Appearance
Netty - 03 黏包半包分析
在Netty中仍然存在黏包半包现象，但是Netty提供了一些处理器可以帮助我们方便快捷地处理黏包半包现象。
1. 短连接
短连接方案不是Netty提供的，但在这里仍作为一种方案展示。 短连接方案是指客户端发送完一条消息就断开连接，以连接作为消息之间的分隔符，所以我们可以把发送消息抽取为一个方法：
java
public static void main(String[] args) throws InterruptedException {
for (int i = 0; i < 10; i++) {
sendMessage("abcdefgh");
}
}
private static void sendMessage(String message) throws InterruptedException {
NioEventLoopGroup group = new NioEventLoopGroup();
Channel channel = new Bootstrap()
.group(group)
.channel(NioSocketChannel.class)
.handler(new ChannelInitializer<NioSocketChannel>() {
@Override
protected void initChannel(NioSocketChannel nioSocketChannel) throws Exception {
nioSocketChannel.pipeline().addLast(new StringEncoder());
}
})
.connect(new InetSocketAddress("127.0.0.1", 8080))
.sync()
.channel();
channel.writeAndFlush(message);
channel.close().sync();
group.shutdownGracefully();
}短连接方案可以解决黏包问题，但是无法解决半包问题！
2. 定长消息
Netty提供了定长消息处理器FixedLengthFrameDecoder，我们可以使用它来指定一条消息的长度：
java
public static void main(String[] args) {
new ServerBootstrap()
.group(new NioEventLoopGroup(1), new NioEventLoopGroup())
.channel(NioServerSocketChannel.class)
.childHandler(new ChannelInitializer<NioSocketChannel>() {
@Override
protected void initChannel(NioSocketChannel nioSocketChannel) throws Exception {
// 指定消息长度为8字节
nioSocketChannel.pipeline().addLast(new FixedLengthFrameDecoder(8));
nioSocketChannel.pipeline().addLast(new LoggingHandler(LogLevel.DEBUG));
}
})
.bind(8080);
}3. 基于分隔符分隔消息
Netty提供了基于特殊符号分隔消息的处理器，我们可以指定特殊符号作为消息的分隔符以切分消息。Netty也提供了现成的基于换行符分隔消息的处理器。
DelimiterBasedFrameDecoder(int maxFrameLength, ByteBuf delimiter)：第一个参数指定了消息的最大长度，第二个参数指定了特殊符。LineBasedFrameDecoder(final int maxLength)：指定换行符(\n或\r\n)作为消息分隔符；
为什么要指定消息的最大长度？如果没有指定消息的最大长度，那么服务端在没有遇到分隔符时会一直接收下去，这显然会导致内存溢出。
下面以换行符作为例子演示基于分隔符分隔消息：
java
public static void main(String[] args) {
new ServerBootstrap()
.group(new NioEventLoopGroup(1), new NioEventLoopGroup())
.channel(NioServerSocketChannel.class)
.childHandler(new ChannelInitializer<NioSocketChannel>() {
@Override
protected void initChannel(NioSocketChannel nioSocketChannel) throws Exception {
// 以换行分割符分隔消息，消息最大长度为1024
nioSocketChannel.pipeline().addLast(new LineBasedFrameDecoder(1024));
nioSocketChannel.pipeline().addLast(new LoggingHandler(LogLevel.DEBUG));
}
})
.bind(8080);
}java
public static void main(String[] args) throws InterruptedException {
NioEventLoopGroup group = new NioEventLoopGroup();
Channel channel = new Bootstrap()
.group(group)
.channel(NioSocketChannel.class)
.handler(new ChannelInitializer<NioSocketChannel>() {
@Override
protected void initChannel(NioSocketChannel nioSocketChannel) throws Exception {
nioSocketChannel.pipeline().addLast(new StringEncoder());
}
})
.connect(new InetSocketAddress("127.0.0.1", 8080))
.sync()
.channel();
for (int i = 0; i < 10; i++) {
channel.writeAndFlush(makeString());
}
channel.close().sync();
group.shutdownGracefully();
}
/**
* 创建随机长度的消息
* @return
*/
private static String makeString(){
int i = new Random().nextInt(100) + 1;
StringBuilder stringBuilder = new StringBuilder();
char c = '0';
for (int j = 0; j < i; j++) {
stringBuilder.append(c);
}
stringBuilder.append("\n");
return stringBuilder.toString();
}4. 基于长度字段分隔消息
Netty提供了LengthFieldBasedFrameDecoder，用于提供基于长度字段的消息分隔方案。
java
LengthFieldBasedFrameDecoder(
int maxFrameLength,
int lengthFieldOffset,
int lengthFieldLength,
int lengthAdjustment,
int initialBytesToStrip)LengthFieldBasedFrameDecoder构造方法有五个参数：
maxFrameLength：指定了消息的最大长度；lengthFieldOffset：指定了长度字段的偏移量；lengthFieldLength：指定了长度字段的长度；lengthAdjustment：指定了内容字段基于长度字段的偏移量；initialBytesToStrip：指定消息分隔后，需要移除的头部字节数；
现在给出几个LengthFieldBasedFrameDecoder示例说明上述参数的作用：
txt
* lengthFieldOffset = 0 // 长度字段偏移量为0
* lengthFieldLength = 2 // 长度字段长度为2
* lengthAdjustment = 0 // 内容字段基于长度字段的偏移量为0
* initialBytesToStrip = 0 // 头部字节移除量为0，表示不移除任何字节
*
* BEFORE DECODE (14 bytes) AFTER DECODE (14 bytes)
* +--------+----------------+ +--------+----------------+
* | Length | Actual Content |----->| Length | Actual Content |
* | 0x000C | "HELLO, WORLD" | | 0x000C | "HELLO, WORLD" |
* +--------+----------------+ +--------+----------------+txt
* lengthFieldOffset = 0
* lengthFieldLength = 2
* lengthAdjustment = 0
* initialBytesToStrip = 2
*
* BEFORE DECODE (14 bytes) AFTER DECODE (12 bytes)
* +--------+----------------+ +----------------+
* | Length | Actual Content |----->| Actual Content |
* | 0x000C | "HELLO, WORLD" | | "HELLO, WORLD" |
* +--------+----------------+ +----------------+txt
* lengthFieldOffset = 2 // 在最开始增加头部，所以长度字段偏移量为2
* lengthFieldLength = 3
* lengthAdjustment = 0
* initialBytesToStrip = 0
*
* BEFORE DECODE (17 bytes) AFTER DECODE (17 bytes)
* +----------+----------+----------------+ +----------+----------+----------------+
* | Header 1 | Length | Actual Content |----->| Header 1 | Length | Actual Content |
* | 0xCAFE | 0x00000C | "HELLO, WORLD" | | 0xCAFE | 0x00000C | "HELLO, WORLD" |
* +----------+----------+----------------+ +----------+----------+----------------+txt
* lengthFieldOffset = 0
* lengthFieldLength = 3
* lengthAdjustment = 2 (= the length of Header 1) // 头部字段在长度和内容之间，内容字段基于长度字段偏移量为2
* initialBytesToStrip = 0
*
* BEFORE DECODE (17 bytes) AFTER DECODE (17 bytes)
* +----------+----------+----------------+ +----------+----------+----------------+
* | Length | Header 1 | Actual Content |----->| Length | Header 1 | Actual Content |
* | 0x00000C | 0xCAFE | "HELLO, WORLD" | | 0x00000C | 0xCAFE | "HELLO, WORLD" |
* +----------+----------+----------------+ +----------+----------+----------------+txt
* lengthFieldOffset = 1 (= the length of HDR1)
* lengthFieldLength = 2
* lengthAdjustment = 1 (= the length of HDR2)
* initialBytesToStrip = 3 (= the length of HDR1 + LEN)
*
* BEFORE DECODE (16 bytes) AFTER DECODE (13 bytes)
* +------+--------+------+----------------+ +------+----------------+
* | HDR1 | Length | HDR2 | Actual Content |----->| HDR2 | Actual Content |
* | 0xCA | 0x000C | 0xFE | "HELLO, WORLD" | | 0xFE | "HELLO, WORLD" |
* +------+--------+------+----------------+ +------+----------------+下面给出代码示例：
java
public static void main(String[] args) {
new ServerBootstrap()
.group(new NioEventLoopGroup(1), new NioEventLoopGroup())
.channel(NioServerSocketChannel.class)
.childHandler(new ChannelInitializer<NioSocketChannel>() {
@Override
protected void initChannel(NioSocketChannel nioSocketChannel) throws Exception {
// 基于长度字段的分隔符
nioSocketChannel.pipeline().addLast(new LengthFieldBasedFrameDecoder(
1024,1,4,1,6
));
nioSocketChannel.pipeline().addLast(new LoggingHandler(LogLevel.DEBUG));
}
})
.bind(8080);
}java
public static void main(String[] args) throws InterruptedException {
NioEventLoopGroup group = new NioEventLoopGroup();
Channel channel = new Bootstrap()
.group(group)
.channel(NioSocketChannel.class)
.handler(new ChannelInitializer<NioSocketChannel>() {
@Override
protected void initChannel(NioSocketChannel nioSocketChannel) throws Exception {
}
})
.connect(new InetSocketAddress("127.0.0.1", 8080))
.sync()
.channel();
for (int i = 0; i < 10; i++) {
String content = makeString();
ByteBuf buffer = ByteBufAllocator.DEFAULT.buffer();
buffer.writeByte(1); // 头部1
buffer.writeInt(content.getBytes(StandardCharsets.UTF_8).length); // 长度
buffer.writeByte(1); // 头部2
buffer.writeBytes(content.getBytes(StandardCharsets.UTF_8)); // 内容
channel.writeAndFlush(buffer);
}
channel.close().sync();
group.shutdownGracefully();
}
/**
* 创建随机长度的消息
* @return
*/
private static String makeString(){
int i = new Random().nextInt(100) + 1;
StringBuilder stringBuilder = new StringBuilder();
char c = '0';
for (int j = 0; j < i; j++) {
stringBuilder.append(c);
}
return stringBuilder.toString();
}