kampyo 0.2.0 → 1.0.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 0d96499aeb214750eddaa4eecdede3a629b05dcd05755c9a3ca798be84f400e3
-  data.tar.gz: '094e4f87fe8c86726c46deb081c4d2fd4f7e356c62d1dfbd15384c9bb3eca6f0'
+  metadata.gz: 7f831a683b8d1f6bf690fffe535b7f301c4e2ed9de3b9bbb58efe65b12c7e38b
+  data.tar.gz: 62e652f7d9a8d0d953247b2b2e8926844dc92ea668d3b94a102ac577c57c2101
 SHA512:
-  metadata.gz: cd8aab55671360c9acb592050383f717fe1a93c4efd3233943b2bd3b7ffb7585c90090561b11c0dde13010c44743ab2e89842b4f317d1428d37b58a1438a0a3a
-  data.tar.gz: 3d85e4fed2888ab5c3322da3685bdedafbfc8fb8c79c6b9f17eb578c4fe1ec793b9c85a34b803b3f1d58e722be3c109a935bf1a5b4f7078bbb005e341ae285cb
+  metadata.gz: eb42364413b79900b512d8e830091334ffd7bee17ac19cf2e3b1c1387139861a7d0324a3ab05c65f3a45f8f690286cf3ade7ac2d0735cea0c971bedf09f46644
+  data.tar.gz: f18179269d98c2ff02c1bd18d928da1dc0190e4d9fbdad22f88b35c10a918d4b04f6fc2ae1d925bfe0bfe576fcb26e3c08bb92432ceb138cfabab1accb11e8f5

data/README.base.md

@@ -20,8 +20,8 @@ Unique features already implemented are subject and predicate, and stylistic inf
 Basic usage with Cabocha.
 
 ```
-text = Kampyo::Text.new
-text.cabocha_parser("今日は雨です")
+cabocha = Kampyo::Cabocha.new
+cabocha.parser("今日は雨です")
 ```
 
 You will get the following result.
@@ -38,8 +38,8 @@ You will get the following result.
 Guess the subject, predicate and sentence system.
 
 ```
-text = Kampyo::Text.new
-text.analysis(text.cabocha_parser("今日は雨です"))
+cabocha = Kampyo::Cabocha.new
+cabocha.analysis(text.parser("今日は雨です"))
 ```
 
 You will get the following result.

data/README.ja.md

@@ -15,8 +15,8 @@ Unique features already implemented are subject and predicate, and stylistic inf
 Basic usage with Cabocha.
 
 ```
-text = Kampyo::Text.new
-text.cabocha_parser("今日は雨です")
+cabocha = Kampyo::Cabocha.new
+cabocha.parser("今日は雨です")
 ```
 
 You will get the following result.
@@ -33,8 +33,8 @@ You will get the following result.
 Guess the subject, predicate and sentence system.
 
 ```
-text = Kampyo::Text.new
-text.analysis(text.cabocha_parser("今日は雨です"))
+cabocha = Kampyo::Cabocha.new
+cabocha.analysis(text.parser("今日は雨です"))
 ```
 
 You will get the following result.

data/README.md

@@ -15,8 +15,8 @@ Unique features already implemented are subject and predicate, and stylistic inf
 Basic usage with Cabocha.
 
 ```
-text = Kampyo::Text.new
-text.cabocha_parser("今日は雨です")
+cabocha = Kampyo::Cabocha.new
+cabocha.parser("今日は雨です")
 ```
 
 You will get the following result.
@@ -33,8 +33,8 @@ You will get the following result.
 Guess the subject, predicate and sentence system.
 
 ```
-text = Kampyo::Text.new
-text.analysis(text.cabocha_parser("今日は雨です"))
+cabocha = Kampyo::Cabocha.new
+cabocha.analysis(text.parser("今日は雨です"))
 ```
 
 You will get the following result.

data/lib/kampyo/cabocha.rb

@@ -0,0 +1,172 @@
+# frozen_string_literal: true
+
+require 'kampyo'
+require 'kampyo/string'
+require 'kampyo/text'
+require 'cabocha'
+
+# Kampyo
+module Kampyo
+  # Text
+  class Cabocha < Text # rubocop:disable Metrics/ClassLength
+    # rubocop:todo Metrics/MethodLength
+    def parser(input) # rubocop:todo Metrics/AbcSize, Metrics/MethodLength
+      # <sentence>
+      #  <chunk id="0" link="1" rel="D" score="0.000000" head="0" func="1">
+      #   <tok id="0" feature="名詞,副詞可能,*,*,*,*,今日,キョウ,キョー" ne="B-DATE">今日</tok>
+      #   <tok id="1" feature="助詞,係助詞,*,*,*,*,は,ハ,ワ" ne="O">は</tok>
+      #  </chunk>
+      #  <chunk id="1" link="-1" rel="D" score="0.000000" head="2" func="3">
+      #   <tok id="2" feature="名詞,一般,*,*,*,*,雨,アメ,アメ" ne="O">雨</tok>
+      #   <tok id="3" feature="助動詞,*,*,*,特殊・デス,基本形,です,デス,デス" ne="O">です</tok>
+      #  </chunk>
+      # </sentence>
+
+      parser = CaboCha::Parser.new
+      tree = parser.parse(input)
+
+      chunks = []
+      tokens = []
+      token_position = 0
+      (0..tree.chunk_size - 1).each do |i| # rubocop:todo Metrics/BlockLength
+        chunk = tree.chunk(i)
+        token_size = chunk.token_size
+
+        chunks << {
+          id: chunks.size + 1,
+          link: chunk.link >= 0 ? chunk.link + 1 : -1,
+          score: chunk.score
+        }
+
+        (token_position..token_position + token_size - 1).each do |j|
+          token = tree.token(j)
+
+          surface = token.surface.to_utf8
+          feature0 = token.feature_list(0).to_utf8
+          feature1 = token.feature_list(1).to_utf8
+          feature6 = token.feature_list(6).to_utf8
+          feature7 = token.feature_list(7).to_utf8
+
+          tokens << {
+            id: tokens.size + 1,
+            chunk: i + 1,
+            surface: surface,
+            feature1: feature0,
+            feature2: feature1,
+            baseform: feature6,
+            reading: feature7,
+            ext_reading: ext_reading(feature7)
+          }
+        end
+
+        token_position += token_size
+      end
+
+      { chunks: chunks, tokens: tokens }
+    end
+    # rubocop:enable Metrics/MethodLength
+
+    def analysis(cabocha) # rubocop:todo Metrics/CyclomaticComplexity, Metrics/AbcSize, Metrics/MethodLength, Metrics/PerceivedComplexity
+      chunks = cabocha[:chunks]
+      tokens = cabocha[:tokens]
+
+      subject_token = nil
+      predicate_token = nil
+      tod = '*'
+
+      # 述語の候補
+      predicate_chunk = chunks.find { |item| item[:link] == -1 }
+      predicate_tokens = tokens.select { |item| item[:chunk] == predicate_chunk[:id] }
+
+      # 主語の候補
+      subject_chunk = chunks.find do |item|
+        item[:link] == predicate_chunk[:id]
+      end
+      unless subject_chunk.nil?
+        subject_tokens = tokens.select do |item|
+          item[:chunk] == subject_chunk[:id] &&
+            item[:feature1] == '名詞' &&
+            %w[一般 固有名詞 サ変接続 接尾 数 副詞可能].include?(item[:feature2])
+        end
+
+        subject_tokens.each do |token|
+          next_token = tokens.find do |item|
+            item[:id] == token[:id] + 1 &&
+              item[:chunk] == token[:chunk] &&
+              %w[は って も が].include?(item[:baseform])
+          end
+
+          next if next_token.nil?
+
+          # 主語として確定する
+          subject_token = token
+          next
+        end
+      end
+
+      predicate_tokens.each do |token| # rubocop:todo Metrics/BlockLength
+        if %w[形容詞 動詞 名詞].include?(token[:feature1]) &&
+           %w[一般 自立 サ変接続 接尾].include?(token[:feature2]) &&
+           token[:baseform] != '*'
+          # 述語として確定する
+
+          predicate_token = token
+        end
+
+        next unless %w[動詞 終助詞 助動詞 助詞].include?(token[:feature1])
+
+        # 文体系を確定する
+        tods = {
+          'れる' => '受身・尊敬・可能・自発',
+          'られる' => '受身・尊敬・可能・自発',
+          'せる' => '使役',
+          'させる' => '使役',
+          'ない' => '打消',
+          'ぬ' => '打消',
+          'ん' => '打消',
+          'う' => '推量・意志・勧誘',
+          'よう' => '推量・意志・勧誘',
+          'まい' => '打消推量・打消意志',
+          'たい' => '希望',
+          'たがる' => '希望',
+          'た' => '過去・完了・存在・確認',
+          'ます' => '丁寧',
+          'そうだ' => '様態・伝聞',
+          'らしい' => '推定',
+          'ようだ' => '比況・例示・推定',
+          'だ' => '断定',
+          'です' => '断定',
+          'な' => '禁止・感動',
+          'か' => '疑問・反語・感動',
+          'の' => '断定・質問',
+          'よ' => '強意・呼びかけ',
+          'ぞ' => '強意',
+          'も' => '確信',
+          'ね' => '感動・念押し',
+          'わ' => '感動・強意',
+          'さ' => '断定'
+        }
+
+        tod = tods[token[:baseform]]
+      end
+
+      last_token = predicate_tokens.last
+
+      if predicate_token.nil? && ['助動詞'].include?(last_token[:feature1])
+        # 述語が確定していないとき最後の助動詞を述語として確定する
+        predicate_token = last_token
+      end
+
+      if ['?', '？'].include?(last_token[:surface])
+        # 述語の最後の形態素が?のとき
+        tod = '疑問・反語・感動'
+      end
+
+      {
+        subject: subject_token,
+        predicate: predicate_token,
+        tod: tod
+      }
+    end
+  end
+end

data/lib/kampyo/mecab.rb

@@ -0,0 +1,42 @@
+# frozen_string_literal: true
+
+require 'kampyo'
+require 'kampyo/string'
+require 'kampyo/text'
+require 'mecab'
+
+# Kampyo
+module Kampyo
+  # Text
+  class Mecab < Text
+    # rubocop:todo Metrics/MethodLength
+    def parser(input) # rubocop:todo Metrics/AbcSize, Metrics/MethodLength
+      result = []
+      parser = MeCab::Tagger.new
+      node = parser.parseToNode(input)
+      while node
+        features = node.feature.split(',')
+        if features[0] != 'BOS/EOS'
+          result << {
+            id: result.size + 1,
+            chunk: 0,
+            surface: node.surface,
+            feature1: features[0],
+            feature2: features[1],
+            baseform: features[6],
+            reading: features[7],
+            ext_reading: ext_reading(features[7]),
+            cost: node.cost,
+            wcost: node.wcost,
+            right_context: node.rcAttr,
+            left_context: node.lcAttr
+          }
+        end
+        node = node.next
+      end
+
+      result
+    end
+    # rubocop:enable Metrics/MethodLength
+  end
+end

data/lib/kampyo/string.rb

@@ -1,5 +1,6 @@
 # frozen_string_literal: true
 
+# String
 class String
   def to_utf8
     force_encoding('UTF-8')

data/lib/kampyo/text.rb

@@ -2,201 +2,15 @@
 
 require 'kampyo'
 require 'kampyo/string'
-require 'cabocha'
-require 'mecab'
 
+# Kampyo
 module Kampyo
+  # Text
   class Text
     def initialize; end
 
-    def cabocha_parser(input)
-      # <sentence>
-      #  <chunk id="0" link="1" rel="D" score="0.000000" head="0" func="1">
-      #   <tok id="0" feature="名詞,副詞可能,*,*,*,*,今日,キョウ,キョー" ne="B-DATE">今日</tok>
-      #   <tok id="1" feature="助詞,係助詞,*,*,*,*,は,ハ,ワ" ne="O">は</tok>
-      #  </chunk>
-      #  <chunk id="1" link="-1" rel="D" score="0.000000" head="2" func="3">
-      #   <tok id="2" feature="名詞,一般,*,*,*,*,雨,アメ,アメ" ne="O">雨</tok>
-      #   <tok id="3" feature="助動詞,*,*,*,特殊・デス,基本形,です,デス,デス" ne="O">です</tok>
-      #  </chunk>
-      # </sentence>
-
-      parser = CaboCha::Parser.new
-      tree = parser.parse(input)
-
-      chunks = []
-      tokens = []
-      token_position = 0
-      (0..tree.chunk_size - 1).each do |i|
-        chunk = tree.chunk(i)
-        token_size = chunk.token_size
-
-        chunks << {
-          id: chunks.size + 1,
-          link: chunk.link >= 0 ? chunk.link + 1 : -1,
-          score: chunk.score
-        }
-
-        (token_position..token_position + token_size - 1).each do |j|
-          token = tree.token(j)
-
-          surface = token.surface.to_utf8
-          feature0 = token.feature_list(0).to_utf8
-          feature1 = token.feature_list(1).to_utf8
-          feature6 = token.feature_list(6).to_utf8
-          feature7 = token.feature_list(7).to_utf8
-
-          tokens << {
-            id: tokens.size + 1,
-            chunk: i + 1,
-            surface: surface,
-            feature1: feature0,
-            feature2: feature1,
-            baseform: feature6,
-            reading: feature7,
-            ext_reading: ext_reading(feature7)
-          }
-        end
-
-        token_position += token_size
-      end
-
-      { chunks: chunks, tokens: tokens }
-    end
-
-    def mecab_parser(input)
-      result = []
-      parser = MeCab::Tagger.new
-      node = parser.parseToNode(input)
-      while node
-        features = node.feature.split(',')
-        if features[0] != 'BOS/EOS'
-          result << {
-            id: result.size + 1,
-            chunk: 0,
-            surface: node.surface,
-            feature1: features[0],
-            feature2: features[1],
-            baseform: features[6],
-            reading: features[7],
-            ext_reading: ext_reading(features[7]),
-            cost: node.cost,
-            wcost: node.wcost,
-            right_context: node.rcAttr,
-            left_context: node.lcAttr
-          }
-        end
-        node = node.next
-      end
-
-      result
-    end
-
     def ext_reading(feature)
       (feature =~ /\A[\p{katakana}|ー]+\z/).nil? ? feature : nil
     end
-
-    def analysis(cabocha)
-      chunks = cabocha[:chunks]
-      tokens = cabocha[:tokens]
-
-      subject_token = nil
-      predicate_token = nil
-      tod = '*'
-
-      # 述語の候補
-      predicate_chunk = chunks.find { |item| item[:link] == -1 }
-      predicate_tokens = tokens.select { |item| item[:chunk] == predicate_chunk[:id] }
-
-      # 主語の候補
-      subject_chunk = chunks.find do |item|
-        item[:link] == predicate_chunk[:id]
-      end
-      unless subject_chunk.nil?
-        subject_tokens = tokens.select do |item|
-          item[:chunk] == subject_chunk[:id] &&
-            item[:feature1] == '名詞' &&
-            %w[一般 固有名詞 サ変接続 接尾 数 副詞可能].include?(item[:feature2])
-        end
-
-        subject_tokens.each do |token|
-          next_token = tokens.find do |item|
-            item[:id] == token[:id] + 1 &&
-              item[:chunk] == token[:chunk] &&
-              %w[は って も が].include?(item[:baseform])
-          end
-
-          next if next_token.nil?
-
-          # 主語として確定する
-          subject_token = token
-          next
-        end
-      end
-
-      predicate_tokens.each do |token|
-        if %w[形容詞 動詞 名詞].include?(token[:feature1]) &&
-           %w[一般 自立 サ変接続 接尾].include?(token[:feature2]) &&
-           token[:baseform] != '*'
-          # 述語として確定する
-
-          predicate_token = token
-        end
-
-        next unless %w[動詞 終助詞 助動詞 助詞].include?(token[:feature1])
-
-        # 文体系を確定する
-        tods = {
-          'れる' => '受身・尊敬・可能・自発',
-          'られる' => '受身・尊敬・可能・自発',
-          'せる' => '使役',
-          'させる' => '使役',
-          'ない' => '打消',
-          'ぬ' => '打消',
-          'ん' => '打消',
-          'う' => '推量・意志・勧誘',
-          'よう' => '推量・意志・勧誘',
-          'まい' => '打消推量・打消意志',
-          'たい' => '希望',
-          'たがる' => '希望',
-          'た' => '過去・完了・存在・確認',
-          'ます' => '丁寧',
-          'そうだ' => '様態・伝聞',
-          'らしい' => '推定',
-          'ようだ' => '比況・例示・推定',
-          'だ' => '断定',
-          'です' => '断定',
-          'な' => '禁止・感動',
-          'か' => '疑問・反語・感動',
-          'の' => '断定・質問',
-          'よ' => '強意・呼びかけ',
-          'ぞ' => '強意',
-          'も' => '確信',
-          'ね' => '感動・念押し',
-          'わ' => '感動・強意',
-          'さ' => '断定'
-        }
-
-        tod = tods[token[:baseform]]
-      end
-
-      last_token = predicate_tokens.last
-
-      if predicate_token.nil? && ['助動詞'].include?(last_token[:feature1])
-        # 述語が確定していないとき最後の助動詞を述語として確定する
-        predicate_token = last_token
-      end
-
-      if ['?', '？'].include?(last_token[:surface])
-        # 述語の最後の形態素が?のとき
-        tod = '疑問・反語・感動'
-      end
-
-      {
-        subject: subject_token,
-        predicate: predicate_token,
-        tod: tod
-      }
-    end
   end
 end

data/lib/kampyo/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Kampyo
-  VERSION = '0.2.0'
+  VERSION = '1.0.0'
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: kampyo
 version: !ruby/object:Gem::Version
-  version: 0.2.0
+  version: 1.0.0
 platform: ruby
 authors:
 - arthur87
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2025-03-12 00:00:00.000000000 Z
+date: 2025-06-14 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rubocop
@@ -66,6 +66,8 @@ files:
 - README.md
 - Rakefile
 - lib/kampyo.rb
+- lib/kampyo/cabocha.rb
+- lib/kampyo/mecab.rb
 - lib/kampyo/string.rb
 - lib/kampyo/text.rb
 - lib/kampyo/version.rb