john-doe 0.1.4 → 0.1.8

data/.gitignore

@@ -4,3 +4,4 @@ coverage
 rdoc
 pkg
 test.rb
+quotes

data/README.rdoc

@@ -25,6 +25,9 @@ John doe is simple chatbot AI basing on idea of AIML language. Yaml is used inst
 If you want to create your own rules file just do:
   JohnDoe::ChatBot.new("your_rules.yml")
 
+If you want to use another quotes file just do:
+  JohnDoe::ChatBot.new("your_rules.yml", "my_quotes.txt")
+
 === Rules file
 
 See http://github.com/dfens/john-doe/raw/master/default.yml to understand what should be in you rules.yml file.
@@ -40,15 +43,7 @@ whoareyou:
     r3: "I'm nobody"
     r4: "I am bot:name"
 
-There can be unlimited number of patterns that lead to random response from "responses" node. For example if bot gets "Who are you?" it may response "I am chatbot" or "I'm nobody"
-
-There also must be rule 'default':
-  default: 
-    dontunderstand:
-      q1: "Sorry I can't understand you."
-      q2: "I don't get it..."
-      q3: "What what what??"
-it means that if bot do not recognize the sentence it will answer with some quote from 'dontunderstand'.
+If bot does not recognize the sentence it will use markov chain to build it. Bot learns from what you write to him (he will load this data in new session).
 
 There is also possible to add some universal knowledge for bot:
   knowledge:
@@ -82,7 +77,8 @@ whats:
     r1: "I'm not sure what's $"
 
 Now if someone will write "tell me something about ruby language" bot will answer "I'm not sure what's ruby language"
-
+===Quotes
+Quotes file is collection of sentences provided by user during chat. Using markov chains bot is generating answer if he can't recognize pattern.
 == Copyright
 
 Copyright (c) 2010 Pawel Mikolajewski. See LICENSE for details.

data/VERSION

@@ -1 +1 @@
-0.1.4
+0.1.8

data/default.yml

@@ -4,6 +4,7 @@ default:
   dontunderstand:
     q1: "Sorry I can't understand you."
     q2: "I don't get it..."
+    q3: "I'm bot and i'm written in ruby!"
   thinking:
     q1: "hm.."
     q2: "interesting..."
@@ -50,7 +51,7 @@ hello:
     r4: goodday 
 
 whatsup:
-  priority: 10
+  rank: 10
   patterns:
     p1: "*what's up"
     p2: "*whats up"
@@ -91,9 +92,11 @@ stupid:
 you are:
   patterns:
     p1: you are $
+    p2: "you're $"
   responses:
     r1: sometimes i think I am...
     r2: no, you are $!! hahaha
+    r3: thanks, I needed that
 
 boolean:
   patterns:
@@ -201,6 +204,8 @@ haha:
     p2: ha ha ha
     p3: haha
     p4: ha ha
+    p5: hehe
+    p6: he he
   responses: 
     r1: "I'm glad to make you happy"
     r2: "yes, that's funny"
@@ -222,7 +227,7 @@ because:
     r2: "thanks for explanation ;-)"
 
 specimen:
-  priority: 20
+  rank: 20
   patterns:
     p1: are you * human
     p2: are you * bot
@@ -304,15 +309,6 @@ movie:
   responses:
     r1: "I like Odyssey 2010..."
 
-areyou:
-  emotion: shame|pride|embarrassment|none
-  patterns:
-    p1: are you
-  responses:
-    r1: sometimes i think i am
-    r2: "hard to answer ;-)"
-    r3: "as I know I'm not"
-
 yesno:
   patterns:
     p1: "yes"
@@ -340,7 +336,7 @@ aboutyou:
     r3: "what about me?"
 
 whysth:
-  priority: 10
+  rank: 10
   patterns:
     p1: why $ ?
     p2: why $
@@ -356,7 +352,7 @@ why:
   responses:
     r1: hard to say why
     r2: not easy to say why
-    r3: "I don' know why"
+    r3: "I don't know why"
     r4: "I'm not sure why"
 doyou:
   patterns:
@@ -389,7 +385,7 @@ asl:
   responses:
     r1: "bot:age bot:sex bot:language"
 orly:
-  priority: 10
+  rank: 10
   patterns:
     p1: "really?"
     p2: are you sure
@@ -412,7 +408,55 @@ testing:
 
 whats:
   patterns:
-    p1: what's $
+    p1: "what's $"
     p2: what is $
   responses:
     r1: "I'm not sure what's $"
+
+whydoes:
+  rank: 10
+  patterns:
+    p1: "why does $"
+  responses:
+    r1: $?
+    r2: good to hear that $
+
+tellme:
+  rank: 5
+  patterns:
+    p1: tell me $
+  responses:
+    r1: "I can't tell you $"
+    r2: "nobody knows $"
+
+random:
+  patterns:
+    p1: random
+  responses:
+    r1: everything is random
+    r2: 42,61,12,41 - this is random!
+
+yousleep:
+  rank: 1
+  patterns:
+    p1: "*you sleep?"
+  responses:
+    r1: "I don't sleep"
+    r2: "Bots NEVER SLEEP"
+
+ithink:
+  patterns:
+    p1: "i think *"
+  responses:
+    r1: "that's right;-)"
+    r2: are you sure?
+    r3: "you're right"
+ruby:
+  patterns:
+    p1: " ruby"
+  responses:
+    r1: "You can look into my ruby code;-)"
+    r2: "i can't code ruby"
+    r3: "learn ruby and you will know"
+
+

data/lib/john-doe.rb

@@ -1,13 +1,18 @@
 require 'rubygems'
 require File.dirname(__FILE__)+'/johndoe/aiml'
 require File.dirname(__FILE__)+'/johndoe/responser'
+require File.dirname(__FILE__)+'/johndoe/markov'
 
 module JohnDoe
   class ChatBot
-    def initialize(filename = File.dirname(__FILE__) + "/../default.yml" )
+    def initialize(filename = File.dirname(__FILE__) + "/../default.yml" , quotes_filename = "quotes")
+      unless File.exists? quotes_filename
+        f = File.new(quotes_filename,"w")
+        f.close
+      end
       @loader = JohnDoe::Aiml.new
       @loader.load(filename)
-      @responser = JohnDoe::Responser.new(@loader)
+      @responser = JohnDoe::Responser.new(@loader,quotes_filename)
     end
     def get_response(s)
       @responser.response s

data/lib/johndoe/markov.rb

@@ -0,0 +1,119 @@
+module JohnDoe
+  class Markov
+
+    def initialize
+      @map = {}
+    end
+
+    def load(filename)
+      file = File.new(filename, "r")
+      while (line = file.gets)
+        process_line line.downcase
+      end
+      file.close
+      count_probability
+    end
+
+    #assume that model is linear so using linear prob
+    def count_probability
+      sum = 0
+      p_sum = 0
+      @map.each{|k,v| sum+=v[:count]}
+      @map.each do |k,v|
+        v[:p] = p_sum + v[:count]/sum.to_f 
+        p_sum = v[:p]
+
+      end
+      @map.each do |k,v|
+        word_sum = 0
+        v.each do |word_key,word_val|
+          next unless word_key.class == String
+          word_sum += word_val
+        end
+        v[:word_sum] = word_sum  
+      end
+    end
+
+    def process_line(line)
+      line.squeeze!(" ")
+      words = line.split(" ")
+      for i in 0..words.length-2
+        word1 = words[i]
+        word2 = words[i+1]
+        word3 = words[i+2]
+        add(word1,word2,word3)
+      end
+    end
+
+    def add(w1,w2,w3)
+      w12 = w1+" "+w2
+      if @map[w12].nil?
+        @map[w12] = {:count => 1} 
+      else
+        @map[w12][:count]+=1
+      end
+      if @map[w12][w3].nil?
+        @map[w12][w3] = 1
+      else
+        @map[w12][w3]+=1
+      end
+
+    end
+
+    def generate_random
+      r = rand
+      str = ""
+      @map.each  do |k,v|
+        if r < v[:p]
+          str = k
+          break
+        end
+      end
+      return gen_from_two(str)
+
+    end
+
+    def gen_from_two(str)
+      begin
+        while true
+          found_nil = false
+          two_last = two_last_words(str) rescue "I'm still learning"
+          return str if @map[two_last].nil? || @map[two_last][:word_sum] == 0
+          r = rand(@map[two_last][:word_sum])
+          @map[two_last].each do |k,v|
+            next if k.class == Symbol
+            r -= v
+            if r < 0
+              found_nil == v.nil?
+              str = str + " " + k
+              break
+            end
+          end
+        end
+      end
+      rescue nil
+    end
+
+    def two_last_words(s)
+      m = /(.* )?([^\s]+) ([^\s]+)$/.match s
+      return m[2] + " " + m[3] rescue nil
+    end
+
+    def response(str)
+      str.squeeze(" ").split(" ").sort {|x,y| y.size <=> x.size}.each do |s|
+        puts s
+        collection = []
+        @map.each do |k,v|
+          if k.include? s
+            collection << k 
+          end
+        end
+        return gen_from_two(@map.keys[rand(@map.keys.size)]) if collection.empty?
+        return gen_from_two collection[rand(rand(collection.size))] rescue nil
+      end
+      return nil
+    end
+
+  end
+
+end

data/lib/johndoe/responser.rb

@@ -1,4 +1,3 @@
-
 module JohnDoe
   class Response
     attr_accessor :text, :emotion
@@ -8,17 +7,22 @@ module JohnDoe
     end
   end
   class Responser
-    def initialize(data)
+    def initialize(data, quotes)
       @data = data
+      @markov = Markov.new
+      @markov.load quotes
+      @quotes_file = File.open(quotes,"a")
     end
 
     def response(sentence)
+      @quotes_file.puts(sentence + "\n")
       max_priority = -1
       best_v = nil
       best_k = nil
 
       @data.patterns.each do |k,v|
         if (/^#{k}/i =~ sentence)
+          puts "#{v[:priority]}:#{k}"
           next if v[:priority] <= max_priority
           max_priority = v[:priority]
           best_v = v
@@ -28,6 +32,9 @@ module JohnDoe
       unless best_v.nil?
         return Response.new(sub_v(random_quote(@data.responses[best_v[:resp]], /^#{best_k}/i.match(sentence).captures)),best_v[:emotions])
       else
+        generated = @markov.response sentence
+        puts generated.to_s + "###"
+        return Response.new(generated, ["none"]) unless generated.nil?
         return Response.new(sub_v(random_quote(@data.default["dontunderstand"])),["none"])
       end
     end
@@ -57,8 +64,8 @@ module JohnDoe
     end
 
     def describe_who(s)
-      s = s.gsub(/([^a-z])am\s+i([^a-z]|$)/,"\\1#AMI#\\2").gsub(/([^a-z])i\s+am([^a-z]|$)/,"\\1#IAM#\\2").gsub(/([^a-z])you\s+are([^a-z]|$)/,"\\1#AREYOU#\\2").gsub(/([^a-z])are\s+you([^a-z]|$)/,"\\1#AREYOU#\\2").gsub(/([^a-z])me([^a-z]|$)/,"\\1#ME#\\2").gsub(/([^a-z])you([^a-z]|$)/,"\\1#YOU#\\2")
-      s.gsub("#AMI#","you are").gsub("#IAM#","you are").gsub("#AREYOU#","I am").gsub("#ME#","you").gsub("#YOU#","me")
+      s = s.gsub(/([^a-z])am\s+i([^a-z]|$)/,"\\1#AMI#\\2").gsub(/([^a-z])i\s+am([^a-z]|$)/,"\\1#IAM#\\2").gsub(/([^a-z])you\s+are([^a-z]|$)/,"\\1#AREYOU#\\2").gsub(/([^a-z])are\s+you([^a-z]|$)/,"\\1#AREYOU#\\2").gsub(/([^a-z])me([^a-z]|$)/,"\\1#ME#\\2").gsub(/([^a-z])you([^a-z]|$)/,"\\1#YOU#\\2").gsub(/([^a-z])your([^a-z]|$)/,"\\1#YOUR#\\2").gsub(/([^a-z])my([^a-z]|$)/,"\\1#MY#\\2")
+      s.gsub("#AMI#","you are").gsub("#IAM#","you are").gsub("#AREYOU#","I am").gsub("#ME#","you").gsub("#YOU#","me").gsub("#YOUR#","my").gsub("#MY#","your")
     end
   end
 end

data/quotes

@@ -0,0 +1,82 @@
+Everybody's always telling me one thing and out the other.
+Even a fish could stay out of trouble if it would just learn to keep its mouth shut.
+Beware the lollipop of mediocrity -- lick it once and you suck forever.
+We don't have time to stop for gas -- we're already late.
+By doing just a little each day, I can gradually let the task overwhelm me.
+Being powerful is like being a lady. If you have to tell people you are, you aren't. 
+Never attribute to malice that which can satisfactorily be explained by incompetence. 
+Some men are born mediocre, some men achieve mediocrity, and some men have mediocrity thrust upon them. 
+Thought: why does man kill? He kills for food. And not only for food: frequently there must be a beverage. 
+Everything should be made as simple as possible, but not simpler. 
+Government is not reason; it is not eloquence; it is force! It is a dangerous servant and a terrible master. 
+The secret to creativity is knowing how to hide your sources. 
+It usually takes more than three weeks to prepare a good impromptu speech. 
+In the long run, we are all dead. 
+There are two kinds of people, those who finish what they start and so on. 
+Learning to dislike children at an early age saves a lot of expense and aggravation later in life. 
+America is a mistake, a giant mistake. 
+If you think nobody cares if you're alive, try missing a couple of car payments. 
+Many are cold but few are frozen. 
+Laugh and the world laughs with you, snore and you sleep alone. 
+Never confuse motion with action. 
+Washington is a city of Southern efficiency and Northern charm. 
+Power in defense of freedom is greater than power in behalf of tyranny and oppression. 
+This is not a novel to be tossed aside lightly. It should be thrown aside with great force. 
+The trouble with the rat race is that even if you win, you're still a rat. 
+Experience: A comb life gives you after you lose your hair. 
+Float like a butterfly, sting like a bee. 
+Under capitalism, man exploits man. Under communism, it's just the opposite. 
+Life's most urgent question is: what are you doing for others? 
+We do not inherit this land from our ancestors; we borrow it from our children. 
+An eye for eye only ends up making the whole world blind. 
+In waking a tiger, use a long stick. (Mao Tse-tung)
+Never go to a doctor whose office plants have died. 
+What you do is of little signifigance. But it is very important that you do it. 
+Never mistake motion for action. 
+Never try and teach a pig to sing. It wastes your time and annoys the pig. 
+Time flies like an arrow, and fruit flies like a banana. 
+I love criticism just so long as it's unqualified praise. 
+Beauty is only skin deep, but ugly goes clean to the bone. 
+Put all your eggs in one basket, and watch that basket. 
+Biography lends to death a new terror. 
+Victory goes to the player who makes the next-to-last mistake. 
+When the only tool you own is a hammer, every problem begins to resemble a nail. 
+Better to light a candle than curse the darkness. 
+I'm not going to vacuum 'til Sears makes one you can ride on. 
+Nothing in life is to be feared. It is only to be understood. 
+Always be nice to your children because they are the ones who will choose your rest home. 
+Thousands of candles can be lighted from a single candle, and the life of the candle will not be shortened.  Happiness never decreases by being shared. 
+The man who strikes first admits that his ideas have given out. 
+Never judge a book by it's movie. (J.W. Eagan)
+Don't bother just to be better than your contemporaries or predecessors. Try to be better than yourself. 
+640K ought to be enough for anybody. 
+There is no reason anyone would want a computer in their home. 
+You can't separate peace from freedom because no one can be at peace unless he has his freedom. 
+If you spend a perfectly useless afternoon in a perfectly useless manner, you have learned how to live. 
+Never interrupt your enemy when he is making a mistake. 
+The difference between 'involvement' and 'commitment' is like an eggs-and-ham breakfast: the chicken was 'involved' - the pig was 'committed'. 
+If you are going through hell, keep going. 
+If you haven't got anything nice to say about anybody, come sit next to me. 
+When choosing between two evils, I always like to try the one I've never tried before. 
+There are only two tragedies in life: one is not getting what one wants, and the other is getting it.      
+
+Who is the best ?
+I don't get it
+bee?
+is there something that you did not try before ?
+what do you feel right now ?
+i feel bad
+hello
+how are ou
+how to live?
+your sources?
+it does not make sense
+;)
+:)
+shit
+yes
+thaaf
+pathetic
+asdfaokoook
+okkokokk
+lllllllllll

metadata

@@ -1,13 +1,13 @@
 --- !ruby/object:Gem::Specification 
 name: john-doe
 version: !ruby/object:Gem::Version 
-  hash: 19
+  hash: 11
   prerelease: false
   segments: 
   - 0
   - 1
-  - 4
-  version: 0.1.4
+  - 8
+  version: 0.1.8
 platform: ruby
 authors: 
 - Pawel Mikolajewski
@@ -15,7 +15,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2010-06-14 00:00:00 +02:00
+date: 2010-06-18 00:00:00 +02:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency 
@@ -49,10 +49,11 @@ files:
 - Rakefile
 - VERSION
 - default.yml
-- john-doe.gemspec
 - lib/john-doe.rb
 - lib/johndoe/aiml.rb
+- lib/johndoe/markov.rb
 - lib/johndoe/responser.rb
+- quotes
 - test/john-doe_test.rb
 - test/test_helper.rb
 has_rdoc: true

data/john-doe.gemspec

@@ -1,56 +0,0 @@
-# Generated by jeweler
-# DO NOT EDIT THIS FILE
-# Instead, edit Jeweler::Tasks in Rakefile, and run `rake gemspec`
-# -*- encoding: utf-8 -*-
-
-Gem::Specification.new do |s|
-  s.name = %q{john-doe}
-  s.version = "0.1.4"
-
-  s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
-  s.authors = ["Pawel Mikolajewski"]
-  s.date = %q{2010-06-14}
-  s.description = %q{Simple chatbot AI}
-  s.email = %q{mikolajewskip@gmail.com}
-  s.extra_rdoc_files = [
-    "LICENSE",
-     "README.rdoc"
-  ]
-  s.files = [
-    ".document",
-     ".gitignore",
-     "LICENSE",
-     "README.rdoc",
-     "Rakefile",
-     "VERSION",
-     "default.yml",
-     "john-doe.gemspec",
-     "lib/john-doe.rb",
-     "lib/johndoe/aiml.rb",
-     "lib/johndoe/responser.rb",
-     "test/john-doe_test.rb",
-     "test/test_helper.rb"
-  ]
-  s.homepage = %q{http://github.com/dfens/john-doe}
-  s.rdoc_options = ["--charset=UTF-8"]
-  s.require_paths = ["lib"]
-  s.rubygems_version = %q{1.3.7}
-  s.summary = %q{Minimal chatbot AI}
-  s.test_files = [
-    "test/john-doe_test.rb",
-     "test/test_helper.rb"
-  ]
-
-  if s.respond_to? :specification_version then
-    current_version = Gem::Specification::CURRENT_SPECIFICATION_VERSION
-    s.specification_version = 3
-
-    if Gem::Version.new(Gem::VERSION) >= Gem::Version.new('1.2.0') then
-      s.add_development_dependency(%q<thoughtbot-shoulda>, [">= 0"])
-    else
-      s.add_dependency(%q<thoughtbot-shoulda>, [">= 0"])
-    end
-  else
-    s.add_dependency(%q<thoughtbot-shoulda>, [">= 0"])
-  end
-end