§sbt Cookbook
§Hooking into Play’s dev mode
When Play runs in dev mode, that is, when using sbt run
, it’s often useful to hook into this to start up additional processes that are required for development. This can be done by defining a PlayRunHook
, which is a trait with the following methods:
beforeStarted(): Unit
- called before the play application is started, but after all “before run” tasks have been completed.afterStarted(addr: InetSocketAddress): Unit
- called after the play application has been started.afterStopped(): Unit
- called after the play process has been stopped.
Now let’s say you want to build a Web application with grunt
before the application is started. First, you need to create a Scala object in the project/
directory to extend PlayRunHook
. Let’s name it Grunt.scala
:
import play.sbt.PlayRunHook
import sbt._
import scala.sys.process.Process
object Grunt {
def apply(base: File): PlayRunHook = {
object GruntProcess extends PlayRunHook {
override def beforeStarted(): Unit = {
Process("grunt dist", base).run
}
}
GruntProcess
}
}
Now you can register this hook in build.sbt
:
PlayKeys.playRunHooks += Grunt(baseDirectory.value)
This will execute the grunt dist
command in baseDirectory
before the application is started whenever you run sbt run
.
Now we want to modify our run hook to execute the grunt watch
command to observe changes and rebuild the Web application when they happen, so we’ll modify the Grunt.scala
file we created before:
import play.sbt.PlayRunHook
import sbt._
import java.net.InetSocketAddress
import scala.sys.process.Process
object Grunt {
def apply(base: File): PlayRunHook = {
object GruntProcess extends PlayRunHook {
var watchProcess: Option[Process] = None
override def beforeStarted(): Unit = {
Process("grunt dist", base).run
}
override def afterStarted(addr: InetSocketAddress): Unit = {
watchProcess = Some(Process("grunt watch", base).run)
}
override def afterStopped(): Unit = {
watchProcess.map(p => p.destroy())
watchProcess = None
}
}
GruntProcess
}
}
Now when the application is started using sbt run
, grunt watch
will be executed to rerun the grunt build whenever files change.
§Add compiler options
For example, you may want to add the feature flag to have details on feature warnings:
[info] Compiling 1 Scala source to ~/target/scala-2.10/classes...
[warn] there were 1 feature warnings; re-run with -feature for details
Simply add -feature
to the scalacOptions
attribute:
scalacOptions += "-feature"
§Add additional asset directory
For example you can add the pictures
folder to be included as an additional asset directory:
unmanagedResourceDirectories in Assets += baseDirectory.value / "pictures"
This will allow you to use routes.Assets.at
with this folder.
§Disable documentation
To speed up compilation you can disable documentation generation:
sources in (Compile, doc) := Seq.empty
publishArtifact in (Compile, packageDoc) := false
The first line will disable documentation generation and the second one will avoid to publish the documentation artifact.
§Configure ivy logging level
By default ivyLoggingLevel
is set on UpdateLogging.DownloadOnly
. You can change this value with:
UpdateLogging.Quiet
only displays errorsUpdateLogging.Full
logs the most
For example if you want to only display errors:
ivyLoggingLevel := UpdateLogging.Quiet
§Fork and parallel execution in test
By default parallel execution is disabled and fork is enabled. You can change this behavior by setting parallelExecution in Test
and/or fork in Test
:
parallelExecution in Test := true
fork in Test := false
Next: Debugging your build
Found an error in this documentation? The source code for this page can be found here. After reading the documentation guidelines, please feel free to contribute a pull request. Have questions or advice to share? Go to our community forums to start a conversation with the community.