johan-- / elixir_wordsets

Elixir program that uses parallel processing to discover symmetrical word combinations

Overview

This Elixir program implements a parallelized algorithm to discover word combinations that, when written on separate lines, read the same both horizontally and vertically. For example:

J O S H
O B O E
S O F A
H E A D

Read an article about this program on my blog.

How to run the program

Assuming you have Elixir installed, open Terminal in this project's root directory and load its mix.exs file into the interactive Elixir console (IEx):

$ iex -S mix

Next set up the processing environment…

iex(1)> Wordsets.init  # initialize the Lexicon and Cache processes 
Loading words…
Indexed lexicon in 6.7 seconds.
:ok

Now search for wordsets that start with a certain word, such as elixir.

iex(2)> Wordsets.find "elixir"                                     
elixir
lately
itonia
xenian
iliahi
ryania

elixir
linene
inanga
xenial
ingate
reales

found 2 in 0.774412 seconds
:ok

Replace "elixir" with any word you like. Be aware, finding all available wordsets for some words is very fast, but for other words it can take a few seconds. I heavily optimized the search algorithm (see the Engine module) and data model (see the Lexicon module), but natural language is inherently irregular.

Compatibility

This program has only been tested on OS X. Its source of words is the file /usr/share/dict/words which is included in the operating system installation by default. Windows users will need to download the file and adjust the @file_path value in lexicon.ex.

The code was written and tested with Elixir v1.0.3