上篇我们讨论了静态数据源(Static Source, snapshot)。这种方式只能在预知数据规模有限的情况下使用,对于超大型的数据库表也可以说是不安全的资源使用方式。Slick3.x已经增加了支持Reactive-Streams功能,可以通过Reactive-Streams API来实现有限内存空间内的无限规模数据读取,这正符合了FunDA的设计理念:高效、便捷、安全的后台数据处理工具库。我们在前面几篇讨论里介绍了Iteratee模式,play-iteratees支持Reactive-Streams并且提供与Slick3.x的接口API,我们就在这篇讨论里介绍如何把Slick-Reactive-Streams转换成fs2-Streams。根据Slick官方文档:Slick可以通过db.stream函数用Reactive-Stream方式来读取后台数据,具体的配置如下:
val disableAutocommit = SimpleDBIO(_.connection.setAutoCommit(false))
val action = queryAction.withStatementParameters(fetchSize = 512)
val publisher = db.stream(disableAutocommit andThen action)
首先,我们需要取消自动提交(disableAutocommit)。fetchSize是缓存数据页长度(每批次读取数据字数),然后用db.stream来构成一个Reactive-Streams标准的数据源publisher。Slick官方网页只提供了下面这个使用publisher的例子:
val fut = publisher.foreach(s => println(s))
Await.ready(fut,Duration.Inf)
除了数据枚举外就没什么用处,也无法提供更细节点的示范。FunDA的具体解决方案是把publisher转换成play-iteratee的Enumerator。play-iteratee支持Reactive-Streams,所以这个Enumerator应该具备协调后台数据和内存缓冲之间关系(back-pressure)的功能。play-iteratee是如下构建Enumerator的;
import play.api.libs.iteratee._
val enumerator = streams.IterateeStreams.publisherToEnumerator(publisher)
enumerator从后台数据库表中产生的数据源通过Iteratee把数据元素enqueue推送给一个fs2的queue:
private def pushData[R](q: async.mutable.Queue[Task,Option[R]]): Iteratee[R,Unit] = Cont {
case Input.EOF => {
q.enqueue1(None).unsafeRun
Done((), Input.Empty)
}
case Input.Empty => pushData(q)
case Input.El(e) => {
q.enqueue1(Some(e)).unsafeRun
pushData(q)
}
}
然后fs2进行dequeue后生成fs2的Stream:
Stream.eval(async.boundedQueue[Task,Option[SOURCE]](queSize)).flatMap { q =>
Task { Iteratee.flatten(enumerator |>> pushData(q)).run }.unsafeRunAsyncFuture()
pipe.unNoneTerminate(q.dequeue)
}
整个构建Stream的过程在FunDA的fdasources包是这样定义的:
package com.bayakala.funda.fdasources
import fs2._
import play.api.libs.iteratee._
import com.bayakala.funda.fdapipes._
import slick.driver.JdbcProfile
object FDADataStream {
class FDAStreamLoader[SOURCE, TARGET](slickProfile: JdbcProfile, convert: SOURCE => TARGET) {
import slickProfile.api._
def fda_typedStream(action: DBIOAction[Iterable[SOURCE],Streaming[SOURCE],Effect.Read])(slickDB: Database)(fetchSize: Int, queSize: Int): FDAPipeLine[TARGET] = {
val disableAutocommit = SimpleDBIO(_.connection.setAutoCommit(false))
val action_ = action.withStatementParameters(fetchSize = fetchSize)
val publisher = slickDB.stream(disableAutocommit andThen action)
val enumerator = streams.IterateeStreams.publisherToEnumerator(publisher)
Stream.eval(async.boundedQueue[Task,Option[SOURCE]](queSize)).flatMap { q =>
Task { Iteratee.flatten(enumerator |>> pushData(q)).run }.unsafeRunAsyncFuture()
pipe.unNoneTerminate(q.dequeue).map {row => convert(row)}
}
}
def fda_plainStream(action: DBIOAction[Iterable[SOURCE],Streaming[SOURCE],Effect.Read])(slickDB: Database)(fetchSize: Int, queSize: Int): FDAPipeLine[SOURCE] = {
val disableAutocommit = SimpleDBIO(_.connection.setAutoCommit(false))
val action_ = action.withStatementParameters(fetchSize = fetchSize)
val publisher = slickDB.stream(disableAutocommit andThen action)
val enumerator = streams.IterateeStreams.publisherToEnumerator(publisher)
Stream.eval(async.boundedQueue[Task,Option[SOURCE]](queSize)).flatMap { q =>
Task { Iteratee.flatten(enumerator |>> pushData(q)).run }.unsafeRunAsyncFuture()
pipe.unNoneTerminate(q.dequeue)
}
}
private def pushData[R](q: async.mutable.Queue[Task,Option[R]]): Iteratee[R,Unit] = Cont {
case Input.EOF => {
q.enqueue1(None).unsafeRun
Done((), Input.Empty)
}
case Input.Empty => pushData(q)
case Input.El(e) => {
q.enqueue1(Some(e)).unsafeRun
pushData(q)
}
}
}
object FDAStreamLoader {
def apply[SOURCE, TARGET](slickProfile: JdbcProfile, converter: SOURCE => TARGET): FDAStreamLoader[SOURCE, TARGET] =
new FDAStreamLoader[SOURCE, TARGET](slickProfile, converter)
}
}
FDADataStream对象内主要实现了fda_typedStream和fda_plainStream。fda_typedStream提供了SOURCE=>TARGET的转换。从Enumerator转换到Stream整个过程和原理我们在FunDA(7)里已经详细介绍过了。下面我们看看FunDA-Example中fda_typedStream的具体应用例子:
package com.bayakala.funda.fdasources.examples
import slick.driver.H2Driver.api._
import com.bayakala.funda.fdasources.FDADataStream._
import com.bayakala.funda.samples._
import com.bayakala.funda.fdarows._
import com.bayakala.funda.fdapipes._
import FDANodes._
import FDAValves._
object Example2 extends App {
val albums = SlickModels.albums
val companies = SlickModels.companies
//数据源query
val albumsInfo = for {
(a,c) <- albums join companies on (_.company === _.id)
} yield (a.title,a.artist,a.year,c.name)
//query结果强类型(用户提供)
case class Album(title: String, artist: String, year: Int, publisher: String) extends FDAROW
//转换函数(用户提供)
def toTypedRow(row: (String, String, Option[Int], String)): Album =
Album(row._1, row._2, row._3.getOrElse(2000), row._4)
val db = Database.forConfig("h2db")
val streamLoader = FDAStreamLoader(slick.driver.H2Driver, toTypedRow _)
val albumStream = streamLoader.fda_typedStream(albumsInfo.result)(db)(512,128)
//定义一个用户作业函数:列印数据内容
def printAlbums: FDATask[FDAROW] = row => {
row match {
case album: Album =>
println("____________________")
println(s"品名:${album.title}")
println(s"演唱:${album.artist}")
println(s"年份:${album.year}")
println(s"发行:${album.publisher}")
fda_next(album)
case _ => fda_skip
}
}
albumStream.through(fda_execUserTask(printAlbums)).run.unsafeRun
}
运算结果:
品名:Keyboard Cat's Greatest Hits
演唱:Keyboard Cat
年份:1999
发行:Sony Music Inc
____________________
品名:Spice
演唱:Spice Girls
年份:1999
发行:Columbia Records
____________________
品名:Whenever You Need Somebody
演唱:Rick Astley
年份:1999
发行:Sony Music Inc
____________________
品名:The Triumph of Steel
演唱:Manowar
年份:1999
发行:The K-Pops Singers
____________________
品名:Believe
演唱:Justin Bieber
年份:1999
发行:Columbia Records
Process finished with exit code 0