celery是python实现的分布式调度框架,有时候想用celery去调用java服务,正好有一个celery-java的库可以使用,能达到这个效果,记录一下:
先添加依赖:
com.geneea.celery
celery-java
1.10.4
先定义任务类TestTask.java
import com.geneea.celery.CeleryTask;
@CeleryTask
public class TestTask {
public int sum(int x, int y){
return x + y;
}
}
再定义工作入口类MyWorker.java
import com.geneea.celery.CeleryWorkerCLI;
public class MyWorker {
public static void main(String[] args) throws Exception {
//注意broker的url的末尾/必须用转义符%2F
args = new String[]{"--queue","celery",
"--concurrency","2",
"--broker","amqp://guest:guest@localhost:5672/%2F"};
CeleryWorkerCLI.main(args);
}
}
运行MyWorker即可启动java版的celery工作节点:
Started consuming tasks from queue celery.
Known tasks:
- cn.main.celery.TestTask
然后定义celery的任务调度程序test.py:
import celery
app = celery.Celery(broker="amqp://guest:guest@localhost:5672//", backend="rpc://guest:guest@localhost:5672//")
#调度任务格式为pacakage.class#func
result = app.signature("cn.main.celery.TestTask#sum", [1, 2]).delay().get()
print("result is %s" % result)
运行test.py程序反应如下:
C:\ProgramData\Anaconda3\python.exe D:/Test/python/celerys/javatest.py
result is 3
Process finished with exit code 0
19:06:43.685 INFO [pool-1-thread-1] c.g.c.CeleryWorkerCore:126 - CeleryTask cn.main.celery.TestTask#sum[2c1476a0-0d64-47c8-afce-6ddc99915ea9] succeeded in 338.9 μs
参考链接: