Kafka.net使用编程入门(一)

最近研究分布式消息队列，分享下！

首先zookeeper 和 kafka 压缩包解压并配置好！

我本机zookeeper环境配置如下:

D:\Worksoftware\ApacheZookeeper3\conf\zoo.cfg

以下是kafka的配置

D:\Worksoftware\Apachekafka2.11\config\server.properties

我已经加了path环境变量，没加的话需要到zookeeper对应bin目录下执行zkServer

然后执行cmd命令:

结果:

然后打开第二个dos窗口,我没加环境变量path,执行kafka命令如下:

重头戏来了，开始kafka C#客户端处理:

首先引用kafka-net.dll,可以用vs2013的nuget下载，

以下是Prorame.cs:

[csharp] view plaincopy

class Program
{
static void Main(string[] args)
{
const string topicName = "test";
var options = new KafkaOptions(new Uri("http://localhost:9092"))
{
Log = new ConsoleLog()
};
Task.Run(() =>
{
var consumer = new Consumer(new ConsumerOptions(topicName, new BrokerRouter(options)) { Log = new ConsoleLog() });
foreach (var data in consumer.Consume())
{
Console.WriteLine("Response: PartitionId={0},Offset={1} :Value={2}", data.Meta.PartitionId, data.Meta.Offset, data.Value.ToUtf8String());
}
});
//创建一个生产者发消息
var producer = new Producer(new BrokerRouter(options))
{
BatchSize = 100,
BatchDelayTime = TimeSpan.FromMilliseconds(2000)
};
Console.WriteLine("打出一条消息按 enter...");
while (true)
{
var message = Console.ReadLine();
if (message == "quit") break;
if (string.IsNullOrEmpty(message))
{
//
SendRandomBatch(producer, topicName, 200);
}
else
{
producer.SendMessageAsync(topicName, new[] { new Message(message) });
}
}
//释放资源
using (producer)
{
}
}
private static async void SendRandomBatch(Producer producer, string topicName, int count)
{
//发送多个消息
var sendTask = producer.SendMessageAsync(topicName, Enumerable.Range(0, count).Select(x => new Message(x.ToString())));
Console.WriteLine("传送了 #{0} messages. Buffered:{1} AsyncCount:{2}", count, producer.BufferCount, producer.AsyncCount);
var response = await sendTask;
Console.WriteLine("已完成批量发送: {0}. Buffered:{1} AsyncCount:{2}", count, producer.BufferCount, producer.AsyncCount);
foreach (var result in response.OrderBy(x => x.PartitionId))
{
Console.WriteLine("主题:{0} PartitionId:{1} Offset:{2}", result.Topic, result.PartitionId, result.Offset);
}
}
}