基于Nodejs的Tcp封包和解包的理解

let head = new Buffer(4);
let jsonStr = JSON.stringify(json);
let body = new Buffer(jsonStr);
//超过一字节的大端写入
head.writeInt32BE(body.byteLength, 0);
let buffer = Buffer.concat([head, body]);

let dataReadStart = 0; //新数据的起始位置
let dataLength = buffer.length; // 要拷贝数据的长度
let availableLen = _bufferLength - _dataLen; // 缓冲区剩余可用空间

// buffer剩余空间不足够存储本次数据
if (availableLen < dataLength) {
 let newLength = Math.ceil((_dataLen + dataLength) / _bufferLength) * _bufferLength;
 let _tempBuffer = Buffer.alloc(newLength);
 
 // 将旧数据复制到新buffer并且修正相关参数
 if (_writePointer < _readPointer) { // 数据存储在旧buffer的尾部+头部的顺序
  let dataTailLen = _bufferLength - _readPointer;
  _buffer.copy(_tempBuffer, 0, _readPointer, _readPointer + dataTailLen);
  _buffer.copy(_tempBuffer, dataTailLen, 0, _writePointer);
 } else { // 数据是按照顺序进行的完整存储
  _buffer.copy(_tempBuffer, 0, _readPointer, _writePointer);
 }
 _bufferLength = newLength;
 _buffer = _tempBuffer;
 _tempBuffer = null;
 _readPointer = 0;
 _writePointer = _dataLen;

 //存储新到来的buffer
 buffer.copy(_buffer, _writePointer, dataReadStart, dataReadStart + dataLength);
 _dataLen += dataLength;
 _writePointer += dataLength;

} else if (_writePointer + dataLength > _bufferLength) {
// 空间够用情况下，但是数据会冲破缓冲区尾部，部分存到缓冲区旧数据后，一部分存到缓冲区开始位置
 // 缓冲区尾部剩余空间的长度
 let bufferTailLength = _bufferLength - _writePointer;

 // 数据尾部位置
 let dataEndPosition = dataReadStart + bufferTailLength;
 buffer.copy(_buffer, _writePointer, dataReadStart, dataEndPosition);

 // data剩余未拷贝进缓存的长度
 let restDataLen = dataLength - bufferTailLength;
 buffer.copy(_buffer, 0, dataEndPosition, dataLength);

 _dataLen = _dataLen + dataLength;
 _writePointer = restDataLen

} else { // 剩余空间足够存储数据，直接拷贝数据到缓冲区
 buffer.copy(_buffer, _writePointer, dataReadStart, dataReadStart + dataLength);
 _dataLen = _dataLen + dataLength;
 _writePointer = _writePointer + dataLength
}

let _dataHeadLen = 4;
timer && clearInterval(timer);
timer = setInterval(()=>{
 // 缓冲区数据不够解析出包头
 if (_dataLen < _dataHeadLen) {
  console.log('数据长度小于包头规定长度，等待数据......')
  clearInterval(timer);
 }
 // 解析包头长度
 // 尾部最后剩余可读字节长度
 let restDataLen = _bufferLength - _readPointer;
 let dataLen = 0;
 let headBuffer = Buffer.alloc(_dataHeadLen);
 // 数据包为分段存储，不能直接解析出包头，先拼接
 if (restDataLen < _dataHeadLen) {
  // 取出第一部分头部字节
  _buffer.copy(headBuffer, 0, _readPointer, _bufferLength)
  // 取出第二部分头部字节
  let unReadHeadLen = _dataHeadLen - restDataLen;
  _buffer.copy(headBuffer, restDataLen, 0, unReadHeadLen)
  dataLen = headBuffer.readUInt32BE(0);

 } else {
  _buffer.copy(headBuffer, 0, _readPointer, _readPointer + _dataHeadLen);
  dataLen = headBuffer.readUInt32BE(0);;
 }

 // 数据长度不够读取，直接返回
 if (_dataLen - _dataHeadLen < dataLen) {
  log.info("缓冲区已有body数据长度小于包头定义body的长度，等待数据......")
  clearInterval(timer);

 } else { // 数据够读，读取数据包 
  let package = Buffer.alloc(dataLen);
  // 数据是分段存储，需要分两次读取
  if (_bufferLength - _readPointer < dataLen) {
   let firstPartLen = _bufferLength - _readPointer;
   // 读取第一部分，直接到字符尾部的数据
   _buffer.copy(package, 0, _readPointer, firstPartLen + _readPointer);
   // 读取第二部分，存储在开头的数据
   let secondPartLen = dataLen - firstPartLen;
   _buffer.copy(package, firstPartLen, 0, secondPartLen);
   _readPointer = secondPartLen; //更新可读起点

  } else { // 直接读取数据
   _buffer.copy(package, 0, _readPointer, _readPointer + dataLen);
   _readPointer += dataLen; //更新可读起点
  }

  _dataLen -= readData.length; //更新数据长度
  // 已经读取完所有数据
  if (_readPointer === _writePointer) {
   clearInterval(timer)
  }

  //开始解包
  callback(package);
   
 }
}, 50);

let headBytes = 4;
let head = new Buffer(headBytes);
buffer.copy(head, 0, 0, headBytes);
let dataLen = head.readUInt32BE();
const body = new Buffer(dataLen);
buffer.copy(body, 0, headBytes, headBytes + dataLen)

let content = null;
try {
 const str = body.toString('utf-8');
 if(str === ''){
  content = null;
 }else{
  content = JSON.parse(body);
 }
} catch (e) {
 log.error('head指定body长度有问题')
}
//传递给业务层
callback(content);