markovite 0.2.4 → 0.2.5

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: a45b08ff10130126c3f0325fca5a1a1d23d31b942e0430d088f73b6ab812faac
-  data.tar.gz: 698a6acf00459bb8f61f8325e48f2c096b36af8de6890126d2e68c2a0d994627
+  metadata.gz: 844cdac9c79870dc34add5123467ba7de998324dbceb333c3358f45b96de15a5
+  data.tar.gz: 3321497a835c6df6a38e34b8f211ffa1dbd1916bfe152ae071b8210b54a79c18
 SHA512:
-  metadata.gz: d2579a001419cf3ca5913d8d55b7d05b98f5628de91c7b128b9276d58fc22cb2fded0237dc0852df1f2c4c15c1d461ac7e4f522fca1d002c9cae254ede2eb4fc
-  data.tar.gz: e5a2b74db3c68cde040a1d935d9c62237d6f076fa2d82893cdcbed0a7b42171ca8e47d71c76bcff15c7c663ef1edd1e231e5dbd1353dc78a2d8023d246c4c381
+  metadata.gz: 8d6a1d275ce80f2deef594d19baf7abd2a52652e6d3e5c034b06a0950777ef3f4ed62411c59c509f0e3e5767aac683b9f32d7075a4de26133588eb4b821042f5
+  data.tar.gz: bb8f7930db79db564bffeb2676200d559a26f202bcc4ed7d49d296b0ad41299d23575cc9e4f1dc0724430645e530d627d15d122379c09242522886843287ef9f

data/Gemfile.lock

@@ -1,7 +1,8 @@
 PATH
   remote: .
   specs:
-    markovite (0.2.4)
+    markovite (0.2.5)
+      msgpack
 
 GEM
   remote: https://rubygems.org/
@@ -10,6 +11,7 @@ GEM
     diff-lcs (1.3)
     interception (0.5)
     method_source (0.9.0)
+    msgpack (1.2.4)
     pry (0.11.3)
       coderay (~> 1.1.0)
       method_source (~> 0.9.0)

data/config/environment.rb

@@ -1,4 +1,5 @@
 require_relative "../lib/markovite/chainer.rb"
 require_relative "../lib/markovite/dict.rb"
 require_relative "../lib/markovite/splitter.rb"
-require_relative "../lib/markovite.rb"
+require_relative "../lib/markovite.rb"
+require "msgpack"

data/lib/markovite.rb

@@ -11,9 +11,33 @@ module Markovite
     MAX_FILENAME_LENGTH = 255
 
     def initialize(filename = nil, dict_depth=DEFAULT_DEPTH)
+      initialize_children
       parse_file(filename, dict_depth) if filename
     end
 
+    def save(filename)
+      raise("Chain is empty") if dictionary.chain.empty?
+      msg_hash = {}
+      msg_hash["sentences"] = dictionary.sentences
+      msg_hash["chain"] = dictionary.chain
+      msg_hash["corpus"] = split.corpus
+      msg_hash["depth"] = dictionary.depth
+      File.open("#{filename}.msg", "w") do |file|
+        test = file.write(msg_hash.to_msgpack)
+      end
+      true
+    end
+
+    def load(filename)
+      raise("Invalid file type") if !is_valid_file_ext?(filename, /.msg\z/i)
+      data = File.read("#{filename}")
+      model = MessagePack.unpack(data)
+      @depth = model["depth"]
+      split.corpus = model["corpus"]
+      dictionary.sentences = model["sentences"]
+      dictionary.chain = model["chain"]
+    end
+
     def self.combine(left_chain, right_chain, dict_depth = nil)
       dict_depth = dict_depth || left_chain.depth
       new_chain = Markovite::Chain.new
@@ -26,12 +50,13 @@ module Markovite
       split.corpus
     end
 
-    def parse_string(text, dict_depth=nil)
-      dict_depth = dict_depth || self.depth || DEFAULT_DEPTH
-      depth_check(dict_depth)
-      if chainer
+    def parse_string(text, dict_depth = nil)
+      if self.depth
+        depth_check(dict_depth)
         add_from_text(text)
       else
+        dict_depth = dict_depth || DEFAULT_DEPTH
+        is_valid_depth?(dict_depth)
         new_from_text(text, dict_depth)
       end
     end
@@ -75,11 +100,7 @@ module Markovite
     private
 
     def depth_check(dict_depth)
-      if chainer
-        raise "Chain depth conflict" if dict_depth != depth
-      elsif !is_valid_depth?(dict_depth)
-        raise "Chain depth must be between #{MIN_DEPTH} and #{MAX_DEPTH}"
-      end
+        raise "Chain depth conflict" if !dict_depth.nil? && dict_depth != depth
     end
 
     def is_valid_depth?(dict_depth)
@@ -91,16 +112,14 @@ module Markovite
     end
 
     def new_from_text(text, dict_depth)
-      #look into refactoring this
       @depth = dict_depth
-      self.split = SplitSentence.new(text)
       @corpus = split.corpus
-      self.dictionary = Dictionary.new(split, depth)
-      self.chainer = Chainer.new(dictionary)
+      dictionary.expand_chain(text)
     end
 
-    def is_valid_file_ext?(filename)
-      re = Regexp.union(FILE_EXT)
+
+    def is_valid_file_ext?(filename, ext = nil)
+      re = ext || Regexp.union(FILE_EXT)
       filename.match(re)
     end
 
@@ -108,6 +127,12 @@ module Markovite
       str.length < 255 && split_words(str).length == 1
     end
 
+    def initialize_children
+      self.split = SplitSentence.new
+      self.dictionary = Dictionary.new({sentence_split: split})
+      self.chainer = Chainer.new(dictionary)
+    end
+
     def add_from_text(text)
       dictionary.expand_chain(text)
     end

data/lib/markovite/chainer.rb

@@ -90,8 +90,9 @@ class Chainer
 
   def remove_markers(sentence)
     #removes BEGINNING and ENDING markers
-    sentence.pop
-    sentence.shift(depth)
+    sentence.shift while sentence.first == BEGINNING
+    sentence.pop while sentence.last == ENDING
+    sentence
   end
 
   def is_valid_sentence?(sentence)

data/lib/markovite/dict.rb

@@ -6,26 +6,33 @@ class Dictionary
   #make this a module???
   BEGINNING = "__BEGIN__"
   ENDING = "__END__"
+  DEFAULT_DEPTH = 2
 
-  attr_accessor :chain, :sentence_split, :sentences
-  attr_reader :depth
-  def initialize(sentence_split = nil, depth = 2)
-    self.sentence_split = sentence_split || SentenceSplit.new
-    # The following line ensures a new array is created for each new key
-    # instead of using the memory address of the first array created    raise exception "First argument must contain a SplitSentence instance" if sentence_split.class != SentenceSplit
-    # as the default value
-    self.chain = Hash.new { |h, k| h[k] = [] }
-    self.sentences = sentence_split.split_text
-    @depth = depth
-    construct_chain
+  attr_accessor :sentence_split, :sentences
+  attr_reader :depth, :chain
+
+  def initialize(attributes)
+    attributes.each {|attribute, value| self.send("#{attribute}=", value)}
+    set_default
   end
 
   def has_sentence(sentence)
     sentences.include?(sentence)
   end
 
+  def chain=(arg)
+    # The following line ensures a new array is created for each new key
+    # instead of using the memory address of the first array created
+    # as the default value
+    @chain = Hash.new { |h, k| h[k] = [] } if self.chain.nil?
+    arg.each {|key, value|chain[key] = value}
+    chain
+  end
+
   def depth=(arg)
-    raise "Depth cannot be changed"
+    raise "Depth cannot be changed" if depth
+    raise "Depth must be integer" if arg.class != Integer
+    @depth = arg
   end
 
   def expand_chain(text)
@@ -36,8 +43,8 @@ class Dictionary
   end
 
   def construct_chain(new_sentences = nil)
+    self.depth = DEFAULT_DEPTH if depth.nil?
     new_sentences = new_sentences || sentences
-    raise "No sentences in memory" if new_sentences.empty?
     new_sentences.each do |sentence|
       words = sentence.split(" ")
       # each chunk is an array that represents a state in the markov chain
@@ -65,4 +72,13 @@ class Dictionary
     sentences.clear
   end
 
+  private
+
+  def set_default
+    self.sentence_split = sentence_split || SentenceSplit.new
+    self.chain = chain || {}
+    self.sentences = sentences || sentence_split.split_text
+    construct_chain if chain.empty?
+  end
+
 end

data/lib/markovite/splitter.rb

@@ -17,13 +17,14 @@ class SplitSentence
     "prof.", "rev.", "rt. hon.", "sr.", "st." #personal
   ]
 
-  #look into detecting abbreviations!
+  attr_reader :corpus
 
-  attr_accessor :corpus
+  def initialize(corpus = nil)
+    self.corpus = corpus || ""
+  end
 
-  def initialize(corpus = "")
-    self.corpus = corpus.dup
-    split_text
+  def corpus=(text)
+    @corpus = text.dup
   end
 
   # We will want to change this to something that splits the words into an
@@ -58,7 +59,7 @@ class SplitSentence
   end
 
   def expand_corpus(text)
-    self.corpus += "\n" + text
+    self.corpus += " #{text}"
   end
 
   private

data/lib/markovite/version.rb

@@ -1,3 +1,3 @@
 module Markovite
-  VERSION = "0.2.4"
+  VERSION = "0.2.5"
 end

data/markovite.gemspec

@@ -28,7 +28,7 @@ Gem::Specification.new do |spec|
   end
 
   spec.require_paths = ["lib", "config"]
-
+  spec.add_runtime_dependency "msgpack"
   spec.add_development_dependency "bundler", "~> 1.16"
   spec.add_development_dependency "rake", "~> 10.0"
   spec.add_development_dependency "rspec", "~> 3.0"

metadata

@@ -1,15 +1,29 @@
 --- !ruby/object:Gem::Specification
 name: markovite
 version: !ruby/object:Gem::Version
-  version: 0.2.4
+  version: 0.2.5
 platform: ruby
 authors:
 - Alexandra Wright
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2018-04-02 00:00:00.000000000 Z
+date: 2018-04-12 00:00:00.000000000 Z
 dependencies:
+- !ruby/object:Gem::Dependency
+  name: msgpack
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
 - !ruby/object:Gem::Dependency
   name: bundler
   requirement: !ruby/object:Gem::Requirement
@@ -106,8 +120,6 @@ files:
 - lib/markovite/splitter.rb
 - lib/markovite/version.rb
 - markovite.gemspec
-- pkg/markovite-0.2.2.gem
-- pkg/markovite-0.2.3.gem
 homepage: https://github.com/f3mshep/ruby_markovify
 licenses:
 - MIT