Oracle Slick Driver是否支持Streaming?

时间:2017-09-12 22:09:31

标签: scala akka slick akka-stream slick-3.0

我在Scala 2.11.8上使用Slick 3.1.0。这是使用光滑的扩展来查询Oracle数据库。

我写了以下代码

type MyTup = (String, String, String)
implicit val actorSystem = ActorSystem()
implicit val materializer = ActorMaterializer()
val db = Database.forConfig("oracle")
val query = sql"""select col1, col2, col3 from FOO.BAR""".as[MyTup]
val source = Source.fromPublisher[MyTup](db.stream(query))
val sink = Sink.foreach[MyTup](row => println(row))
val flow = source to sink
flow.run()
readLine()
db.close()

当我运行此代码时,此代码运行很长时间而不打印任何内容....这似乎意味着它试图在打印之前读取整个结果(因此它不是流式传输)。

我已经在MySQL上完成了流式传输,我知道它需要特殊配置来启用流式传输

https://github.com/slick/slick/issues/1218

oracle是否还需要一些特殊的配置,比如MySql?

我的build.sbt

libraryDependencies ++= Seq(
   "com.typesafe.slick" % "slick-extensions_2.11" % "3.1.0",
   "com.typesafe.slick" % "slick_2.11" % "3.1.0",
   "com.typesafe.slick" % "slick-codegen_2.11" % "3.1.0",
   "oracle" % "ojdbc7_2.11" % "7.0.0",
   "com.typesafe.slick" % "slick-hikaricp_2.11" % "3.1.0",
   "com.typesafe.akka" % "akka-stream_2.11" % "2.5.4"
)

1 个答案:

答案 0 :(得分:0)

行。我找到了答案。是的Oracle驱动程序确实支持流式传输,它不需要像mysql那样的任何额外配置。

这是我的application.conf

oracle {
  driver = "slick.jdbc.OracleProfile$"
  dataSourceClass = "slick.jdbc.DatabaseUrlDataSource"
  properties {
    driver = "oracle.jdbc.OracleDriver"
    url = "jdbc:oracle:thin:user/password@//server:1521/serviceName"
  }
  numThreads = 10
  connectionTimeout = 10000
}

我的build.sbt

"com.typesafe.slick" %% "slick" % "3.2.1",
"com.typesafe.slick" %% "slick-hikaricp" % "3.2.1",
"oracle" % "ojdbc7_2.11" % "7.0.0",
"com.typesafe.akka" %% "akka-stream" % "2.5.4"

请注意,我没有使用光滑的扩展名,因为它们已经过时了。新的Oracle驱动程序(OracleProfile)位于光滑的核心版本中。

这里唯一的技巧是条目"oracle" % "ojdbc7_2.11" % "7.0.0"是假的。因为我下载了jar并在本地sbt中发布了它。 (oracle不提供oracle驱动程序作为sbt / maven依赖)

现在这是我的代码,它可以非常精确地传输数据:)

implicit val actorSystem = ActorSystem()
implicit val actorMaterializer = ActorMaterializer()
val db = Database.forConfig("oracle")
val query = sql"select member_id from foo".as[Long]
val publisher = db.stream(query)
val source = Source.fromPublisher[Long](publisher)
val sink = Sink.foreach[Long](x => println(x))
val graph = RunnableGraph.fromGraph(GraphDSL.create(sink){implicit builder => s =>
  import GraphDSL.Implicits._
  source ~> s.in
  ClosedShape
})
val future = graph.run()
Await.result(future, Duration.Inf)
db.close()