\documentclass{article}

\usepackage{../style}

\usepackage{../langs}

\usepackage{marvosym}

%cheat sheet

%http://worldline.github.io/scala-cheatsheet/

\begin{document}

\fnote{\copyright{} Christian Urban, King's College London, 2020}

\section*{Scala in 6CCS3CFL}

For the coursework in this module you are free to use any programming

language you like, but I will show you all my code using Scala---I

\noindent

The main difference to the Scala I showed you in PEP is that in CFL

I will use the Ammonite REPL

\begin{quote}

\url{https://ammonite.io/#Ammonite-REPL}

\end{quote}

\noindent

This is a drop-in replacement for the original Scala REPL and

works very similarly, for example

\begin{lstlisting}[language={},numbers=none,basicstyle=\ttfamily\small]

$ amm

Loading...  

res0: Int = 3

\end{lstlisting} %% $

\noindent

Ammonite uses the same Scala compiler, just adds some useful features

on top of it. It is quite main-stream in the Scala community and it should

therefore be very easy for you to install \texttt{amm}.

The big advantage of Ammonite is that it comes with some additional

libraries already built-in and also allows one to easily break up code

into smaller modules. For example reading and writing files in

Ammonite can be achieved with

\begin{lstlisting}[numbers=none,language=Scala]

scala> import ammonite.ops._

scala> read(pwd / "file.name")   

res1: String = """..."""

scala> write.over(pwd / "file.name", "foo bar")

\end{lstlisting}

\noindent

The latter writes the string \code{"foo bar"} into the file

\code{"file.name"}, which is located in the current working

directory. For loading and accessing code from another Scala file, you

can import it as follows:

\begin{lstlisting}[numbers=none,language=Scala]

import $file.name-of-the-file

import name-of-the-file._

\end{lstlisting}  %% $

\noindent

This assumes the other Scala file is called

\texttt{name-of-the-file.sc} and requires the file to be in the same

directory where \texttt{amm} is working in. This will be very convenient 

for the compiler we implement in CFL, because it allows us to easily

break up the code into the lexer, parser and code generator.

Another feature which exists in Ammonite, but not yet in the

current version of Scala (it will be in the next version called dotty)

is that you can mark functions as \texttt{@main}. For example

\begin{lstlisting}[numbers=none,language=Scala]

@main

def foo() = ...

\end{lstlisting}

\noindent

This means you can now call that function from the command line like

\begin{lstlisting}[numbers=none,language=Scala]

$ amm file.sc foo

\end{lstlisting} %% $

\noindent

If you want to specify an argument on the commandline, say an int and

a string, then you can write

\begin{lstlisting}[numbers=none,language=Scala]

@main

def bar(i: Int, s: String) = ...

\end{lstlisting}

\noindent

and then call

\begin{lstlisting}[numbers=none,language=Scala]

$ amm file.sc 42 foobar

\end{lstlisting} %% $

\noindent

What is also good in Ammonite that you can specify more than one

function to be ``main'' and then specify on the command line which

function you want to run as entry-point.

Another feature you might like to use is that Ammonite can ``watch'' files.

This means it can automatically re-run a file when it is saved. For this

you have to call \texttt{amm} with the option \texttt{-w}, as in

\begin{lstlisting}[numbers=none,language=Scala]

$ amm -w file.sc

\end{lstlisting} %% $

\noindent Of course this requires that you use \texttt{println} for

inspecting any data, as otherwise nothing will be displayed at the

commandline.

\bigskip

\noindent

To sum up, Ammonite is a really useful addition to the Scala ecosystem.

You can find more information about how to use it in the first five chapters

of the ``Hands-on Scala'' book by Li Haoyi. These chapters are

free and can be used as a reference, see:

\begin{center}

\url{https://www.handsonscala.com/part-i-introduction-to-scala.html}

\end{center}

\end{document}

%%% Local Variables: 

%%% mode: latex

%%% TeX-master: t

%%% End: