Netty源码解析(八) —— channel的read操作

客户端channel在建立连接之后会关注read事件,那么read事件在哪触发的呢?
NioEventLoop中

            /**
             * 读事件和 accept事件都会经过这里,但是拿到的unsafe对象不同  所以后续执行的read操作也不一样
             * NioServerChannel进行accept操作
             * NioChannel进行read操作
             */
            if ((readyOps & (SelectionKey.OP_READ | SelectionKey.OP_ACCEPT)) != 0 || readyOps == 0) {
                unsafe.read();

io.netty.channel.nio.AbstractNioByteChannel.NioByteUnsafe#read

        @Override
        public final void read() {
            final ChannelConfig config = config();
            //若 inputClosedSeenErrorOnRead = true ,移除对 SelectionKey.OP_READ 事件的感兴趣。
            if (shouldBreakReadReady(config)) {
                //清楚读事件
                clearReadPending();
                return;
            }

            final ChannelPipeline pipeline = pipeline();
            final ByteBufAllocator allocator = config.getAllocator();
            //获取并重置allocHandle对象
            final RecvByteBufAllocator.Handle allocHandle = recvBufAllocHandle();
            allocHandle.reset(config);

            ByteBuf byteBuf = null;
            boolean close = false;//是否关闭
            try {
                do {
                    //申请bytebuf
                    byteBuf = allocHandle.allocate(allocator);
                    //读取数据,设置最后读取字节数
                    allocHandle.lastBytesRead(doReadBytes(byteBuf));
                    //没读到数据
                    if (allocHandle.lastBytesRead() <= 0) {
                        // 梅毒到数据  释放buf
                        byteBuf.release();
                        byteBuf = null;
                        //如果最后读取的字节为小于 0 ,说明对端已经关闭
                        close = allocHandle.lastBytesRead() < 0;
                        if (close) {
                            // There is nothing left to read as we received an EOF.
                            readPending = false;
                        }
                        break;
                    }

                    //读到了数据
                    allocHandle.incMessagesRead(1);
                    readPending = false;
                    //通知pipline read事件
                    pipeline.fireChannelRead(byteBuf);
                    byteBuf = null;
                } while (allocHandle.continueReading());//循环判断是否继续读取

                //读取完成
                allocHandle.readComplete();
                //通知pipline读取完成
                pipeline.fireChannelReadComplete();

                if (close) {//关闭连接
                    closeOnRead(pipeline);
                }
            } catch (Throwable t) {
                handleReadException(pipeline, byteBuf, t, close, allocHandle);
            } finally {
                .....
                if (!readPending && !config.isAutoRead()) {
                    removeReadOp();
                }
            }
        }
    }
  1. 若 inputClosedSeenErrorOnRead = true ,移除对 SelectionKey.OP_READ 事件的感兴趣。
  2. 获取并重置allocHandle对象(代理alloctor的一些功能)
  3. 读取数据,设置最后读取字节数
  4. 通知pipline read事件
  5. 通知pipline读取完成
    这次先跟着主线走,然后再回头看细节,直接定位到事件通知
    io.netty.channel.DefaultChannelPipeline#fireChannelRead
    /**
     * 有数据读入的时候会调用(InBound)  也可以手动调用
     * @param msg 客户端连接的channel
     * @return
     */
    @Override
    public final ChannelPipeline fireChannelRead(Object msg) {
        //pipline节点类的静态方法 穿进去的head
        AbstractChannelHandlerContext.invokeChannelRead(head, msg);
        return this;
    }

最后调用到io.netty.channel.DefaultChannelPipeline.HeadContext#channelRead

        /**
         * 被调用read  可以向下传递
         * @param ctx
         * @param msg
         * @throws Exception
         */
        @Override
        public void channelRead(ChannelHandlerContext ctx, Object msg) throws Exception {
            ctx.fireChannelRead(msg);
        }

ctx.fireChannelRead(msg);向下个节点传递,如果自定义了handler来处理就可以拦截channelRead的bytebuf数据来进行处理,负责一直向下传递到TailContext节点处理
io.netty.channel.DefaultChannelPipeline.TailContext#channelRead

        @Override
        public void channelRead(ChannelHandlerContext ctx, Object msg) throws Exception {

            //如果不去自定义handler处理byteBuf  最终会到TailContext 来处理
            onUnhandledInboundMessage(msg);
        }
    protected void onUnhandledInboundMessage(Object msg) {
        try {
            //日志提醒没有处理
            logger.debug(
                    "Discarded inbound message {} that reached at the tail of the pipeline. " +
                            "Please check your pipeline configuration.", msg);
        } finally {
            //释放byteBuf内存
            ReferenceCountUtil.release(msg);
        }
    }

数据到达TailContext节点之后,再onUnhandledInboundMessage方法中打印数据未处理日志,然后释放bytebuf内存
io.netty.channel.nio.AbstractNioByteChannel#doReadBytes读取操作的方法

    @Override
    /**
     * 读取数据
     */
    protected int doReadBytes(ByteBuf byteBuf) throws Exception {
        //获取handle对象
        final RecvByteBufAllocator.Handle allocHandle = unsafe().recvBufAllocHandle();
        //设置读索引=写索引
        allocHandle.attemptedBytesRead(byteBuf.writableBytes());
        //读取无数据到buf
        return byteBuf.writeBytes(javaChannel(), allocHandle.attemptedBytesRead());
    }
  1. allocHandle.lastBytesRead()小于0证明没有读取到数据,要释放bytebuf
  2. 读取到数据,allocHandle.incMessagesRead(1);
  3. allocHandle.continueReading() 循环判断是否继续读取

        @Override
        public boolean continueReading() {
            return continueReading(defaultMaybeMoreSupplier);
        }

        @Override
        public boolean continueReading(UncheckedBooleanSupplier maybeMoreDataSupplier) {
            // Keep reading if we are allowed to read more bytes, and our last read filled up the buffer we provided.
            //最后读取的字节数大于0  并且等于最大可写入的字节数
            return bytesToRead > 0 && maybeMoreDataSupplier.get();
        }
        private final UncheckedBooleanSupplier defaultMaybeMoreSupplier = new UncheckedBooleanSupplier() {
            @Override
            public boolean get() {
                //最后读取的字节数 是否等于最大可写入字节数
                return attemptedBytesRead == lastBytesRead;
            }
        };

最后读取的字节数大于0,并且最后读取的数据==尝试可写入的大小,即证明可以继续读取
出现异常时候调用io.netty.channel.nio.AbstractNioByteChannel.NioByteUnsafe#handleReadException

        private void handleReadException(ChannelPipeline pipeline, ByteBuf byteBuf, Throwable cause, boolean close,
                RecvByteBufAllocator.Handle allocHandle) {
            if (byteBuf != null) {
                if (byteBuf.isReadable()) {
                    readPending = false;
                    //把已经读取到的数据  通知到pipline中
                    pipeline.fireChannelRead(byteBuf);
                } else {
                    //释放bytebuf
                    byteBuf.release();
                }
            }
            //读取完成
            allocHandle.readComplete();
            //通知pipline读取完成
            pipeline.fireChannelReadComplete();
            //通知pipline异常
            pipeline.fireExceptionCaught(cause);
            if (close || cause instanceof IOException) {
                closeOnRead(pipeline);
            }
        }

io.netty.channel.DefaultChannelPipeline#fireExceptionCaught

    @Override
    public final ChannelPipeline fireExceptionCaught(Throwable cause) {
        AbstractChannelHandlerContext.invokeExceptionCaught(head, cause);
        return this;
    }

最终调用到TailContext的io.netty.channel.DefaultChannelPipeline.TailContext#exceptionCaught

        @Override
        public void exceptionCaught(ChannelHandlerContext ctx, Throwable cause) throws Exception {
            onUnhandledInboundException(cause);
        }
    protected void onUnhandledInboundException(Throwable cause) {
        try {
            logger.warn(
                    "An exceptionCaught() event was fired, and it reached at the tail of the pipeline. " +
                            "It usually means the last handler in the pipeline did not handle the exception.",
                    cause);
        } finally {
            ReferenceCountUtil.release(cause);
        }
    }

和读取操作一样,最终也是要再发生异常的时候释放buf的内存

猜你喜欢

转载自blog.csdn.net/u011702633/article/details/82051329