[r-t] Scala, and its application to ringing problems

[Mark Davies]

In the investigation of Philip's Helixoid-based extents, I've been using 
a language called Scala. It is a relatively new language, although now 
pretty mature, which combines object-oriented programming in the style 
of Java with the functional approach championed by languages such as 
Lisp, Haskell, Clojure and Erlang. It might seem madness that these two 
very different paradigms can be made to work together, but Scala not 
only succeeds in this challenge, but makes it look and feel like the 
most natural thing in the world.

I have been dabbling with Scala for about 18 months, but this is the 
first time I have used it for a serious exercise in composition. It has 
proved extremely well suited to the rapid prototyping and good execution 
speeds required for composing tasks. In fact, I'm so taken with it that 
I thought it worth disseminating my experiences to the ringing/composing 
community.

The website for the Scala language is here:

<http://www.scala-lang.org/>

If you are interested in it, or indeed interested in programming in 
general, you can do no better than read the Scala book, which as well as 
teaching you the language happens in my view to be the best book on 
programming ever written. It is an absolutely cracking read, and there 
is not much which you need to know about programming which it does not 
contain, nor which it does not elucidate beautifully.

<http://www.amazon.co.uk/Programming-In-Scala-2nd-Edition/dp/0981531644/ref=sr_1_1?s=books&ie=UTF8&qid=1340897990&sr=1-1>

So how is this language useful to the change-ringing composer? I'd like 
to show some examples of how simple, concise and quick Scala can be, 
lifted from the code I wrote for the Helixoid project. As explained 
yesterday, the project started off with a simple search for Helixoid 
methods, but soon evolved into a fairly complex series of pipelines 
which re-processed those methods already found, in order to extract 
candidates which passed a series of requirement filters, including (a) 
having no negative cross-falseness with Plain Bob, and (b) possessing 
interesting course-splices. In all these tasks Scala proved beautifully 
suitable and efficient. It enabled me to try things rapidly and make 
progress in places where, with other languages, the coding barrier might 
have been too high for my limited spare time.

I will break down my discussion into a series of articles in order to 
reduce the length of each email. In the first article, which I'll send 
shortly, I'll show how Scala's functional programming style can 
dramatically simplify the job of writing machine searches. Happy to take 
questions, and to hear anyone else's experience of using the language.

MBD