所以我在java中使用SocketChannels来发送和接收项目的数据。该项目与第三方交互,第三方需要数据的ByteBuffer,但前缀为4个字节,这是它的长度。
接收时,我将收到一个ByteBuffer,我需要从前面解构4字节长度并提取数据。我收到的邮件不是固定的长度。
我目前的实施如下:
public PedResponse send(ByteBuffer message) {
String returnString;
try {
message.flip();
while (message.hasRemaining()) {
socketChannel.write(message);
}
ByteBuffer readBuffer = ByteBuffer.allocate(5000);
int bufferSize = socketChannel.read(readBuffer);
if (bufferSize == -1) {
socketChannel.close();
} else {
readBuffer.flip();
}
returnString = new String(deconstructMessage(readBuffer.array()), "UTF-8").trim();
response = parser.parseResponse(returnString);
} catch (IOException e) {
e.printStackTrace();
}
return response;
}
private ByteBuffer constructMessage(byte[] data) {
// Construct a buffer where we'll put the data to send
ByteBuffer sendBuffer = ByteBuffer.allocate(4 + data.length);
// it's the default, but included for clarity
sendBuffer.order(ByteOrder.BIG_ENDIAN);
// Put the 4-byte length, then the data
sendBuffer.putInt(data.length);
sendBuffer.put(data);
// Extract the actual bytes from our sendBuffer
return sendBuffer;
}
public byte[] deconstructMessage(byte[] data) {
byte[] filteredByteArray = Arrays.copyOfRange(data, 4, data.length - 4);
return filteredByteArray;
}
//Currently not being used.
private byte[] deconstructMessage(byte[] data) {
// Construct a buffer where we'll put the data to send
ByteBuffer receiveBuffer = ByteBuffer.allocate(data.length);
// it's the default, but included for clarity
receiveBuffer.order(ByteOrder.BIG_ENDIAN);
// Get the 4-byte length, then the data
receiveBuffer.getInt(data.length);
receiveBuffer.get(data);
// Extract the actual bytes from our receivedBuffer
byte[] dataReceived = receiveBuffer.array();
return dataReceived;
}
正如你所看到的,我正在创建一个大小为5000的新ByteBuffer,但理想情况下我并不想这样做。所以我的问题是,我怎么能不使用这个超大的ByteBuffer,而只是读取我收到的数据,所以我可以使用我未使用的解构消息方法?