sbt编译需要很长时间才能完成。

Question

我现在使用Akka、Play Framework、Scala和构建工具SBT运行一个相对较小的项目。然而，我已经建立了一个比这个更大的项目，它已经建立了相当快。但是现在，经过一段时间之后，我正在尝试构建我的项目，而且编译需要很长时间(总时间: 904 s)。

> ~/nlp-search/code(branch:feature/demo-app-api*) » sbt compile         
> renienj@Rajani [info] Loading project definition from
> /Users/renienj/nlp-search/code/project [info] Set current project to
> nlp-search (in build file:/Users/renienj/nlp-search/code/) [info]
> Compiling 115 Scala sources and 1 Java source to
> /Users/renienj/nlp-search/code/target/scala-2.11/classes... [warn]
> /Users/renienj/nlp-search/code/app/api/parsedquery/ParsedQuery.scala:14:
> imported `format' is permanently hidden by definition of value format
> in object ParsedQuery [warn]   import Sentence.format [warn]          
> ^ [warn]
> /Users/renienj/nlp-search/code/app/api/parsedquery/Sentence.scala:14:
> imported `format' is permanently hidden by definition of value format
> in object Sentence [warn]   import Phrase.format [warn]               
> ^ [warn]
> /Users/renienj/nlp-search/code/app/api/parsedquery/Sentence.scala:15:
> imported `format' is permanently hidden by definition of value format
> in object Sentence [warn]   import  Filter.format [warn]              
> ^ [warn]
> /Users/renienj/nlp-search/code/app/api/parsedquery/Sentence.scala:16:
> imported `format' is permanently hidden by definition of value format
> in object Sentence [warn]   import Order.format [warn]               
> ^ [warn]
> /Users/renienj/nlp-search/code/app/backend/intelligenceServices/nlp/executor/StreamExecutor.scala:23:
> abstract type pattern OT is unchecked since it is eliminated by
> erasure [warn]     case x: OT => [warn]             ^ [warn]
> /Users/renienj/nlp-search/code/app/backend/intelligenceServices/merchRules/io/DAO.scala:27:
> method newTermName in trait Names is deprecated: Use TermName instead
> [warn]       String.valueOf(x.map(x => x.toChar)).unpickle[MerchRule]
> [warn]                                                    ^ [warn]
> /Users/renienj/nlp-search/code/app/backend/solrAccess/cache/SolrCacheProvider.scala:24:
> method newTermName in trait Names is deprecated: Use TermName instead
> [warn]       x.pickle.value [warn]         ^ [warn]
> /Users/renienj/nlp-search/code/app/backend/solrAccess/cache/SolrCacheProvider.scala:30:
> method newTermName in trait Names is deprecated: Use TermName instead
> [warn]       BinaryPickle(x).unpickle[DocumentList] [warn]            
> ^ [warn] there were 9 feature warnings; re-run with -feature for
> details [warn] 9 warnings found [success] Total time: 904 s, completed
> Apr 9, 2016 5:07:09 PM

我的build.sbt：

name := """nlp-search"""

version := """3.0-RC1"""

lazy val root = (project in file(".")).enablePlugins(PlayScala)

scalaVersion := "2.11.7"

resolvers += Resolver.mavenLocal

val akkaVersion = "2.3.12"

libraryDependencies ++= Seq(
  "com.typesafe.akka"     %% "akka-actor"       % akkaVersion,
  "com.typesafe.akka"     %% "akka-contrib"     % akkaVersion,
  "com.typesafe.akka"     %% "akka-remote"      % akkaVersion,
  "com.typesafe.akka"     %% "akka-cluster"     % akkaVersion,
  "com.typesafe.akka"     %% "akka-testkit"     % akkaVersion,
  "org.apache.solr"       % "solr-solrj"        % "4.10.0",
  "org.apache.thrift"     % "libthrift"         % "0.9.2",
  "org.json"              % "json"              % "20080701",
  "org.skife.com.typesafe.config" % "typesafe-config" % "0.3.0",
  "net.debasishg"         %% "redisclient"      % "2.13",
  "org.scala-lang.modules" %% "scala-pickling"  % "0.10.0",
  "edu.stanford.nlp"      % "stanford-parser"   % "3.5.2",
  "edu.stanford.nlp" % "stanford-corenlp" % "3.5.2",
  "org.scalatest"         % "scalatest_2.11"    % "2.2.4" % "test",
  "com.kohls.search.qp"   % "api"               % "1.0.0",
  "com.typesafe.akka"     %% "akka-stream-experimental" % "1.0"
)

libraryDependencies += specs2 % Test

resolvers += "scalaz-bintray" at "https://dl.bintray.com/scalaz/releases"

publishArtifact in (Compile, packageDoc) := false

publishArtifact in packageDoc := false

sources in (Compile,doc) := Seq.empty

从早上开始我一直在挣扎。我真的不明白为什么要花这么长时间。

先谢谢你。

Answer 1

不是真正的答案，但也许是暗示…如果您有很多由宏应用程序(通常是您自己的案例类的JSON阅读器和作者)生成的关联，那么编译速度可能会大大减慢。您可以尝试用这样的方法(例如，upickle)来缓存植入者：

object JsonSerialization {
  implicit lazy val userReader = { val userReader = (); macroR[User] }
  implicit lazy val userWriter = { val userWriter = (); macroR[Writer] }
}

您可能还想看看无形的cachedImplicit，这在this question中也提到过。