guillaume 0.0.1

data/guillaume.gemspec

@@ -0,0 +1,23 @@
+# coding: utf-8
+lib = File.expand_path('../lib', __FILE__)
+$LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
+require 'guillaume/version'
+
+Gem::Specification.new do |spec|
+  spec.name          = "guillaume"
+  spec.version       = Guillaume::VERSION
+  spec.authors       = ["Jamey J. DeOrio"]
+  spec.email         = ["jdeorio@gmail.com"]
+  spec.description   = %q{Guillaume is a generative poetry creation tool using ngrams for predictive writing and a defined poetics for guiding its creation.}
+  spec.summary       = "A generative poetry creation tool"
+  spec.homepage      = "http://rubygems.org/gems/guillaume"
+  spec.license       = "MIT"
+
+  spec.files         = `git ls-files -z`.split("\x0")
+  spec.executables   = spec.files.grep(%r{^bin/}) { |f| File.basename(f) }
+  spec.test_files    = spec.files.grep(%r{^(test|spec|features)/})
+  spec.require_paths = ["lib"]
+
+  spec.add_development_dependency "bundler", "~> 1.3"
+  spec.add_development_dependency "rake", "~> 0"
+end

data/lib/guillaume/base.rb

@@ -0,0 +1,2 @@
+module Guillaume
+end

data/lib/guillaume/line.rb

@@ -0,0 +1,29 @@
+class Guillaume::Line
+  def initialize(seed = nil)
+    @line_parts = seed.split(/ /)
+  end
+
+  def build(ngrams)
+    begin
+      @line_parts << next_word(ngrams)
+    end until @line_parts.last.nil?
+    @line_parts.join(" ").strip
+  end
+
+  def next_word(ngrams, options = { look_behind: 1 })
+    search_words = @line_parts[-options[:look_behind]..-1]
+    choose_word(ngram_matches(ngrams, search_words))
+  end
+
+  def ngram_matches(ngrams, search_words)
+    ngrams.select { |gram| gram[0..search_words.count-1] == search_words }
+  end
+
+  def choose_word(matches)
+    possible_choices(matches).sample
+  end
+
+  def possible_choices(matches)
+    matches.map { |res| res.last }.sort
+  end
+end

data/lib/guillaume/poem.rb

@@ -0,0 +1,51 @@
+class Guillaume::Poem
+  attr_accessor :corpora, :lines, :poem
+
+  def initialize(corpora, options = { seed: nil, max_stanzas: 10 })
+    @corpora = corpora
+    @first_seed = options[:seed].nil? ? random_seed(@corpora.unigrams) : options[:seed]
+    @max_stanzas = options[:max_stanzas]
+    @lines = 0
+    @stanzas = 0
+    @poem = write
+  end
+
+  def random_seed(ngrams)
+    ngrams.select { |word| word.first[0] =~ /[A-Z]/ }.sample.first
+  end
+
+  def write(lines_memo = [])
+    if rand(@max_stanzas) < @stanzas
+      return lines_memo
+    else
+      $LOGGER.info("Writing stanza #{@stanzas + 1}...")
+      lines_memo += stanza
+      @stanzas += 1
+      write(lines_memo)
+    end
+  end
+
+  #
+  # TODO: which_gram
+  # TODO: thesaurus stuff for topic staying/straying
+  # TODO: explicit num_lines option
+  #
+  def stanza(lines_memo = [])
+    #which_gram = :bigrams
+
+    if rand(100) <= lines_memo.count * 10 # never more than 10 lines this way
+      return lines_memo << ""
+    else
+      line = Guillaume::Line.new(random_seed(@corpora.unigrams)).build(@corpora.bigrams)
+      if line.length > 80
+        line = Guillaume::Poetics.enjamb 40, line # 40% chance to break a long line
+      end
+      lines_memo << line
+      stanza(lines_memo)
+    end
+  end
+
+  def formatted
+    @formatted ||= @poem.join("\n")
+  end
+end

data/lib/guillaume/poetics.rb

@@ -0,0 +1,13 @@
+class Guillaume::Poetics
+  def self.enjamb(chance=100, line)
+    if rand(100) <= chance
+      line_parts = line.split(/(?<=[\.,;!-])/)
+      if line_parts.count > 1 # if the line contains any . , ; ! -
+        random_index = rand(line_parts.count)
+        line_parts[random_index] = line_parts[random_index] + "\n"
+        line_parts.join.strip
+      end
+    end
+    line
+  end
+end

data/lib/guillaume/source_text.rb

@@ -0,0 +1,34 @@
+require "uri"
+
+class Guillaume::SourceText
+  def initialize(file_name)
+    @file_name = file_name
+    @raw_source = File.open(file_name, "rb") { |f| f.read }
+    @sentences = normalized.split(/(?<=\.|\?|!) /).reject(&:empty?)
+  end
+
+  def normalized
+    @raw_source.gsub(/[\r\n ]+/, " ").strip
+  end
+
+  def ngrams(n)
+    $LOGGER.info("Grabbing #{n}-grams from #{@file_name}...")
+    result = []
+    @sentences.each do |sentence|
+      result += sentence.split(" ").each_cons(n).to_a
+    end
+    result
+  end
+
+  def unigrams
+    @unigrams ||= ngrams 1
+  end
+
+  def bigrams
+    @bigrams ||= ngrams 2
+  end
+
+  def trigrams
+    @trigrams ||= ngrams 3
+  end
+end

data/lib/guillaume/version.rb

@@ -0,0 +1,3 @@
+module Guillaume
+  VERSION = "0.0.1"
+end

data/lib/guillaume.rb

@@ -0,0 +1,13 @@
+require "net/http"
+require "logger"
+require "guillaume/version"
+require "guillaume/source_text"
+require "guillaume/line"
+require "guillaume/poem"
+require "guillaume/poetics"
+
+$LOGGER = Logger.new(STDOUT)
+$LOGGER.level = Logger::ERROR
+
+module Guillaume
+end

data/spec/data/zone.txt

@@ -0,0 +1,4 @@
+   A la fin tu es las de
+
+ce monde ancien
+

data/spec/source_text_spec.rb

@@ -0,0 +1,23 @@
+describe Guillaume::SourceText do
+  let(:file_path) { "spec/data/zone.txt" }
+
+  subject do
+    Guillaume::SourceText.new(file_path)
+  end
+
+  it "can normalize the input file" do
+    expect(subject.normalized).to eq("A la fin tu es las de ce monde ancien")
+  end
+
+  it "has unigrams" do
+    expect(subject.unigrams).to eq([%w(A), %w(la), %w(fin), %w(tu), %w(es), %w(las), %w(de), %w(ce), %w(monde), %w(ancien)])
+  end
+
+  it "has bigrams" do
+    expect(subject.bigrams).to eq([%w(A la), %w(la fin), %w(fin tu), %w(tu es), %w(es las), %w(las de), %w(de ce), %w(ce monde), %w(monde ancien)])
+  end
+
+  it "has trigrams" do
+    expect(subject.trigrams).to eq([%w(A la fin), %w(la fin tu), %w(fin tu es), %w(tu es las), %w(es las de), %w(las de ce), %w( de ce monde), %w(ce monde ancien)])
+  end
+end

data/spec/spec_helper.rb

@@ -0,0 +1,5 @@
+require "guillaume"
+
+RSpec.configure do |config|
+  config.color = true
+end

metadata

@@ -0,0 +1,100 @@
+--- !ruby/object:Gem::Specification
+name: guillaume
+version: !ruby/object:Gem::Version
+  version: 0.0.1
+platform: ruby
+authors:
+- Jamey J. DeOrio
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2014-07-03 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: bundler
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.3'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.3'
+- !ruby/object:Gem::Dependency
+  name: rake
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0'
+description: Guillaume is a generative poetry creation tool using ngrams for predictive
+  writing and a defined poetics for guiding its creation.
+email:
+- jdeorio@gmail.com
+executables:
+- guillaume
+extensions: []
+extra_rdoc_files: []
+files:
+- ".gitignore"
+- ".rspec"
+- Gemfile
+- LICENSE.txt
+- README.markdown
+- Rakefile
+- bin/guillaume
+- config/database.yaml.example
+- data/call_of_the_wild.txt
+- data/heart_of_darkness.txt
+- data/moby_dick.txt
+- data/the_picture_of_dorian_gray.txt
+- guillaume.gemspec
+- lib/guillaume.rb
+- lib/guillaume/base.rb
+- lib/guillaume/line.rb
+- lib/guillaume/poem.rb
+- lib/guillaume/poetics.rb
+- lib/guillaume/source_text.rb
+- lib/guillaume/version.rb
+- spec/data/zone.txt
+- spec/source_text_spec.rb
+- spec/spec_helper.rb
+homepage: http://rubygems.org/gems/guillaume
+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: 2.2.2
+signing_key: 
+specification_version: 4
+summary: A generative poetry creation tool
+test_files:
+- spec/data/zone.txt
+- spec/source_text_spec.rb
+- spec/spec_helper.rb