interscript 0.1.3 → 0.1.9

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 643981da933194b2464ea279e9d31b9fcd9d32519c5cd236ed805855c93755ad
-  data.tar.gz: f54c4303bb02f0a873cfdf96287d78321648cee19c685bf338cb9f8e2f642c56
+  metadata.gz: 7813dba0d0cc7493ed3b9279c61283c8d305f1e05584a44aa700e9b72acb2f06
+  data.tar.gz: f4c87e24d7c2719b4f358198967d55e0c17d8aaac354311cb9eecc800a592b2d
 SHA512:
-  metadata.gz: 2d8cfd0d60e2d41d8b1e31b4e61353b0bc7fd5ac4fc426d4304ccc86bc0bb6d84b4b4a2a6e44bb342afa6c20202a4bca4180a1f5037c73072e246038c6f36f1f
-  data.tar.gz: 2a5fffac1de98702494f69d55b2de5200684195b0f7948619bfa2ae9f3f97810c731868f2550578f5ad97a9db9fa72d9c2abad24451437b7e08673dfc1cd97d8
+  metadata.gz: 8e23000fe8fb016dec9351241787608c892d7af48682259c10345ce417f94a626ea06a5bc8b1e7a3f084da8c0502d13b96bd06c9a53da31530f5c487b06fe4e9
+  data.tar.gz: b2d8cb122b2c1bbb2d989d832802a007a60ecc6d0f2984e323a983ec94ae9505664b98e0d9c33e0b2711f5fa31f4e657a720648dcb572915f69006a1719a9610

data/README.adoc

@@ -1,6 +1,7 @@
 = Interscript: Interoperable Script Conversion Systems, with a Ruby implementation
 
-image:https://github.com/interscript/interscript/workflows/test/badge.svg["Build Status", link="https://github.com/interscript/interscript/actions?workflow=test"]
+image:https://github.com/interscript/interscript/workflows/test/badge.svg["Ruby build status", link="https://github.com/interscript/interscript/actions?workflow=test"]
+image:https://github.com/interscript/interscript/workflows/js/badge.svg["JavaScript build status", link="https://github.com/interscript/interscript/actions?workflow=js"]
 
 == Introduction
 
@@ -22,9 +23,9 @@ The goal is to achieve interoperable transliteration schemes allowing quality co
 These transliteration systems are used in the demo:
 
 `bgnpcgn-rus-Cyrl-Latn-1947`:: BGN/PCGN Romanization of Russian
-`iso-rus-Cyrl-Latn-iso9`::     ISO 9 Romanization of Russian
+`iso-rus-Cyrl-Latn-9-1995`::     ISO 9 Romanization of Russian
 `icao-rus-Cyrl-Latn-9303`::    ICAO MRZ Romanization of Russian
-`bas-rus-Cyrl-Latn-bss`::      Bulgaria Academy of Science Streamlined System for Russian
+`bas-rus-Cyrl-Latn-2017-bss`::      Bulgaria Academy of Science Streamlined System for Russian
 
 image:demo/20191118-interscript-demo-cast.gif["interscript screencast"]
 
@@ -51,9 +52,7 @@ Interscript depends on Python and the https://github.com/sequitur-g2p/sequitur-g
 
 [source,sh]
 ----
-pip3 install setuptools numpy
-curl -sSL -o sequitur-g2p.zip https://github.com/sequitur-g2p/sequitur-g2p/archive/806273f.zip
-pip3 install sequitur-g2p.zip
+pip3 install -r requirments.txt
 ----
 
 Interscript depends on Ruby. Once you manage to install Ruby, it's easy.
@@ -95,7 +94,7 @@ interscript rus-Cyrl.txt \
   --output=bgnpcgn-rus-Latn.txt
 
 interscript rus-Cyrl.txt \
-  --system=iso-rus-Cyrl-Latn-iso9 \
+  --system=iso-rus-Cyrl-Latn-9-1995 \
   --output=iso-rus-Latn.txt
 
 interscript rus-Cyrl.txt \
@@ -103,7 +102,7 @@ interscript rus-Cyrl.txt \
   --output=icao-rus-Latn.txt
 
 interscript rus-Cyrl.txt \
-  --system=bas-rus-Cyrl-Latn-bss \
+  --system=bas-rus-Cyrl-Latn-2017-bss \
   --output=bas-rus-Latn.txt
 ----
 
@@ -149,7 +148,7 @@ tests:
 
 map:
   rules:
-    - pattern: (?<=[АаЕеЁёИиОоУуЫыЭэЮюЯяЙйЪъЬь])\u0415 # Е after a, e, ё, и, о, у, ы, э, ю, я, й, ъ, ь
+    - pattern: (?<=[АаЕеЁёИиОоУуЫыЭэЮюЯяЙйЪъЬь])\u0415 # Е after a, e, ё, и, о, у, ы, э, ю, я, й, ъ, ь
       result: Ye
     - pattern: \b\u0415 # Е initially
       result: Ye
@@ -167,7 +166,7 @@ The subsection `rules` is placed under the `map` key. All rules are applied in o
 
 Each rule has `pattern` and `result` elements.
 
-Pattern is a regex expression. It should be representing as a string without `//` or `%r{}` parentheses. For example `\b\u0415`. In case a rule is depend on previous or next content, lookahead or lookbehind could be used. For example a rule with the pattern `(?<=[АаЕеЁёИиОоУуЫыЭэЮюЯяЙйЪъЬь])\u0415` find every Е after upper or lower case symbols a, e, ё, и, о, у, ы, э, ю, я, й, ъ, ь.
+Pattern is a regex expression. It should be representing as a string without `//` or `%r{}` parentheses. For example `\b\u0415`. In case a rule is depend on previous or next content, lookahead or lookbehind could be used. For example a rule with the pattern `(?<=[АаЕеЁёИиОоУуЫыЭэЮюЯяЙйЪъЬь])\u0415` find every Е after upper or lower case symbols a, e, ё, и, о, у, ы, э, ю, я, й, ъ, ь.
 
 Result is a replacement a for pattern's match. It can contain a string, an Unicode characters specified by a hexadecimal number, a captured group reference. String with hexadecimal number or captured group reference should be double quoted. For example `"Y\u00eb"` or `"\\1\u00b7\\2"`. Captured group are referred by double backslash and group's number.
 
@@ -256,7 +255,7 @@ the system code identifying a script conversion system has the following compone
 e.g. `bgnpcgn-rus-Cyrl-Latn-1947`:
 
 `bgnpcgn`:: the authority identifier
-`rus`:: an ISO 639-2 3-letter language code that this system applies to
+`rus`:: an ISO 639-{1,2,3,5} language code that this system applies to (For 639-2, use (T) code)
 `Cyrl`:: an ISO 15924 script code, identifying the source script
 `Latn`:: an ISO 15924 script code, identifying the target script
 `1947`:: an identifier unit within the authority to identify this system

data/aliases.json

@@ -0,0 +1 @@
+{"pan_Deva2Latn_ALA_1997":"alalc-pan-Guru-Latn-1997","kor_Hang2Latn_ALA_1997":"alalc-kor-Hang-Latn-1997","asm_Deva2Latn_ALA_1997":"alalc-asm-Deva-Latn-1997","aze_Cyrl2Latn_ALA_1997":"alalc-aze-Cyrl-Latn-1997","ukr_Cyrl2Latn_GUP_1996":"ua-ukr-Cyrl-Latn-1996","tha_Thai2Latn_RIT_1968":"royin-tha-Thai-Latn-1968","bul_Cyrl2Latn_BGN_1952":"bgnpcgn-bul-Cyrl-Latn-1952","tam_Taml2Latn_ALA_1997":"alalc-tam-Taml-Latn-1997","kor_Hang2Latn_GKN_2002":"kp-kor-Hang-Latn-2002","ell_Grek2Latn_ELOT743_1996":"bgnpcgn-ell-Grek-Latn-1996","zho_Hani2Latn_AcadSin_2002":"acadsin-zho-Hani-Latn-2002","ara_Arab2Latn_SES_1930":"ses-ara-Arab-Latn-1930","tgk_Cyrl2Latn_BGN_1994":"bgnpcgn-tgk-Cyrl-Latn-1994","fas_Arab2Latn_BGN_1958":"bgnpcgn-fas-Arab-Latn-1956","sin_Sinh2Latn_ALA_1997":"alalc-sin-Sinh-Latn-1997","uas_Arab2Latn_BGN_2007":"bgnpcgn-urd-Arab-Latn-2007","ukr_Cyrl2Latn_ALA_1997":"alalc-ukr-Cyrl-Latn-1997","bak_Cyrl2Latn_BGN_2007":"bgnpcgn-bak-Cyrl-Latn-2007","tam_Taml2Latn_ALA_2011":"alalc-tam-Taml-Latn-2011","ara_Arab2Latn_BGN_1956":"bgnpcgn-ara-Arab-Latn-1956","ell_Grek2Latn_ALA_1997":"alalc-ell-Grek-Latn-1997","rus_Cyrl2Latn_GOST_1983":"gost-rus-Cyrl-Latn-16876-71-1983","mar_Deva2Latn_ALA_1997":"alalc-mar-Deva-Latn-1997","bel_Cyrl2Latn_ALA_1997":"alalc-bel-Cyrl-Latn-1997","kat_Geor2Latn_ALA_1997":"alalc-kat-Geor-Latn-1997","bul_Cyrl2Latn_ALA_1997":"alalc-bul-Cyrl-Latn-1997","ara_Arab2Latn_ALA_1997":"alalc-ara-Arab-Latn-1997","mon_Cyrl2Latn_ALA_1997":"alalc-mon-Cyrl-Latn-1997","div_Thaa2Latn_GMV_1988":"bgnpcgn-div-Thaa-Latn-1988","hin_Deva2Latn_ALA_1997":"alalc-hin-Deva-Latn-1997","bel_Cyrl2Latn_GBO_1998":"by-bel-Cyrl-Latn-1998","ukr_Cyrl2Latn_BGN_1965":"bgnpcgn-ukr-Cyrl-Latn-1965","rus_Cyrl2Latn_ALA_1997":"alalc-rus-Cyrl-Latn-1997","tir_Thai2Latn_RIT_2000":"royin-tha-Thai-Latn-1999","guj_Gujr2Latn_ALA_1997":"alalc-guj-Gujr-Latn-1997","tel_Telu2Latn_ALA_1997":"alalc-tel-Telu-Latn-1997","mkd_Cyrl2Latn_BGN_1981":"bgnpcgn-mkd-Cyrl-Latn-1981","ori_Orya2Latn_ALA_2011":"alalc-ori-Orya-Latn-2011","aze_Arab2Latn_ALA_1997":"alalc-aze-Arab-Latn-1997","ori_Orya2Latn_ALA_1997":"alalc-ori-Orya-Latn-1997","div_Thaa2Latn_ALA_1997":"alalc-div-Thaa-Latn-1997","rue_Cyrl2Latn_BGN_2016":"bgnpcgn-rue-Cyrl-Latn-2016","guj_Gujr2Latn_ALA_2011":"alalc-guj-Gujr-Latn-2011","kat_Geor2Latn_BGN_1981":"bgnpcgn-kat-Geor-Latn-1981","kor_Hang2Latn_MOCT_2000":"moct-kor-Hang-Latn-2000","sin_Sinh2Latn_ALA_2011":"alalc-sin-Sinh-Latn-2011","amh_Ethi2Latn_BGN_1967":"bgnpcgn-amh-Ethi-Latn-1967","srp_Cyrl2Latn_BGN_2005":"bgnpcgn-srp-Cyrl-Latn-2005","srp_Cyrl2Latn_ALA_1997":"alalc-srp-Cyrl-Latn-1997","mal_Mlym2Latn_ALA_2012":"alalc-mal-Mlym-Latn-2012","kat_Geor2Latn_GGG_2002":"ggg-kat-Geor-Latn-2002","mon_Cyrl2Latn_BGN_1964":"bgnpcgn-mon-Cyrl-Latn-1964","mal_Mlym2Latn_ALA_1997":"alalc-mal-Mlym-Latn-1997","ben_Beng2Latn_ALA_1997":"alalc-ben-Beng-Latn-1997","kor_Hang2Latn_MR_1939":"bgn-kor-Hang-Latn-1943","zho_Hani2Latn_GCH_1979":"sac-zho-Hans-Latn-1979","bul_Cyrl2Latn_BGN_2013":"bgnpcgn-bul-Cyrl-Latn-2013","ell_Grek2Latn_BGN_1962":"bgnpcgn-ell-Grek-Latn-1962","amh_Ethi2Latn_ALA_1997":"alalc-amh-Ethi-Latn-1997","pan_Deva2Latn_ALA_2011":"alalc-pan-Guru-Latn-2011","zho_Hani2Latn_WDG_1979":"var-zho-Hani-Latn-wd-1979","rus_Cyrl2Latn_BGN_1947":"bgnpcgn-rus-Cyrl-Latn-1947","bel_Cyrl2Latn_BGN_1979":"bgnpcgn-bel-Cyrl-Latn-1979","tat_Cyrl2Latn_BGN_2005":"bgnpcgn-tat-Cyrl-Latn-2007"}

data/lib/interscript.rb

@@ -1,56 +1,33 @@
 # frozen_string_literal: true
 
-require "yaml"
 require "interscript/mapping"
 
 # Transliteration
 module Interscript
 
-  class << self
-    def root_path
-      @root_path ||= Pathname.new(File.dirname(__dir__))
-    end
-
-    def transliterate_file(system_code, input_file, output_file, maps)
-      input = File.read(input_file)
-      output = transliterate(system_code, input, maps)
-
-      File.open(output_file, 'w') do |f|
-        f.puts(output)
-      end
-      puts "Output written to: #{output_file}"
-    end
+  class InvalidSystemError < StandardError; end
+  class ExternalProcessNotRecognizedError < StandardError; end
+  class ExternalProcessUnavailableError < StandardError; end
 
-    def import_python_modules
-      begin
-        pyimport :g2pwrapper
-      rescue
-        pyimport :sys
-        sys.path.append(root_path.to_s+"/lib/")
-        pyimport :g2pwrapper
-      end
-    end
+  if RUBY_ENGINE == 'opal'
+    require "interscript/opal"
+    extend Opal
+  else
+    require "interscript/fs"
+    extend Fs
+  end
 
-    def external_process(process_name, string)
-      import_python_modules
-      case process_name
-        when 'sequitur.pythainlp_lexicon'
-          return g2pwrapper.transliterate('pythainlp_lexicon', string)
-        when 'sequitur.wiktionary_phonemic'
-          return g2pwrapper.transliterate('wiktionary_phonemic', string)
-      else
-        puts "Invalid Process"
-      end
-    end
+  class << self
 
     def transliterate(system_code, string, maps={})
-      if (!maps.has_key?system_code)
+      system_code = map_resolve(system_code)
+
+      unless maps.has_key? system_code
         maps[system_code] = Interscript::Mapping.for(system_code)
       end
       # mapping = Interscript::Mapping.for(system_code)
       mapping = maps[system_code]
 
-
       # First, apply chained transliteration as specified in the list `chain`
       chain = mapping.chain.dup
       while chain.length > 0
@@ -63,17 +40,11 @@ module Interscript
       title_case = mapping.title_case
       downcase = mapping.downcase
 
-      # charmap = mapping.characters&.sort_by { |k, _v| k.size }&.reverse&.to_h
-      # dictmap = mapping.dictionary&.sort_by { |k, _v| k.size }&.reverse&.to_h
       charmap = mapping.characters_hash
       dictmap = mapping.dictionary_hash
       trie = mapping.dictionary_trie
 
-      # Segmentation
-      string = external_process(mapping.segmentation, string) if mapping.segmentation
-
-      # Transliteration/Transcription
-      string = external_process(mapping.transcription, string) if mapping.transcription
+      string = external_processing(mapping, string)
 
       pos = 0
       while pos < string.to_s.size
@@ -81,14 +52,15 @@ module Interscript
         wordmatch = ""
 
         # Using Trie, find the longest matching substring
-        while (pos + m < string.to_s.size) && (trie.partial_word?string[pos..pos+m]) 
+        while (pos + m < string.to_s.size) && (trie.partial_word?string[pos..pos+m])
           wordmatch = string[pos..pos+m] if trie.word?string[pos..pos+m]
           m += 1
         end
+
         m = wordmatch.length
         if m > 0
           repl = dictmap[string[pos..pos+m-1]]
-          string[pos..pos+m-1] = repl
+          string = sub_replace(string, pos, m, repl)
           pos += repl.length
         else
           pos += 1
@@ -109,32 +81,62 @@ module Interscript
       #     offsets[pos] += result.size - match[0].size
       #   end
       # end
+
       mapping.rules.each do |r|
-        output.gsub!(/#{r['pattern']}/, r['result'])
+        next unless output
+        re = mkregexp(r["pattern"])
+        output = output.gsub(re, r["result"])
       end
 
       charmap.each do |k, v|
-        while (match = output&.match(/#{k}/))
+        re = mkregexp(k)
+        while (match = output&.match(re))
           pos = match.offset(0).first
           result = !downcase && up_case_around?(output, pos) ? v.upcase : v
-          result = result[0] if result.is_a?(Array) # if more than one, choose the first one
-          output[pos, match[0].size] = add_separator(separator, pos, result)
+
+          # if more than one, choose the first one
+          result = result[0] if result.is_a?(Array)
+
+          output = sub_replace(
+            output,
+            pos,
+            match[0].size,
+            add_separator(separator, pos, result)
+          )
         end
       end
 
       mapping.postrules.each do |r|
-        output.gsub!(/#{r['pattern']}/, r['result'])
+        next unless output
+        re = mkregexp(r["pattern"])
+        output = if r["result"] == "upcase"
+                   output.gsub(re, &:upcase)
+                 else
+                   output.gsub(re, r["result"])
+                 end
       end
 
-      if output
-        output.sub!(/^(.)/,  &:upcase) if title_case
-        if word_separator != ''
-          output.gsub!(/#{word_separator}#{separator}/,word_separator)
-          output.gsub!(/#{word_separator}(.)/, &:upcase) if title_case
+      return unless output
+
+      re = mkregexp('^(.)')
+      output = output.gsub(re, &:upcase) if title_case
+      if word_separator != ''
+        re = mkregexp("#{word_separator}#{separator}")
+        output = output.gsub(re, word_separator)
+
+        if title_case
+          re = mkregexp("#{word_separator}(.)")
+          output = output.gsub(re, &:upcase)
         end
       end
 
-      output ? output.unicode_normalize : output
+      output.unicode_normalize
+    end
+
+    def map_resolve(map)
+      map = aliases[map] if aliases.key? map
+      raise ArgumentError, "Map #{map} doesn't exist" unless map_exist? map
+      map
     end
 
     private
@@ -147,11 +149,11 @@ module Interscript
       return false if string[pos] == string[pos].downcase
 
       i = pos - 1
-      i -= 1 while i.positive? && string[i] !~ /[[:alpha:]]/
+      i -= 1 while i.positive? && string[i] !~ mkregexp('[[:alpha:]]')
       before = i >= 0 && i < pos ? string[i].to_s.strip : ''
 
       i = pos + 1
-      i += 1 while i < string.size - 1 && string[i] !~ /[[:alpha:]]/
+      i += 1 while i < string.size - 1 && string[i] !~ mkregexp('[[:alpha:]]')
       after = i > pos ? string[i].to_s.strip : ''
 
       before_uc = !before.empty? && before == before.upcase
@@ -159,5 +161,6 @@ module Interscript
       # before_uc && (after.empty? || after_uc) || after_uc && (before.empty? || before_uc)
       before_uc || after_uc
     end
+
   end
 end

data/lib/interscript/command.rb

@@ -1,16 +1,17 @@
 require 'thor'
 require 'interscript'
-
+require 'json'
 module Interscript
   # Command line interface
   class Command < Thor
     desc '<file>', 'Transliterate text'
     option :system, aliases: '-s', required: true, desc: 'Transliteration system'
     option :output, aliases: '-o', required: false, desc: 'Output file'
+    option :map, aliases: '-m', required: false, default: "{}", desc: 'Transliteration mapping json'
 
     def translit(input)
       if options[:output]
-        Interscript.transliterate_file(options[:system], input, options[:output])
+        Interscript.transliterate_file(options[:system], input, options[:output], JSON.parse(options[:map]))
       else
         puts Interscript.transliterate(options[:system], IO.read(input))
       end

data/lib/interscript/fs.rb

@@ -0,0 +1,96 @@
+require 'pathname'
+
+module Interscript
+  module Fs
+    def sub_replace(string, pos, size, repl)
+      string[pos..pos + size - 1] = repl
+      string
+    end
+
+    def root_path
+      @root_path ||= Pathname.new(File.join(File.dirname(__dir__), ".."))
+    end
+
+    def transliterate_file(system_code, input_file, output_file, maps={})
+      input = File.read(input_file)
+      output = transliterate(system_code, input, maps)
+
+      File.open(output_file, 'w') do |f|
+        f.puts(output)
+      end
+
+      puts "Output written to: #{output_file}"
+      output_file
+    end
+
+    def import_python_modules
+      begin
+        pyimport :g2pwrapper
+      rescue
+        pyimport :sys
+        sys.path.append(root_path.to_s + "/lib/")
+        pyimport :g2pwrapper
+      end
+    end
+
+    def external_process(process_name, string)
+      import_python_modules
+
+      case process_name
+      when 'sequitur.pythainlp_lexicon'
+        return g2pwrapper.transliterate('pythainlp_lexicon', string)
+      when 'sequitur.wiktionary_phonemic'
+        return g2pwrapper.transliterate('wiktionary_phonemic', string)
+      else
+        raise ExternalProcessNotRecognizedError.new
+      end
+
+    rescue
+      raise ExternalProcessUnavailableError.new
+    end
+
+    def external_processing(mapping, string)
+      # Segmentation
+      string = external_process(mapping.segmentation, string) if mapping.segmentation
+
+      # Transliteration/Transcription
+      string = external_process(mapping.transcription, string) if mapping.transcription
+
+      string
+    end
+
+    def aliases (refresh: false)
+      file = root_path.join("./aliases.json").to_s
+      if !refresh && File.exist?(file)
+        JSON.load(File.read(file))
+      elsif !refresh && @aliases
+        @aliases
+      else
+        @aliases = {}
+        Dir[root_path.join('./maps/*.yaml').to_s].each do |yaml_file|
+          org_name = File.basename(yaml_file, ".yaml")
+          map = YAML.load_file(yaml_file)
+          (map["alias"] || {}).each do |k,v|
+            @aliases[v["code"]] = org_name
+          end
+        end
+
+        # Try to save it to a file, but not force it.
+        File.write("aliases.json", JSON.dump(@aliases)) rescue nil
+
+        @aliases
+      end
+    end
+
+    private
+
+    def map_exist?(map)
+      File.exist?(root_path.join("./maps/" + map + ".yaml").to_s)
+    end
+
+    def mkregexp(regexpstring)
+      /#{regexpstring}/u
+    end
+
+  end
+end

data/lib/interscript/mapping.rb

@@ -1,7 +1,8 @@
 require 'rambling-trie'
+require 'yaml' unless RUBY_ENGINE == 'opal'
+require 'json'
 
 module Interscript
-  class InvalidSystemError < StandardError; end
 
   class Mapping
     attr_reader(
@@ -35,7 +36,10 @@ module Interscript
     def initialize(system_code, options = {})
       @system_code = system_code
       @depth = options.fetch(:depth, 0).to_i
-      @system_path = options.fetch(:system_code, default_path)
+
+      unless RUBY_ENGINE == 'opal'
+        @system_path = options.fetch(:system_code, default_path)
+      end
 
       load_and_serialize_system_mappings
     end
@@ -45,10 +49,10 @@ module Interscript
     end
 
     def load_and_serialize_system_mappings
-      if depth < 5
-        mappings = load_system_mappings
-        serialize_system_mappings(mappings)
-      end
+      return if depth >= 5
+
+      mappings = load_system_mappings
+      serialize_system_mappings(mappings)
     end
 
     private
@@ -64,6 +68,18 @@ module Interscript
     end
 
     def load_system_mappings
+      if RUBY_ENGINE == 'opal'
+        load_opal_mappings
+      else
+        load_fs_mappings
+      end
+    end
+
+    def load_opal_mappings
+      JSON.parse(`Opal.global.InterscriptMaps[#{system_code}]`)
+    end
+
+    def load_fs_mappings
       YAML.load_file(system_path.join(system_code_file))
     rescue Errno::ENOENT
       raise Interscript::InvalidSystemError.new("No system mappings found")
@@ -100,24 +116,27 @@ module Interscript
 
     def include_inherited_mappings(mappings)
       inherit_systems = [].push(mappings["map"]["inherit"]).flatten
-      for inherit_system in inherit_systems do
-        if (inherit_system)
-          inherited_mapping = Mapping.for(inherit_system, depth: depth + 1)
-
-          @rules = [inherited_mapping.rules, rules].flatten
-          @postrules = [inherited_mapping.postrules, postrules].flatten
-          @characters = (inherited_mapping.characters|| {}).merge(characters)
-          @dictionary = (inherited_mapping.dictionary|| {}).merge(dictionary)
-        end
+
+      inherit_systems.each do |inherit_system|
+        next unless inherit_system
+
+        inherited_mapping = Mapping.for(inherit_system, depth: depth + 1)
+
+        @rules = [rules, inherited_mapping.rules].flatten
+        @postrules = [inherited_mapping.postrules, postrules].flatten
+        @characters = (inherited_mapping.characters|| {}).merge(characters)
+        @dictionary = (inherited_mapping.dictionary|| {}).merge(dictionary)
       end
+
+      @characters.compact! # the feature to ignore characters from inherited
     end
 
-    def build_hashes()
+    def build_hashes
       @characters_hash = characters&.sort_by { |k, _v| k.size }&.reverse&.to_h
       @dictionary_hash = dictionary&.sort_by { |k, _v| k.size }&.reverse&.to_h
     end
 
-    def build_trie()
+    def build_trie
       @dictionary_trie = Rambling::Trie.create
       dictionary_trie.concat dictionary.keys
     end