Practical Functional Programming in Ruby 

~ NDA ~ ! If you meet Diego Echeverri, please praise him for the code he has written. By the way, don’t show this code to anyone else, it is not my code. __________________________________ (Date, Signature) creator of awesome things Ruby 

The Problem: ! Detecting Cheaters 

No content

Many different attributes… ! Score, Mouse Movement, Click Rate, … 

A Solution: ! Classifiers 

ScoreCheatClassifier class ScoreCheatClassifier < CheaterClassifier SCORE_FILTER_MAX = 2_000_000 SCORE_FILTER_MIN = 1_400_000 ! def initialize(last_score) @last_score = last_score end ! def cheater? if @last_score > SCORE_FILTER_MAX :yes elsif @last_score < SCORE_FILTER_MIN :no else :maybe end end end 

LevelClassifier class LevelClassifier < CheaterClassifier BASE_LEVEL = 50 BASE_SCORE = 1_700_000 ! def initialize(score, level) @score = score @level = level end ! def cheater? if @score > BASE_SCORE :yes else @level >= BASE_LEVEL ? :maybe : :yes end end end 

Hooking them together ! in a lazy fashion 

PlusClassifier class PlusClassifier < CheaterClassifier def initialize(a, b) @a = a @b = b end ! def cheater? partial_result = @a.cheater? if partial_result == :maybe @b.cheater? else partial_result end end end 

CheaterClassifier (Base Class) class CheaterClassifier def +(other) PlusClassifier.new(self, other) end end 

MainClassifier (Entry Point) class MainClassifier < CheaterClassifier def initialize(last_score, level) classifiers = [ ScoreCheatClassifier.new(last_score), LevelClassifier.new(last_score, level) ] @classifier = classifiers.reduce { |a,b| a + b } end end 

LevelClassifier ScoreCheatClassifier MainClassifier PlusClassifier data: Score, Level Is it a cheater yet? Reduce available Classifiers 

Sincerely yours, Lukas "Rampensau" Rieder