markov_noodles 1.0.0

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: 25c202c27f6207b467221cbf0d8a009ee19638f6
+  data.tar.gz: 747f38c46ad2775a86b10c6469d434597911aef4
+SHA512:
+  metadata.gz: 6b097a3bb1b3973b4fd1780c5b8d3d1000918e4ef42dab87f7c2401b1f67cb86eb8cf2869778ba81b613d374c11a766ecde8060b590853ee1c00da5274c1693c
+  data.tar.gz: 9ff568d2373c0003c4e2982d6a0156815bbe672f0560f274c963d855521956a698d20c3a2d1be76534c0317ae15d4c0b9a2b1154abf417cfad199b8cc062e8c2

data/lib/markov_noodles.rb

@@ -0,0 +1,95 @@
+require 'msgpack'
+
+class MarkovNoodles
+  attr_reader :dictionary
+  attr_reader :depth
+
+  def initialize(depth = 2)
+    @depth = depth
+    @dictionary = {}
+  end
+
+  def analyse_file(filename)
+    analyse_string(File.read(filename))
+  end
+
+  def analyse_string(text)
+    current_words = Array.new(depth)
+    text_array = split_text_to_array(text)
+    while text_array.length > 0
+      next_word = text_array.shift
+      add_words(current_words.dup, next_word)
+      current_words.push next_word
+      current_words.shift
+    end
+  end
+
+  def save_dictionary(filename)
+    File.open(filename, 'w') do |file|
+      file.write @dictionary.to_msgpack
+    end
+  end
+
+  def load_dictionary(filename)
+    @dictionary = MessagePack.unpack(File.read(filename))
+  end
+
+  def generate_n_sentences(n)
+    text = ''
+    n.times do |i|
+      text.concat(generate_sentence)
+      is_last_sentence = i == (n - 1)
+      text.concat(' ') unless is_last_sentence
+    end
+    text
+  end
+
+  def generate_sentence
+    current_words = Array.new(depth)
+    sentence_array = []
+    loop do
+      new_word = current_words.last
+      sentence_array.push new_word if new_word
+      break if end_word?(new_word)
+      next_word_options = @dictionary[current_words]
+      if next_word_options.nil? && !end_word?(new_word)
+        new_word.concat('.')
+        break
+      end
+      next_word = next_word_options.sample
+      current_words.push next_word
+      current_words.shift
+    end
+    sentence_array.join(' ')
+  end
+
+  private
+
+  # Splits a text into array and inserts @depth nils after each sentence.
+  #
+  # This way generated texts can start with any word that is at the beginning of
+  # any sentence in analysed text, instead of always starting with the first
+  # word from the text.
+  def split_text_to_array(text)
+    text_array = []
+    text.split.each do |word|
+      text_array.push word
+      if end_word?(word)
+        depth.times do
+          text_array.push nil
+        end
+      end
+    end
+    text_array
+  end
+
+  def add_words(preceding, followedby)
+    @dictionary[preceding] ||= []
+    @dictionary[preceding].push followedby
+  end
+
+  # Checks if word ends with one of following characters .?$
+  def end_word?(word)
+    !word.nil? && !(word =~ /^*+[?\.!]$/).nil?
+  end
+end

metadata

@@ -0,0 +1,46 @@
+--- !ruby/object:Gem::Specification
+name: markov_noodles
+version: !ruby/object:Gem::Version
+  version: 1.0.0
+platform: ruby
+authors:
+- Filip Defar
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2015-08-11 00:00:00.000000000 Z
+dependencies: []
+description: Noodles uses Markov chains to generate superficially real-looking text
+  given a sample document.
+email: filip.defar@gmail.com
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- lib/markov_noodles.rb
+homepage: https://github.com/dabrorius/noodles
+licenses:
+- MIT
+metadata: {}
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubyforge_project: 
+rubygems_version: 2.4.5
+signing_key: 
+specification_version: 4
+summary: Simple markov chain implementation
+test_files: []
+has_rdoc: