集合对象能不能像文件那样可以算出字节数来？

集合是存放对象的类，要想计算出集合中的字节数，必须要将存放在集合中的对象实现一个方法：

这里类的成员变量以及它们的setter and getter就不提供了

public void serialize(DataOutputStream dataStream)

    throws ApplicationException{

       try{

           dataStream.writeUTF(getJXFS()) ;

           dataStream.writeUTF(getZBXZ());

           dataStream.writeUTF(getCQ());

           dataStream.writeUTF(getCJ());

           dataStream.writeUTF(getKHMC());

           dataStream.writeUTF(getCCBH());

           dataStream.writeUTF(getBHLX());

           dataStream.writeLong(MPRL==null? 0:MPRL.longValue());

           dataStream.writeUTF(getXH());

           dataStream.writeUTF(getJXZB());

           dataStream.writeUTF(getCCRQ());

           dataStream.writeUTF(getXLMC());

           dataStream.writeUTF(getTYRQ());

           dataStream.writeUTF(getBH());

           dataStream.writeUTF(getLQFS());

           dataStream.writeUTF(getAZRQ());

           dataStream.writeUTF(getKHBH());

           dataStream.writeUTF(getMC());

           dataStream.writeLong(RL==null? 0:RL.longValue());

           dataStream.flush();

       }catch(IOException e){

           throw new ApplicationException(e);

       }

    }

   

    public static UeiByq deserialize(DataInputStream dataStream)

    throws ApplicationException{

       try{

           UeiByq entity = new UeiByq();

           entity.JXFS = dataStream.readUTF() ;

           entity.ZBXZ = dataStream.readUTF();

           entity.CQ = dataStream.readUTF();

           entity.CJ = dataStream.readUTF();

           entity.KHMC = dataStream.readUTF();

           entity.CCBH = dataStream.readUTF();

           entity.BHLX = dataStream.readUTF();

           entity.MPRL = new Long(dataStream.readLong());

           entity.XH = dataStream.readUTF();

           entity.JXZB = dataStream.readUTF();

           entity.CCRQ = dataStream.readUTF();

           entity.XLMC = dataStream.readUTF();

           entity.TYRQ = dataStream.readUTF();

           entity.BH = dataStream.readUTF();

           entity.LQFS = dataStream.readUTF();

           entity.AZRQ = dataStream.readUTF();

           entity.KHBH = dataStream.readUTF();

           entity.MC = dataStream.readUTF();

           entity.RL = new Long(dataStream.readLong());

           return entity;

       }catch(IOException e){

           throw new ApplicationException(e);

       }

    }

即实现序列化和反序列化，要注意的是序列化的时候，属性是什么顺序，反序列化的时候属性就是什么顺序,还要注意的是：DataOutputStream是一个字符流，所以它包装类字节流，一般字符流是有缓冲区的(byte[])，对象的序列化，反序列化肯定是要占用缓冲区的，因此缓冲区要定义一个合适的大小；

需要注意的是:序列化的时候，调用dataStream.writeUTF(getKHBH());的时候一定要保证属性KHBH(字符串类型的)不能为空，所以在get的时候要处理null的情况，处理如下：

Public String getKHBH(){return this.KHBH==null?””:this.KHBH.trim()}

包装类型的属性的判空操作直接在流输出的时候进行处理，见上面的serialize方法。

这样集合所包含的总字节数就可以通过循环来累加了

（一个对象的字节数：dataStream.size()）

思路二：

ByteArrayOutputStream bos = new ByteArrayOutputStream();

ObjectOutputStream oos = new ObjectOutputStream(bos);

oos.writeObject(Object obj);

int size = bos.toByeArray().length;//对象的字节长度,注意输出流的缓存是动态变化的，我测试的好像是几百万字节，超过一千万就内存溢出了