medieval_latina 2.1.1 → 3.0.0

data/lexicons/Latin09.pls

@@ -0,0 +1,10 @@
+<lexicon version='1.0' xmlns='http://www.w3.org/2005/01/pronunciation-lexicon' alphabet='ipa' xml:lang='en-US'>
+  <lexeme>
+    <grapheme>
+      vulnero
+    </grapheme>
+    <phoneme>
+      ˈvul.ne.ro
+    </phoneme>
+  </lexeme>
+</lexicon>

data/lib/medieval_latina/lexicon.rb

@@ -0,0 +1,27 @@
+class MedievalLatina
+  class Lexicon
+    LEXICON_DIR = File.expand_path("../../lexicons", __dir__)
+
+    def self.list_files
+      Dir.glob(File.join(LEXICON_DIR, "*.pls"))
+    end
+
+    def self.get_file_path(file_name)
+      path = File.join(LEXICON_DIR, file_name)
+      return path if File.exist?(path)
+      raise "Lexicon file not found: #{file_name}"
+    end
+
+    def self.get_file_content(file_name)
+      path = get_file_path(file_name)
+      File.read(path)
+    end
+
+    def self.file_names_with_contents
+      list_files.each_with_object({}) do |file_path, files_content|
+        file_name, extension = File.basename(file_path).split(".")
+        files_content[file_name] = get_file_content("#{file_name}.#{extension}")
+      end
+    end
+  end
+end

data/lib/medieval_latina/lexicon_builder.rb

@@ -1,61 +1,44 @@
-require "builder"
+require "cgi"
+require "rexml/document"
 
 class MedievalLatina
   class LexiconBuilder
-    def self.write
-      new.call.write
-    end
-
-    def initialize
-      @hash = parts.each_with_object({}) do |part, hash|
-        hash.merge(part)
-      end
-
-      FREQUENCY_LIST.each do |word, metadata|
-        if metadata.key?(:ipa)
-          @hash[word] = metadata[:ipa]
-        end
-      end
+    include REXML
 
-      @xml = Builder::XmlMarkup.new(indent: 2)
+    def initialize(words)
+      @document = Document.new
+      @words = words
     end
 
     def call
-      xml.instruct! :xml, encoding: "UTF-8"
-      xml.lexicon(xmlns: URL, version: "1.0") do
-        grouped_hash.each do |phonetics, words|
-          xml.lexeme do
-            words.each { |word| xml.grapheme word }
-            xml.phoneme phonetics
-          end
-        end
-      end
+      document.add_element "lexicon", SPECIFICATION
 
-      self
-    end
+      words.each do |word, pronunciation|
+        lexeme = Element.new("lexeme")
+        grapheme = Element.new("grapheme")
+        phoneme = Element.new("phoneme")
 
-    def write
-      File.open("lexicon.pls", "w") do |file|
-        file.write(xml.target!)
-      end
-    end
-
-    private
+        grapheme.text = CGI.unescapeHTML(word)
+        phoneme.text = pronunciation
 
-    attr_reader :hash, :xml
+        lexeme.add_element(grapheme)
+        lexeme.add_element(phoneme)
 
-    def grouped_hash
-      hash.group_by do |_, phonetics|
-        phonetics
-      end.transform_values do |pairs|
-        pairs.map(&:first)
+        document.root.add_element(lexeme)
       end
-    end
 
-    def parts
-      [ADJECTIVES, ADVERBS, NOUNS, VERBS]
+      document
     end
 
-    URL = "http://www.w3.org/2005/01/pronunciation-lexicon".freeze
+    private
+
+    attr_reader :document, :words
+
+    SPECIFICATION = {
+      "version" => "1.0",
+      "xmlns" => "http://www.w3.org/2005/01/pronunciation-lexicon",
+      "alphabet" => "ipa",
+      "xml:lang" => "en-US"
+    }.freeze
   end
 end

data/lib/medieval_latina/version.rb

@@ -1,3 +1,3 @@
 class MedievalLatina
-  VERSION = "2.1.1".freeze
+  VERSION = "3.0.0".freeze
 end

data/lib/medieval_latina.rb

@@ -1,5 +1,6 @@
-require "medieval_latina/dictionary"
+require "json"
 require "medieval_latina/initializer"
+require "medieval_latina/lexicon"
 require "medieval_latina/lexicon_builder"
 require "medieval_latina/version"
 require "set"
@@ -10,8 +11,8 @@ class MedievalLatina
       if word?(string)
         metadata = DICTIONARY.fetch(string, {})
 
-        if metadata.key?(:pronunciation)
-          metadata[:pronunciation]
+        if metadata.key?("pronunciation")
+          metadata["pronunciation"]
         else
           new(string).call
         end
@@ -23,6 +24,15 @@ class MedievalLatina
     rejoin_words(prepared_words)
   end
 
+  def self.dictionary
+    @data ||= load_data
+  end
+
+  def self.load_data
+    file_path = File.join(File.dirname(__FILE__), "../data/dictionary.json")
+    JSON.parse(File.read(file_path))
+  end
+
   def self.prepare_text(text)
     text.scan(/[\p{Alnum}'-]+|[[:punct:]]+/).map do |string|
       if word?(string)
@@ -38,31 +48,47 @@ class MedievalLatina
   end
 
   def self.adjective?(word)
-    ADJECTIVES.key?(prepare_word(word))
+    adjectives.key?(prepare_word(word))
   end
 
   def self.adverb?(word)
-    ADVERBS.key?(prepare_word(word))
+    adverbs.key?(prepare_word(word))
   end
 
   def self.noun?(word)
-    NOUNS.key?(prepare_word(word))
+    nouns.key?(prepare_word(word))
   end
 
   def self.verb?(word)
-    VERBS.key?(prepare_word(word))
+    verbs.key?(prepare_word(word))
   end
 
   def self.adjectives
-    ADJECTIVES.keys
+    DICTIONARY.select do |word, metadata|
+      metadata["part"] == "Adjective"
+    end
   end
 
   def self.adverbs
-    ADVERBS.keys
+    DICTIONARY.select do |word, metadata|
+      metadata["part"] == "Adverb"
+    end
   end
 
   def self.nouns
-    NOUNS.keys
+    DICTIONARY.select do |word, metadata|
+      metadata["part"] == "Noun"
+    end
+  end
+
+  def self.pronunciations_for(words)
+    words.map(&:downcase).each_with_object({}) do |word, hash|
+      metadata = DICTIONARY[word]
+
+      if metadata && metadata["ipa"]
+        hash[word] = metadata["ipa"]
+      end
+    end
   end
 
   def self.rejoin_words(array)
@@ -75,7 +101,9 @@ class MedievalLatina
   end
 
   def self.verbs
-    VERBS.keys
+    DICTIONARY.select do |word, metadata|
+      metadata["part"] == "Verb"
+    end
   end
 
   def self.word?(string)
@@ -83,13 +111,7 @@ class MedievalLatina
   end
 
   def self.words
-    [
-      ADJECTIVES,
-      ADVERBS,
-      DICTIONARY,
-      NOUNS,
-      VERBS
-    ].flat_map(&:keys).each_with_object(Set.new) { |word, set| set.add(word) }
+    DICTIONARY.keys.to_set
   end
 
   def initialize(word)
@@ -123,6 +145,15 @@ class MedievalLatina
     x: ->(rest) { "ks" }
   }
   CONSONENT_TEAMS = {gn: "n-y", qu: "kw"}.freeze
+  PARTS_OF_SPEECH = [
+    "Adjective",
+    "Adverb",
+    "Conjunction",
+    "Noun",
+    "Preposition",
+    "Pronoun",
+    "Verb"
+  ].to_set.freeze
   SOFT_C = ["e", "i", "ae", "oe"].freeze
   SOFT_G = SOFT_C
   SOFT_T = ["i"].freeze
@@ -164,13 +195,13 @@ class MedievalLatina
     end
 
     def to_team
-      "#{character}#{rest[0]}".intern
+      :"#{character}#{rest[0]}"
     end
   end
 
   class Error < StandardError; end
 
-  DICTIONARY = FREQUENCY_LIST.each_with_object({}) do |(word, metadata), hash|
+  DICTIONARY = dictionary.each_with_object({}) do |(word, metadata), hash|
     hash[word] = metadata
 
     sanitized_word = I18n.transliterate(word)

data/medieval_latina.gemspec

@@ -15,7 +15,7 @@ Gem::Specification.new do |spec|
   spec.description = description
   spec.homepage = "https://github.com/jaysonvirissimo/medieval_latina"
   spec.license = "MIT"
-  spec.required_ruby_version = Gem::Requirement.new(">= 2.7.0")
+  spec.required_ruby_version = Gem::Requirement.new(">= 3.0.0")
   spec.metadata["allowed_push_host"] = "https://rubygems.org/"
   spec.metadata["homepage_uri"] = spec.homepage
   spec.metadata["source_code_uri"] = github_uri
@@ -29,4 +29,8 @@ Gem::Specification.new do |spec|
   spec.executables = spec.files.grep(%r{^exe/}) { |f| File.basename(f) }
   spec.require_paths = ["lib"]
   spec.add_dependency "i18n"
+  spec.add_development_dependency "rake", "~> 12.0"
+  spec.add_development_dependency "rspec", "~> 3.0"
+  spec.add_development_dependency "nokogiri"
+  spec.add_development_dependency "standard"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: medieval_latina
 version: !ruby/object:Gem::Version
-  version: 2.1.1
+  version: 3.0.0
 platform: ruby
 authors:
 - Jayson Virissimo
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2023-11-30 00:00:00.000000000 Z
+date: 2024-04-28 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: i18n
@@ -24,6 +24,62 @@ dependencies:
     - - ">="
       - !ruby/object:Gem::Version
         version: '0'
+- !ruby/object:Gem::Dependency
+  name: rake
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '12.0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '12.0'
+- !ruby/object:Gem::Dependency
+  name: rspec
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '3.0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '3.0'
+- !ruby/object:Gem::Dependency
+  name: nokogiri
+  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'
+- !ruby/object:Gem::Dependency
+  name: standard
+  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: |
   There are good text-to-speech engines for English and classical Latin, but none for medieval Latin.
   MedievalLatina converts Latin text to a kind of phonetic spelling that can be read by English text-to-speech engines.
@@ -33,6 +89,7 @@ executables: []
 extensions: []
 extra_rdoc_files: []
 files:
+- ".DS_Store"
 - ".github/workflows/tests.yml"
 - ".gitignore"
 - ".rspec"
@@ -45,11 +102,22 @@ files:
 - Rakefile
 - bin/build
 - bin/console
+- bin/lint
 - bin/setup
-- lexicon.pls
+- data/dictionary.json
+- lexicons/Latin00.pls
+- lexicons/Latin01.pls
+- lexicons/Latin02.pls
+- lexicons/Latin03.pls
+- lexicons/Latin04.pls
+- lexicons/Latin05.pls
+- lexicons/Latin06.pls
+- lexicons/Latin07.pls
+- lexicons/Latin08.pls
+- lexicons/Latin09.pls
 - lib/medieval_latina.rb
-- lib/medieval_latina/dictionary.rb
 - lib/medieval_latina/initializer.rb
+- lib/medieval_latina/lexicon.rb
 - lib/medieval_latina/lexicon_builder.rb
 - lib/medieval_latina/version.rb
 - medieval_latina.gemspec
@@ -69,14 +137,14 @@ required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
-      version: 2.7.0
+      version: 3.0.0
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.4.1
+rubygems_version: 3.5.3
 signing_key:
 specification_version: 4
 summary: Transform medieval Latin text into phonetic English