JayVerb 0.0.0

data/lib/japanese/to_romaji.rb

@@ -0,0 +1,195 @@
+module Japanese
+  module ToRomaji
+  
+    BASIC_HIRAGANA_MAP = {"あ": "a",
+                          "い": "i",
+                          "う": "u",
+                          "え": "e",
+                          "お": "o",
+                          "か": "ka",
+                          "き": "ki",
+                          "く": "ku",
+                          "け": "ke",
+                          "こ": "ko",
+                          "さ": "sa",
+                          "し": "shi",
+                          "す": "su",
+                          "せ": "se",
+                          "そ": "so",
+                          "た": "ta",
+                          "ち": "chi",
+                          "つ": "tsu",
+                          "て": "te",
+                          "と": "to",
+                          "な": "na",
+                          "に": "ni",
+                          "ぬ": "nu",
+                          "ね": "ne",
+                          "の": "no",
+                          "は": "ha",
+                          "ひ": "hi",
+                          "ふ": "fu",
+                          "へ": "he",
+                          "ほ": "ho",
+                          "ま": "ma",
+                          "み": "mi",
+                          "む": "mu",
+                          "め": "me",
+                          "も": "mo",
+                          "や": "ya",
+                          "ゆ": "yu",
+                          "よ": "yo",
+                          "ら": "ra",
+                          "り": "ri",
+                          "る": "ru",
+                          "れ": "re",
+                          "ろ": "ro",
+                          "わ": "wa",
+                          "を": "wo",
+                          "ん": "n",
+                          "が": "ga",
+                          "ぎ": "gi",
+                          "ぐ": "gu",
+                          "げ": "ge",
+                          "ご": "go",
+                          "ざ": "za",
+                          "じ": "ji",
+                          "ず": "zu",
+                          "ぜ": "ze",
+                          "ぞ": "zo",
+                          "だ": "da", 
+                          "ぢ": "dchi",
+                          "づ": "dzu",
+                          "で": "de",
+                          "ど": "do",
+                          "ば": "ba",
+                          "び": "bi",
+                          "ぶ": "bu",
+                          "べ": "be",
+                          "ぼ": "bo",
+                          "ぱ": "pa",
+                          "ぴ": "pi",
+                          "ぷ": "pu",
+                          "ぺ": "pe",
+                          "ぽ": "po" 
+                          }
+  
+    SMALL_COMBO_HIRAGANA_MAP = {
+      "きゃ": "kya",
+      "きゅ": "kyu",
+      "きょ": "kyo",
+      "しゃ": "sha",
+      "しゅ": "shu",
+      "しょ": "sho",
+      "ちゃ": "cha",
+      "ちゅ": "chu",
+      "ちょ": "cho",
+      "にゃ": "nya",
+      "にゅ": "nyu",
+      "にょ": "nyo",
+      "ひゃ": "hya",
+      "ひゅ": "hyu",
+      "ひょ": "hyo",
+      "みゃ": "mya",
+      "みゅ": "myu",
+      "みょ": "myo",
+      "りゃ": "rya",
+      "りゅ": "ryu",
+      "りょ": "ryo",
+      "ぎゃ": "gya",
+      "ぎゅ": "gyu",
+      "ぎょ": "gyo",
+      "じゃ": "ja",
+      "じゅ": "ju",
+      "じょ": "jo",
+      "ぢゃ": "dja",
+      "ぢゅ": "dju",
+      "ぢょ": "djo",
+      "びゃ": "bya",
+      "びゅ": "byu",
+      "びょ": "byo",
+      "ぴゃ": "pya",
+      "ぴゅ": "pyu",
+      "ぴょ": "pyo",
+      "てぃ": "ti"
+    }
+  
+    DOUBLED_CONSONANT_HIRAGANA_MAP = {
+      "っか": "kka",
+      "っき": "kki",
+      "っく": "kku",
+      "っけ": "kke",
+      "っこ": "kko",
+      "っさ": "ssa",
+      "っし": "sshi",
+      "っす": "ssu",
+      "っせ": "sse",
+      "っそ": "sso",
+      "った": "tta",
+      "っち": "cchi",
+      "っつ": "ttsu",
+      "って": "tte",
+      "っと": "tto",
+      "っば": "bba",
+      "っび": "bbi",
+      "っぶ": "bbu",
+      "っべ": "bbe",
+      "っぼ": "bbo",
+      "っぱ": "ppa",
+      "っぴ": "ppi",
+      "っぷ": "ppu",
+      "っぺ": "ppe",
+      "っぽ": "ppo"
+    }
+  
+    TRIPLE_CHAR_HIRAGANA_MAP = {
+      "っきゃ": "kkya",
+      "っきゅ": "kkyu",
+      "っきょ": "kkyo",
+      "っしゃ": "ssha",
+      "っしゅ": "sshu",
+      "っしょ": "ssho",
+      "っちゃ": "ccha",
+      "っちゅ": "cchu",
+      "っちょ": "ccho",
+      "っじゃ": "jja",
+      "っじゅ": "jju",
+      "っじょ": "jjo",
+      "っびゃ": "bbya",
+      "っびゅ": "bbyu",
+      "っびょ": "bbyo",
+      "っぴゃ": "ppya",
+      "っぴゅ": "ppyu",
+      "っぴょ": "ppyo"
+    }
+  
+    SPECIALS = {
+      "ゃ": "ya",
+      "ゅ": "yu",
+      "ょ": "yo",
+      "っ": "",
+      "ぁ": "a",
+      "ぃ": "i",
+      "ぅ": "u",
+      "ぇ": "e",
+      "ぉ": "o",
+      "。": "."
+    }
+  
+    # Converts a hiragana string into Roman letters
+    def self.convert_hiragana(string)
+      hierarchy = [TRIPLE_CHAR_HIRAGANA_MAP, DOUBLED_CONSONANT_HIRAGANA_MAP, SMALL_COMBO_HIRAGANA_MAP, BASIC_HIRAGANA_MAP, SPECIALS]
+      hierarchy.length.times do |n|
+        # Stringify the keys so that you can use them in the Regex in the loop.
+        hierarchy[n].stringify_keys!
+        hierarchy[n].each do |k, v|
+          mtch = Regexp.new k
+          if mtch.match(string)
+            string.gsub!(mtch, v)
+          end
+        end
+      end
+      return string
+    end
+  end
+end

data/lib/japanese/verb_identifier.rb

@@ -0,0 +1,112 @@
+require 'active_support'
+require 'active_support/core_ext'
+
+module Japanese
+  module VerbIdentifier
+  
+    # List of consonant stem verbs ending in -iru
+    CONSONANT_VERBS =     %w(脂ぎる びびる 契る 散る どじる 愚痴る 入る 走る 穿る 迸る
+                            熱る いびる 弄る 炒る 要る 限る 齧る 呪る 切る 霧る 切る
+                            きしる 軋る 抉る 参る 混じる　交じる 滅入る 見縊る 漲る
+                            毟る 捩じる 握る 罵る 陥る 思い入る 思い切る せびる 知る
+                            謗る 滾る 魂消る 迸る とちる 野次る 過る 横切る 嘲る 駄弁る
+                            彫る 選る 啁る 耽る 伏せる 侍る 減る 撚る 翻る 火照る 帰る
+                            返る 反る 還る 孵る 陰る 駆ける 蹴る くねる 覆る 練る のめる
+                            滑る 阿る 競る 挵る 喋る 茂る 湿気る そべる 滑る 猛る 照る
+                            抓める 抓る うねる 蘇る 甦る 放る 括る 抉る 捥る 捩る 詰る
+                            )
+                           
+    RU_IRREGULAR_MAPPING = {
+      "する": "v-suru",
+      "来る": "v-kuru",
+      "有る": "v-aru",
+      "ある": "v-aru",
+      "居らっしゃる": "v5r-i",
+      "いらっしゃる": "v5r-i"
+    }
+  
+    AMBIGUITY_IF_HIRAGANA_ONLY = %w(いる きる える へる かえる ねる しめる)
+  
+    E_HIRAGANA = %w(え け せ て へ ね め れ げ ぜ で べ ぺ)
+    I_HIRAGANA = %w(い き し ち ひ に み り ぎ じ ぢぃ び ぴ)
+  
+    def resolve_ru_verb_class
+      return nil if ambiguous?
+      unless irregular?
+        if ends_in_iru_eru? && is_consonant_verb?
+          self.part_of_speech = "v5r"
+        elsif ends_in_ru? && !ends_in_iru_eru?
+          self.part_of_speech = "v5r"
+        elsif ends_in_iru_eru? && !is_consonant_verb?
+          self.part_of_speech = "v1"
+        end
+      else
+        RU_IRREGULAR_MAPPING.stringify_keys!
+        RU_IRREGULAR_MAPPING.each do |k, v|
+          self.part_of_speech = v if self.kanji == k
+        end
+      end
+    end
+  
+    def resolve_verb_class
+      # Check if the word is a special case of "k" or "u" verb
+      if self.kanji == "行く"
+        self.part_of_speech = "v5k-s"
+      elsif self.kanji == "問う"
+        self.part_of_speech = "v5u-s"
+      else
+        case self.kanji[-1]
+          when "ぶ"
+            self.part_of_speech = "v5b"
+          when "ぐ"
+            self.part_of_speech = "v5g"
+          when "く"
+            self.part_of_speech = "v5k" unless self.kanji == "行く"
+          when "む"
+            self.part_of_speech = "v5m"
+          when "ぬ"
+            self.part_of_speech = "v5n"
+          when "す"
+            self.part_of_speech = "v5s"
+          when "る"
+            resolve_ru_verb_class
+          when "つ"
+            self.part_of_speech = "v5t"
+          when "う"
+            self.part_of_speech = "v5u" unless self.kanji == "問う"
+        end
+      end
+    end
+  
+      # Include into the word class and use on word instances
+      def ends_in_ru?
+        unless self.kanji.blank?
+          self.kanji[-1] == "る" ? true : false
+        else 
+          self.hiragana[-1] == "る" ? true : false
+        end
+      end
+  
+      # Returns true if the class of the verb cannot be determined without more information
+      def ambiguous?
+        self.kanji == self.hiragana && self.kanji.in?(AMBIGUITY_IF_HIRAGANA_ONLY) ? true : false
+      end
+  
+      def ends_in_iru_eru?
+        self.hiragana[-2].in?(E_HIRAGANA) || self.hiragana[-2].in?(I_HIRAGANA) ? true : false
+      end
+  
+      def is_consonant_verb?
+        if self.ends_in_iru_eru? && !ambiguous?
+          self.kanji.in?(CONSONANT_VERBS) ? true : false
+        end
+      end
+  
+      # Verifies the verb is not a "ru" irregular
+      def irregular?
+        irregulars = RU_IRREGULAR_MAPPING.stringify_keys!
+        ireg = irregulars.keys
+        self.kanji.in?(ireg) ? true : false
+      end
+  end
+end

data/lib/jay_verb.rb

@@ -0,0 +1,59 @@
+require 'active_support'
+require 'active_support/core_ext'
+require 'japanese/verb_identifier'
+require 'japanese/conjugator'
+require 'japanese/to_romaji'
+
+class JayVerb
+  include Japanese::VerbIdentifier
+  include Japanese::Conjugator
+  include Japanese::ToRomaji
+    
+  attr_accessor :kanji, :hiragana, :romaji, :part_of_speech, :passive_forms,
+                :stem_form, :hiragana_forms, :romaji_forms, :negative_stem,
+                :base, :passive_dictionary_form, :causative_dictionary_form,
+                :causative_passive_dictionary_form, :causative_forms, :conjugations,
+                :causative_passive_forms_hiragana, :causative_passive_forms_romaji,
+                :passive_forms_hiragana, :passive_forms_romaji, :causative_passive_forms,
+                :causative_passive_forms_hiragana, :causative_passive_forms_romaji,
+                :causative_forms_hiragana, :causative_forms_romaji, :has_imperative,
+                :has_causative, :has_causative_passive, :has_passive, :has_volitional
+                  
+  def initialize(kanji, hiragana)
+    @kanji = kanji
+    @hiragana = hiragana
+    @romaji = Japanese::ToRomaji.convert_hiragana(String.new(hiragana))
+    @part_of_speech = self.resolve_verb_class
+    self.set_verb_behavior_types
+    @hiragana_forms = {}
+    @romaji_forms = {}
+    @passive_forms = {}
+    @causative_forms = {}
+    @conjugations = {}
+    @causative_passive_forms = {}
+    @causative_passive_forms_hiragana = {}
+    @causative_passive_forms_romaji = {}
+    @passive_forms_hiragana = {}
+    @passive_forms_romaji ={}
+    @causative_forms_hiragana = {}
+    @causative_forms_romaji = {}
+    self.process_verb
+  end
+    
+  # Defaults all has_* attributes to true for now. When self.process_verb is
+  # called, this will conjugate the verb for all possible modes. This may
+  # not make sense for all verbs —- for example, some verbs may never
+  # be realistically used in the imperative sense, such as "存在しろ!"
+  # ("Exist!"), which while grammatically correct, may sound nonsensical.
+  def set_verb_behavior_types
+    if part_of_speech.in?(%w(v1 v5b v5g v5k v5k-s v5m v5n v5r v5r-i v5s v5t v5u v5u-s v-aru v-kuru v-suru))
+      attrs = %w(imperative passive volitional causative causative_passive)
+      attrs.each do |a|
+        unless self.send("has_#{a}") == false
+          eval "self.has_#{a} = true"
+        end
+      end
+    end
+  end
+end
+# JayVerb.new("言う", "いう")

metadata

@@ -0,0 +1,47 @@
+--- !ruby/object:Gem::Specification
+name: JayVerb
+version: !ruby/object:Gem::Version
+  version: 0.0.0
+platform: ruby
+authors:
+- Zack Kayser
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2016-11-19 00:00:00.000000000 Z
+dependencies: []
+description: A gem for working with and conjugating Japanese verbs
+email: zkayser@gmail.com
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- lib/japanese/conjugator.rb
+- lib/japanese/to_romaji.rb
+- lib/japanese/verb_identifier.rb
+- lib/jay_verb.rb
+homepage: 
+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: 2.4.6
+signing_key: 
+specification_version: 4
+summary: Japanese verb conjugator
+test_files: []