I decided to kick off the first package in my FiniteDifferences project on an easy note — implementing a class for polynomial functions. After writing this class, I wanted to import it into the Scala REPL so that I could test user friendliness and identify bugs. This process was more difficult than I anticipated, so I wanted to capture the method(s) I ended up using here.

For those who don’t know, REPL stands for Read-Eval-Print-Loop. A REPL is a programming environment that takes single user inputs, evaluates them, and returns the result to the user. The command-line Scala REPL can be accessed by opening a terminal and typing “scala” (assuming it is installed on your machine). Why use it? It’s a very fast way to quickly test code segments and run small experiments (and is one of Alvin Alexander’s best practices).

Two Methods

Note: The package I am writing is called com.ztbarry.nicefunctions and the class I wanted to import is called Polynomial.

Method 1: Compile and Import

This guide shows how to compile a class via the command line and then import it into a Scala REPL. I’ve adopted it to my Polynomial class here:

1) Navigate to the folder that contains the package’s classes. In this case my package is called nicefunctions and I can navigate to it’s folder by typing

$ cd ~/Documents/FiniteDifferences/src/main/scala/com/ztbarry/nicefunctions

2) Compile the class using scalac. This part is easy:

$ scalac Polynomial.scala

3) Start the REPL with the folder that contains the class in classpath. Classpath tells the compiler where to look for additional classes and can be specified by typing -cp pathtofolder after scala. Since I navigated to the nicefunctions folder in Step 1, I can reference that folder with a dot . on the command line.

$ scala -cp .

4) Import the class. It is important to note that the formal name of my package is com.ztbarry.nicefunctions, not simply nicefunctions. So to import the Polynomial class I need to type:

scala> import com.ztbarry.nicefunctions.Polynomial

5) Use the class!. I’ll create the polynomial \(4.2 + 2.1x + 0.7x^2\):

scala> val parabola = new Polynomial(Vector[Double](4.2, 2.1, 0.7))

Method 2: SBT

I was unable to get the above method to work initially because I was trying to import all of nicefunctions by typing import com.ztbarry.nicefunctions instead of typing import com.ztbarry.nicefunctions._ (this silly mistake cost me hours!). Luckily, I posted to stackoverflow and someone was kind enough to help.

They pointed me towards using the command sbt console, which “load[s] up the same REPL environment as $ scala but with the addition of all of the project code” (source). Many of the same steps as Method 1 apply, but I’ll list all of the steps for this method for clarity:

1) Navigate to the folder that contains the package’s classes. In this case my package is called nicefunctions and I can navigate to it’s folder by typing

$ cd ~/Documents/FiniteDifferences/src/main/scala/com/ztbarry/nicefunctions

2) Load SBT console.

$ sbt console

3) Import the class. It is important to note that the formal name of my package is com.ztbarry.nicefunctions, not simply nicefunctions. So to import the Polynomial class I need to type:

scala> import com.ztbarry.nicefunctions.Polynomial

4) Use the class!. I’ll create the polynomial \(4.2 + 2.1x + 0.7x^2\):

scala> val parabola = new Polynomial(Vector[Double](4.2, 2.1, 0.7))

Note: I could have run sbt console from within the FiniteDifferences directory to include the project code for all packages within the project, including com.ztbarry.nicefunctions. This is the method I’ll be using going forward.

Parting Thoughts

When you’re learning a new language or technology, some things that seem simple can take a lot longer than anticipated. This is all part of the learning experience. Often times, the extra time leads to additional learnings. If I had typed import com.ztbarry.nicefunctions._ instead of import com.ztbarry.nicefunctions when trying Method 1 for the first time, I wouldn’t have found Method 2 as quickly - even though it is much more useful. Hopefully this post helps someone get moving faster!