neoremind / kraps-rpc Goto Github PK

I find usage of them at

new NettyRpcEnv(sparkConf, javaSerializerInstance, config.advertiseAddress,
    config.securityManager, config.numUsableCores)
val startNettyRpcEnv: Int => (NettyRpcEnv, Int) = { actualPort =>
     nettyEnv.startServer(config.bindAddress, actualPort)

It seems advertiseAddress is used to construct RpcAddress while bindAddress is used to start server. Why are the two address needed?

Thanks :)

in some cases, we need return a set of results on server side, for example, executing a JDBC query:

val rs = con.execute("select * from person")
while(rs.hasNext){
  val result = rs.next
  ...
}

how to use kraps-rpc to handle this application case?
should I send a set of requests(with cursor information) one by one on client side?

needs your kindly help!

Do you have any versions for scala 2.12?
I want to use kraps in my project, which depends on Scala2.12 :(

1.3.1 Asynchronous invocation Client in README.md
val endPointRef: RpcEndpointRef = rpcEnv.setupEndpointRef(RpcAddress("localhost", 52345), "hell-service")
should not be
val endPointRef: RpcEndpointRef = rpcEnv.setupEndpointRef(RpcAddress("localhost", 52345), "hello-service") ?