softprops / guavapants   0.1.1

MIT License GitHub

one pair of pants that makes both your guava _and_ scala types look good

Scala versions: 2.10

guava pants

Build Status

pants that look good on both your scala and guava types.


This library targets scala 2.10 and higher.

via sbt.

Add the following to your sbt build definition

resolvers += "softprops-maven" at ""

libraryDependencies += "me.lessis" %% "guavapants" % "0.1.1"


This library attempts to address the problem that scala.collection.JavaConverters addresses for resolving interface discrepancies between Scala collections and Java collections, for Guava types when things start getting uncomfortable. Guava pants provides convenient asGuava enrichments on Scala types that map well to equivalent Guava types and asScala enrichments on Guava types that map well to equivalent Scala types.

Most usage follows the same convention as JavaConverters. Just import GuavaConverters into scope.

import guavapants.GuavaConverters._

Then call the asGuava or asScala method on the target value to achieve a more comfortable fit for which ever interface you are trying to satisfy.


Guava defines primitives for Function objects that look very familiar to your everyday Scala functions. They are a bit limited in that they are only defined for functions that take one argument, most likey a design decision for the use case of transforming collections.

Given a scala function, you can switch to a Guava type by calling asGuava on it.

val gf:[Int, String] =
  ((_: Int).toString).asGuava


In Guava, there is a special name attributed to functions that take one argument and return a Boolean value called a Predicate. In Guava the return type is a primitive Boolean type. This works out because Scala's Boolean type can not be null ( try for yourself! ).

Given a scala function returning a Boolean value, you can switch to a Guava type by calling asGuava on it.

val gp:[Int] =
  ((_:Int) % 0 == 0).asGuava


Guava defines a special interface for functions which take no arguments called Suppliers. In Scala, this is just a function that takes no arguments.

Given a scala function which takes no arguments, you can switch to a Guava type by calling asGuava on it.

val gs:[Int] =
  (() => 32).asGuava


Guava shares a primitive similar to Scala's built-in Option type called Optional It serves the same purpose in indicating the presence or absence of some underlying value.

Given a scala Option type, you can convert to a Guava type by calling asGuava on it.

val go:[Int] =


Guava can be a little aggressive in the problems it tries to solve as one library. Among the more interesting things found in its collection are abstractions for Futures which you can register hooks on called ListenableFutures. There is a strong correlation to Scala's built-in Futures in their design.

Given a scala Future, you can convert to a Guava type by calling asGuava on it.

import scala.concurrent.Future
val gf:[Int] =


Things you do not need to know but may appreciate.

Value Types

Guava pants was designed to be as efficient as possible at runtime. In java code bases where Guava is used, function interfaces are often exposed prolifically ( with good reason ). If you are calling into these interfaces from Scala you will need be interfacing with them a lot at runtime. The types of conversions this library does (implicit conversions) between types has historically had a less than desirable runtime cost. In scala 2.10, there was a new kind of type, a value type that was introduced to help reduce and eliminate this runtime cost. Guava pants wholeheartedly was designed to take advantage of this Scala 2.10 feature.


You may also want to consider using the bijection guava module.


Did I miss something? Let me know.

Doug Tangren (softprops) 2014