println("循环里面的Count："+count);
		byte b = in.get();
		if(b == 1){
			//收到下一条的起始数据了,证明此条数据已经残缺 重置缓冲区 作废此信息
			in.buf().position(1);//移动游标,mina不允许不使用数据就返回
			context.reset();//重置
			context.limit_data = true;//用于判断是否是前一次包破损后遗留下的新数据包  decodable中判断
			session.setAttribute(CONTEXT, context);
			return MessageDecoderResult.OK;//给他返回个正常解码,然后才能继续解码
		}
		new String(context.byteStr,BeanUtil.charset);
		//如果标题没读完 继续读
		if(!context.isReadName){
			context.byteStr[count] = b;
			if(count == context.strLength-1){
				//标题读完  byte[]转换为字符串
				context.fileName = new String(context.byteStr,BeanUtil.charset);
				//System.out.println(context.fileName);
				//count = -1;
				context.isReadName = true;
				//跳出程序
				break;
			}
		}
		/*
		if(context.isReadName && count != -1){
			//如果读取完了标题那么读取其他内容
			//byteFile[count] = b;
		}
		//byteFile[count] = b;
		 */
		//这里并未判断是否后面还有数据就加了1
		if(in.buf().position()<=in.buf().limit())
		count++;
          }
 
	context.count = count;
	session.setAttribute(CONTEXT, context);
	//如果内容全部读完  那么就存入并返回数据
	//System.out.println("Count:"+context.count+";----StrLen:"+context.strLength);
	if(context.count == context.strLength-1){
		//这里就代表数据全部接收完了 返回出去
		
		BaseMessage message = new BaseMessage();
		message.setDataType(context.dataType);
		StringBean bean = new StringBean();
		bean.setFileName(context.fileName);
		bean.setStrLength(context.strLength);
		message.setData(bean);
		outPut.write(message);
		//重置
		context.reset();
 
	}else{
		return  MessageDecoderResult.NEED_DATA;
	}
	//this.finishDecode(session, outPut);
	
	return MessageDecoderResult.OK;
}