sbt-avrohugger

sbt plugin for generating Scala case classes and ADTs from Apache Avro schemas, datafiles, and protocols.

Install the plugin

Add the following lines to the file myproject/project/plugins.sbt in your project directory:

addSbtPlugin("com.julianpeeters" % "sbt-avrohugger" % "0.15.1")

Import the plugin settings

To activate the plugin, import its settings by adding one of the following lines to your myproject/build.sbt file:

To get the 'generate' task for generating standard Scala Case Classes use:

sbtavrohugger.SbtAvrohugger.avroSettings

To get the generate-specific task for generating Scala Case Classes that are compatible with the Avro Specific API, use:

sbtavrohugger.SbtAvrohugger.specificAvroSettings

To get the generate-scavro task for generating Scala Case Class wrapper classes (Java generated classes supplied separately) for use with Scavro, use:

sbtavrohugger.SbtAvrohugger.scavroSettings

Scope

All settings and tasks are in the avro scope. E.g., to execute the generate task directly, just run avro:generate.

Settings

Name	Name in shell	Default	Description
sourceDirectory	source-directory	src/main/avro	Path containing *.avsc, *.avdl, and/or *.avro files.
scalaSource	scala-source	$sourceManaged/main/compiled_avro	Path for the generated *.scala or *.java files.
avroScalaCustomTypes	avro-scala-custom-types	Map.empty[String, Class[_]]	Map for reassigning array to Array, List, or Seq.
avroScalaCustomNamespace	avro-scala-custom-namespace	Map.empty[String, String]	Map for reassigning namespaces.
avroScalaCustomEnumStyle	avro-scala-custom-enum-style	Map.empty[String, String]	Map for reassigning enum style to java enum or case object.

Changing Settings

Settings can be overridden by adding lines to myproject/build.sbt:

(scalaSource in avroConfig) := new java.io.File("myscalaSource")

avro-scala-custom-types and avro-scala-custom-namespace require additional imports (I'm not sure why these aren't picked up with the other settings, anybody know why we have to import them separately?):

import sbtavrohugger.AvrohuggerSettings.{
  avroScalaCustomTypes,
  avroScalaCustomNamespace,
  avroScalaCustomEnumStyle
}

(avroScalaCustomTypes in avroConfig) := Map("array"->classOf[Array[_]])

(avroScalaCustomNamespace in avroConfig) := Map("example"->"overridden")

(avroScalaCustomEnumStyle in avroConfig) := Map("enum"->"java enum")

Tasks

Each task is automatically executed every time the project is compiled.*

• as of Intellij IDEA 14.1.4, and possibly for other IDEs, the task must be run manually from the integrated Terminal: avro:generate.

Name	Name in shell	Description
generate	generate	Compiles the Avro files into Scala case classes.
generateSpecific	generate-specific	Compiles the Avro files into Scala case classes implementing SpecificRecord.
generateScavro	generate-scavro	Compiles the Avro files into Scala case class Scavro wrapper classes.

Datatypes

Supports generating case classes with arbitrary fields of the following datatypes:

• INT → Int
• LONG → Long
• FLOAT → Float
• DOUBLE → Double
• STRING → String
• BOOLEAN → Boolean
• NULL → Null
• MAP → Map
• ENUM → generate: scala.Enumeration, generate-specific: Java Enum
• BYTES → Array[Byte]
• FIXED → //TODO
• ARRAY → List (generate-scavro: Array). To reassign, please see Settings above.
• UNION → Option
• RECORD → case class

Future

• support for more avro datatypes
• decimal support for avrohugger via logical types
• more codegen situations, e.g. exploding Spark Rows?

Credits

sbt-avrohugger is based on sbt-avro by Juan Manuel Caicedo, and depends on avrohugger.

Contributors

• Marius Soutier
• Brennan Saeta
• Daniel Lundin
• Vince Tse
• Jerome Wacongne
• Ryan Koval
• Saket
• Jon Morra

Fork away, just make sure the tests pass before you send a pull request.

Criticism is appreciated.