interscript 0.1.6 → 0.1.7

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 7557ea1ca381562c61be7dbbeab2ea4adc42ef57ce857ef86acd62a08e5ce588
-  data.tar.gz: 5099d4a7bf07817155d620db716af452607b8565f4e08d37aa805f486556e0e1
+  metadata.gz: 82fae2b248d9c86139b7f188da2ac72699696c9768e54a4510a6f1af2b933dc9
+  data.tar.gz: 3e2cc24b8d33f5a8ed0f8b475e4d109049439221274ecda9ee1b9c7743896e07
 SHA512:
-  metadata.gz: f8a738a34aba269c0a01b4d123f01cc423a3c296541f302f65a6043bd2b618ef2c702f32d469866ee7c488624985b45ed2eee2ad9bbe2955f230b5b45472c364
-  data.tar.gz: a3a66af7fcb9d8c82bcf927b17ea68686298bccec2f6e8dfac25796e7013b36f68fa9f130bd2fb9e7bc214a528896cafdd41a2f464886f09c504f376c19065d8
+  metadata.gz: 5f9925a97d17f0433446a898f63d18869a73e92f9975d8259c916dae242fc5b15ff93dd6d3c28ca2ff5bcbda29489d4ea59af26d4e16f2e1416d701354a6f6e2
+  data.tar.gz: 77321c4a1001cabda8cc057950682037b637176fd05637ce78a6ea698d8c55e238e6bbcc850dda97828c719afbdae1192a8e7a93fd2307a9d52196342fb015f5

data/lib/interscript.rb

@@ -90,7 +90,8 @@ module Interscript
       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
 
@@ -118,12 +119,15 @@ module Interscript
 
       return unless output
 
-      output = output.sub(/^(.)/, &:upcase) if title_case
+      re = mkregexp('^(.)')
+      output = output.gsub(re, &:upcase) if title_case
       if word_separator != ''
-        output = output.gsub(/#{word_separator}#{separator}/u, word_separator)
+        re = mkregexp("#{word_separator}#{separator}")
+        output = output.gsub(re, word_separator)
 
         if title_case
-          output = output.gsub(/#{word_separator}(.)/u, &:upcase)
+          re = mkregexp("#{word_separator}(.)")
+          output = output.gsub(re, &:upcase)
         end
       end
 
@@ -140,11 +144,11 @@ module Interscript
       return false if string[pos] == string[pos].downcase
 
       i = pos - 1
-      i -= 1 while i.positive? && string[i] !~ Regexp.new(ALPHA_REGEXP)
+      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] !~ Regexp.new(ALPHA_REGEXP)
+      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

data/lib/interscript/fs.rb

@@ -2,8 +2,6 @@ require 'pathname'
 
 module Interscript
   module Fs
-    ALPHA_REGEXP = '[[:alpha:]]'
-
     def sub_replace(string, pos, size, repl)
       string[pos..pos + size - 1] = repl
       string

data/lib/interscript/mapping.rb

@@ -122,7 +122,7 @@ module Interscript
 
         inherited_mapping = Mapping.for(inherit_system, depth: depth + 1)
 
-        @rules = [inherited_mapping.rules, rules].flatten
+        @rules = [rules, inherited_mapping.rules].flatten
         @postrules = [inherited_mapping.postrules, postrules].flatten
         @characters = (inherited_mapping.characters|| {}).merge(characters)
         @dictionary = (inherited_mapping.dictionary|| {}).merge(dictionary)

data/lib/interscript/opal.rb

@@ -1,14 +1,34 @@
+require "onigmo"
+require "onigmo/core_ext"
+
+# Increase this if there are out-of-memory errors. This setting is
+# tested to be big enough to handle all the maps provided.
+Onigmo::FFI.library.memory.grow(128)
+
 module Interscript
   module Opal
-    ALPHA_REGEXP = '\p{L}'
-
     def mkregexp(regexpstring)
-      flags = 'u'
-      if regexpstring.include? "(?i)"
-        regexpstring = regexpstring.gsub("(?i)", "").gsub("(?-i)", "")
-        flags = 'ui'
+      @cache ||= {}
+      if s = @cache[regexpstring]
+        s
+      else
+        # JS regexp is more performant than Onigmo. Let's use the JS
+        # regexp wherever possible, but use Onigmo where we must.
+        # Let's allow those characters to happen for the regexp to be
+        # considered compatible: ()|.*+?{} ** BUT NOT (? **.
+        if /[\\$^\[\]]|\(\?/.match?(regexpstring)
+          # Ruby caches its regexps internally. We can't GC. We could
+          # think about freeing them, but we really can't, because they
+          # may be in use.
+
+          # Uncomment those to keep track of Onigmo/JS regexp compilation.
+          # print '#'
+          @cache[regexpstring] = Onigmo::Regexp.new(regexpstring)
+        else
+          # print '.'
+          @cache[regexpstring] = Regexp.new(regexpstring)
+        end
       end
-      Regexp.new("/#{regexpstring}/#{flags}")
     end
 
     def sub_replace(string, pos, size, repl)
@@ -19,9 +39,19 @@ module Interscript
       string
     end
 
+    # name is unused
     def load_map_json(name, json)
-      `Opal.global.InterscriptMaps[#{name}] = #{json}`
+      JSON.load(json).each do |k,v|
+        `Opal.global.InterscriptMaps[#{k}] = #{JSON.dump(v)}`
+      end
     end
 
   end
 end
+
+class String
+  # Opal has a wrong implementation of String#unicode_normalize
+  def unicode_normalize
+    self.JS.normalize
+  end
+end

data/lib/interscript/opal/entrypoint.rb

@@ -0,0 +1,12 @@
+require "opal"
+require "onigmo/onigmo-wasm"
+
+module Interscript
+  def self.on_load(&block)
+    WebAssembly.wait_for("onigmo/onigmo-wasm", &block)
+  end
+end
+
+Interscript.on_load do
+  require "interscript"
+end

data/lib/interscript/opal/map_translate.rb

@@ -0,0 +1,7 @@
+module Interscript
+  module OpalMapTranslate
+    def self.translate_regexp(src)
+      src
+    end
+  end
+end

data/lib/interscript/version.rb

@@ -1,3 +1,3 @@
 module Interscript
-  VERSION = "0.1.6"
+  VERSION = "0.1.7"
 end

data/maps/acadsin-zho-Hani-Latn-2002.yaml

@@ -1,10 +1,14 @@
 ---
 authority_id: acadsin
 id: 2002
-language: zho
+language: iso-639-2:zho
 source_script: Hani
 destination_script: Latn
 name: Chinese Tongyong Pinyin Academica Sinica 2002 System
+alias:
+  ogc11122:
+    code: zho_Hani2Latn_AcadSin_2002
+    description: Chinese Tongyong Pinyin Academica Sinica 2002 System
 url:
 description: Chinese Tongyong Pinyin Academica Sinica 2002 System
 

data/maps/alalc-amh-Ethi-Latn-1997.yaml

@@ -1,10 +1,14 @@
 ---
 authority_id: alalc
 id: 1997
-language: amh
+language: iso-639-2:amh
 source_script: Ethi
 destination_script: Latn
 name: ALA-LC Romanization Table -- Amharic (1997)
+alias:
+  ogc11122:
+    code: amh_Ethi2Latn_ALA_1997
+    description: Amharic ALA-Library of Congress 1997 System
 url: http://catdir.loc.gov/catdir/cpso/romanization/amharic.pdf
 creation_date: 1997
 description: |

data/maps/alalc-amh-Ethi-Latn-2011.yaml

@@ -1,7 +1,7 @@
 ---
 authority_id: alalc
 id: 2011
-language: amh
+language: iso-639-2:amh
 source_script: Ethi
 destination_script: Latn
 name: ALA-LC Romanization Table -- Amharic (2011)

data/maps/alalc-ara-Arab-Latn-1997.yaml

@@ -1,10 +1,14 @@
 ---
 authority_id: alalc
 id: 1997
-language: ara
+language: iso-639-2:ara
 source_script: Arab
 destination_script: Latn
 name: ALA-LC Romanization Table -- Arabic (1997)
+alias:
+  ogc11122:
+    code: ara_Arab2Latn_ALA_1997
+    description: Arabic ALA-Library of Congress 1997 System
 url: http://catdir.loc.gov/catdir/cpso/romanization/arabic.pdf
 creation_date: 1997
 description: |

data/maps/alalc-asm-Deva-Latn-1997.yaml

@@ -1,11 +1,15 @@
 ---
 authority_id: alalc
 id: 1997
-language: asm
+language: iso-639-2:asm
 source_script: Deva
 destination_script: Latn
 name: Assamese Romanization, 1997
-url: https://www.loc.gov/catdir/cpso/romanization/assamese.pdf
+alias:
+  ogc11122:
+    code: asm_Deva2Latn_ALA_1997
+    description: Assamese ALA-Library of Congress 1997 System
+url: http://catdir.loc.gov/catdir/cpso/romanization/assamese.pdf
 creation_date: 1997
 description: |
   ALA-LC Romanization table for Assamese
@@ -47,7 +51,7 @@ tests:
   - source: "কমুম্বাইৰ মেয়ৰৰ দেহত কোভিড পজিটিভ"
     expected: "kamaumabaāira maeẏarara daehata kaobhaiḍa pajaiṭaibha"
   - source: "টুইটাৰযোগে খোদ সদৰী কৰে এই কথা"
-    expected: "ṭauiṭaāraযogae khaoda sadaraī karae ei kathaā"
+    expected: "ṭauiṭaāraẏaogae khaoda sadaraī karae ei kathaā"
   - source: "লখিমপুৰ জিলাৰ নাৰায়ণপুৰৰ বৰপথাৰত আজি প্ৰশান্তি ধাম নামেৰে এখন বৃদ্ধাশ্ৰমৰ শুভাৰম্ভ কৰা হয়"
     expected: "lakhaimapaura jailaāra naāraāẏaṇapaurara barapathaārata ājai paraśaānatai dhaāma naāmaerae ekhana baṛdadhaāśaramara śaubhaāramabha karaā haẏa"
 
@@ -119,6 +123,7 @@ map:
 
     #Semivowels
     'য়': 'ya'
+    'য': 'ẏa'
     'য়': 'ẏa'
     'ৰ': 'ra'
     'ল': 'la'
@@ -156,4 +161,5 @@ map:
     '\u09c8': 'ai'
     '\u09cb': 'o'
     '\u09cc': 'au'
+    '।': '.'
     '\u09CD': '' # Used for joining

data/maps/alalc-asm-Deva-Latn-2012.yaml

@@ -0,0 +1,40 @@
+---
+authority_id: alalc
+id: 2012
+language: iso-639-2:asm
+source_script: Deva
+destination_script: Latn
+name: Assamese Romanization, 2012
+url: https://www.loc.gov/catdir/cpso/romanization/assamese.pdf
+creation_date: 1997
+description: |
+  ALA-LC Romanization table for Assamese
+
+notes:
+
+  - Only the vowel forms that appear at the beginning of a syllable are listed; the forms used for
+    vowels following a consonant can be found in grammars; no distinction between the two is
+    made in transliteration.
+
+  - |
+    The vowel a is implicit after all consonants and consonant clusters and is supplied in
+    transliteration, with the following exceptions:
+
+    a) when another vowel is indicated by its appropriate sign; and
+    b) when the absence of any vowel is indicated by the subscript sign ( ◌् ) called halanta or
+       birāma.
+
+  - Candrabindu before guttural, palatal, cerebral, and dental occlusives is transliterated n̐. Before
+    labials, sibilants, semivowels, aspirates, vowels, and in final position it is transliterated m̐.
+
+  - When doubled, avagraha is transliterated by two apostrophes ( ’’ ).
+
+tests:
+  - source: "ৰাজ্যিক স্বাস্থ্য মন্ত্ৰী পীয়ুষ হাজৰিকাৰ বিৰুদ্ধে দাখিল কৰা হৈছে এজাহাৰ।"
+    expected: "raājaẏaika sabaāsathaẏa manataraī paīyausha haājaraikaāra bairaudadhae daākhaila karaā haaichae ejaāhaāra."
+  - source: "কোৰোনা মহামাৰীৰ এই সময়ত সভাখনত হাজাৰ হাজাৰ লোকে মাস্ক পৰিধান নকৰাৰ লগতে সামাজিক দূৰত্ব নমনাৰ অভিযোগ উত্থাপন কৰা হৈছে"
+    expected: "kaoraonaā mahaāmaāraīra ei samayata sabhaākhanata haājaāra haājaāra laokae maāsaka paraidhaāna nakaraāra lagatae saāmaājaika daūrataba namanaāra abhaiẏaoga utathaāpana karaā haaichae"
+
+map:
+
+  inherit: "alalc-asm-Deva-Latn-1997"

data/maps/alalc-aze-Cyrl-Latn-1997.yaml

@@ -1,10 +1,14 @@
 ---
 authority_id: alalc
 id: 1997
-language: aze
+language: iso-639-2:aze
 source_script: Cyrl
 destination_script: Latn
 name: ALA-LC Romanization Table -- Azerbaijani (1997)
+alias:
+  ogc11122:
+    code: aze_Cyrl2Latn_ALA_1997
+    description: Azerbaijani ALA-Library of Congress 1997 System
 url: https://transliteration.eki.ee/pdf/Azerbaijani.pdf
 creation_date: 1997
 description: |

data/maps/alalc-bel-Cyrl-Latn-1997.yaml

@@ -1,10 +1,14 @@
 ---
 authority_id: alalc
 id: 1997
-language: bel
+language: iso-639-2:bel
 source_script: Cyrl
 destination_script: Latn
 name: ALA-LC Romanization Table -- Byelorussian (1997)
+alias:
+  ogc11122:
+    code: bel_Cyrl2Latn_ALA_1997
+    description: Byelorussian ALA-Library of Congress 1997 System
 url: http://catdir.loc.gov/catdir/cpso/romanization/beloruss.pdf
 creation_date: 1997
 

data/maps/alalc-ben-Beng-Latn-2017.yaml

@@ -1,7 +1,7 @@
 ---
 authority_id: alalc
 id: 2017
-language: ben
+language: iso-639-2:ben
 source_script: Beng
 destination_script: Latn
 name: Bengali Romanization, 2017

data/maps/alalc-bul-Cyrl-Latn-1997.yaml

@@ -1,10 +1,14 @@
 ---
 authority_id: alalc
 id: 1997
-language: bul
+language: iso-639-2:bul
 source_script: Cyrl
 destination_script: Latn
 name: ALA-LC Romanization Table -- Bulgarian (1997)
+alias:
+  ogc11122:
+    code: bul_Cyrl2Latn_ALA_1997
+    description: Bulgarian ALA-Library of Congress Bulgarian 1997 System
 url: http://www.rechtertie.nl/databases/judd/downloads/Bulgarian.pdf
 creation_date: 1997
 

data/maps/alalc-ell-Grek-Latn-1997.yaml

@@ -1,10 +1,14 @@
 ---
 authority_id: alalc
 id: 1997
-language: ell
+language: iso-639-2:ell
 source_script: Grek
 destination_script: Latn
 name: Greek Romanization, 1997
+alias:
+  ogc11122:
+    code: ell_Grek2Latn_ALA_1997
+    description: Greek ALA-Library of Congress 1997 System
 url: http://catdir.loc.gov/catdir/cpso/romanization/greek.pdf
 creation_date: 1997
 description: |

data/maps/alalc-ell-Grek-Latn-2010.yaml

@@ -1,7 +1,7 @@
 ---
 authority_id: alalc
 id: 2010
-language: ell
+language: iso-639-2:ell
 source_script: Grek
 destination_script: Latn
 name: Greek Romanization, 2010
@@ -27,7 +27,6 @@ tests:
 
         Giannēs Makrygiannēs.
 
-
   - source: ΑΘΗΝΑ
     expected: ATHĒNA
   - source: μπαμπάκι

data/maps/alalc-guj-Gujr-Latn-1997.yaml

@@ -0,0 +1,266 @@
+---
+authority_id: alalc
+id: 1997
+language: iso-639-2:guj
+source_script: Gujr
+destination_script: Latn
+name: Gujarati Romanization, 1997
+alias:
+  ogc11122:
+    code: guj_Gujr2Latn_ALA_1997
+    description: Gujarati ALA-Library of Congress 1997 System
+url: http://catdir.loc.gov/catdir/cpso/romanization/gujarati.pdf
+creation_date: 1997
+description: |
+  ALA-LC Romanization table for Gujarati
+
+notes:
+
+  - Only the vowel forms that appear at the beginning of a syllable are listed; the forms used for
+    vowels following a consonant can be found in grammars; no distinction between the two is
+    made in transliteration.
+
+  - |
+    The vowel a is implicit after all consonants and consonant clusters and is supplied in
+    transliteration, with the following exceptions:
+    a) when another vowel is indicated by its appropriate sign and
+    b) when the absence of any vowel is indicated by the subscript symbol ( ્ ) called halanta or
+    virāma.
+
+  - |
+    Exception: Anusvāra is transliterated by:
+    a) ṅ before gutturals,
+    b) ñ before palatals,
+    c) ṇ before cerebrals,
+    d) n before dentals, and
+    e) m before labials.
+
+  - When doubled, avagraha is transliterated by two apostrophes ( ’’ ).
+
+tests:
+  - source: "અમિત શાહનો કોરોના રિપોર્ટ ૨ ઓગસ્ટે પોઝિટિવ આવ્યો હતો, ત્યારથી તેમનું સ્વાસ્થ્ય સારું નથી"
+    expected: "amita śāhanȏ kȏrȏnā ripȏrṭa 2 ȏgasṭȇ pȏjhiṭiva āvyȏ hatȏ, tyārathī tȇmanuṃ svāsthya sāruṃ nathī"
+  - source: "મેદાંતા હોસ્પિટલમાં તેમનો ઇલાજ ચાલી રહ્યો હતો"
+    expected: "mȇdāntā hȏspiṭalamāṃ tȇmanȏ ilāja cālī rahyȏ hatȏ"
+  - source: "ભારતના વિશ્વનાથન આનંદે શેનયાનમાં પહેલો ફિડે શતરંજ વિશ્વ કપ જીત્યો"
+    expected: "bhāratanā viśvanāthana ānandȇ śȇnayānamāṃ pahȇlȏ phiḍȇ śatarañja viśva kapa jītyȏ"
+  - source: "ભારતીય વડા પ્રધાન જવાહરલાલ નેહરુએ ૪૦ લાખ હિન્દુઓ અને મુસલમાનોના પારસ્પરિક સ્થાનાંતરણનું સૂચન આપ્યું"
+    expected: "bhāratīya vaḍā pradhāna javāharalāla nȇharuȇ 40 lākha hinduȏ anȇ musalamānȏnā pārasparika sthānāntaraṇanuṃ sūcana āpyuṃ"
+  - source: "લિબિયાના એલ અજિજિયામાં ધરતી પર સૌથી વધુ તાપમાન નોંધાયું. એ વખતે છાયામાં નોંધવામાં આવેલું તાપમાન ૫૮ ડિગ્રી સેલ્સિયસ હતું."
+    expected: "libiyānā ȇla ajijiyāmāṃ dharatī para sauthī vadhu tāpamāna nȏndhāyuṃ. ȇ vakhatȇ chāyāmāṃ nȏndhavāmāṃ āvȇluṃ tāpamāna 58 ḍigrī sȇlsiyasa hatuṃ."
+  - source: "પ્રથમ વિશ્વયુદ્ધઃ જર્મની અને ફ્રાન્સ વચ્ચે એસ્નેની લડાઈ શરૂ થઈ હતી"
+    expected: "prathama viśvayuddhaḥ jarmanī anȇ phrānsa vaccȇ ȇsnȇnī laḍāī śarū thaī hatī"
+  - source: "એન્ગ્લો-મિસ્ત્ર યુદ્ધઃ તેલ અલ કેબિરનું યુદ્ધ લડવામાં આવ્યું હતું."
+    expected: "ȇnglȏ-mistra yuddhaḥ tȇla ala kȇbiranuṃ yuddha laḍavāmāṃ āvyuṃ hatuṃ."
+  - source: "પુરાવા ન હતા, એ જ કારણે કેસ ચાલ્યો નહીં, પણ તેમને નજરકેદ રાખવામાં આવ્યા"
+    expected: "purāvā na hatā, ȇ ja kāraṇȇ kȇsa cālyȏ nahīṃ, paṇa tȇmanȇ najarakȇda rākhavāmāṃ āvyā"
+  - source: "સરદાર પટેલે નક્કી કર્યું હતું કે કાશ્મીર ભારતનો હિસ્સો બનશે; ૯૧ વર્ષ પહેલાં લાહોર જેલમાં ભૂખહડતાળ દરમિયાન શહીદ થયા હતા જતીન દાસ"
+    expected: "saradāra paṭȇlȇ nakkī karyuṃ hatuṃ kȇ kāśmīra bhāratanȏ hissȏ banaśȇ; 91 varsha pahȇlāṃ lāhȏra jȇlamāṃ bhūkhahaḍatāḷa daramiyāna śahīda thayā hatā jatīna dāsa"
+  - source: "કોરોના પ્રોટોકોલ વચ્ચે આજે મેડિકલ પ્રવેશ પરીક્ષા લેવાશેઃ એન્ટ્રી ટચ ફ્રી રહેશે, એડમિટ કાર્ડ બાર કોડથી ચેક થશે"
+    expected: "kȏrȏnā prȏṭȏkȏla vaccȇ ājȇ mȇḍikala pravȇśa parīkshā lȇvāśȇḥ ȇnṭrī ṭaca phrī rahȇśȇ, ȇḍamiṭa kārḍa bāra kȏḍathī cȇka thaśȇ"
+  - source: "અલ્ ક઼`ઇદ્ માં હવામાન"
+    expected: "al ka`id māṃ havāmāna"
+  - source: "મંત્રાલય તથા ખ઼.ય ના વિ૨ષ્ઠ અધિકા૨ીઓ ઉપસ્થિત ૨હ્યા હતા"
+    expected: "mantrālaya tathā kha.ya nā vi2shṭha adhikā2īȏ upasthita 2hyā hatā"
+
+
+map:
+
+  rules:
+    # note 3
+    - pattern: \u0A82(?=[કખગઘઙ])
+      result: ṅ
+    - pattern: \u0A82(?=[ચછજઝઞ])
+      result: ñ
+    - pattern: \u0A82(?=[ટઠડઢણ])
+      result: ṇ
+    - pattern: \u0A82(?=[તથદધન])
+      result: n
+    - pattern: \u0A82(?=[પફબભમ])
+      result: m
+
+    # note[2(a,b)]
+    - pattern: ([ક]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result: 'k'    
+    - pattern: ([ખ]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result: 'kh'   
+    - pattern: ([ગ]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result: 'g'    
+    - pattern: ([ઘ]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result: 'gh'   
+    - pattern: ([ઙ]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result: 'ṅ'   
+    - pattern: ([ચ]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result: 'c'    
+    - pattern: ([છ]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result: 'ch'   
+    - pattern: ([જ]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result: 'j'    
+    - pattern: ([ઝ]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result: 'jh'   
+    - pattern: ([ઞ]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result: 'ñ'   
+    - pattern: ([ટ]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result:  'ṭ'  
+    - pattern: ([ઠ]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result:  'ṭh' 
+    - pattern: ([ડ]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result:  'ḍ'  
+    - pattern: ([ઢ]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result:  'ḍh' 
+    - pattern: ([ણ]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result:  'ṇ'  
+    - pattern: ([ત]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result: 't'    
+    - pattern: ([થ]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result: 'th'   
+    - pattern: ([દ]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result: 'd'    
+    - pattern: ([ધ]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result: 'dh'   
+    - pattern: ([ન]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result: 'n'    
+    - pattern: ([પ]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result: 'p'    
+    - pattern: ([ફ]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result: 'ph'   
+    - pattern: ([બ]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result: 'b'    
+    - pattern: ([ભ]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result: 'bh'   
+    - pattern: ([મ]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result: 'm'    
+    - pattern: ([ય]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result: 'y'    
+    - pattern: ([ર]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result: 'r'    
+    - pattern: ([લ]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result: 'l'    
+    - pattern: ([ળ]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result: 'ḷ'    
+    - pattern: ([વ]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result: 'v'    
+    - pattern: ([શ]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result: 'ś'   
+    - pattern: ([ષ]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result: 'sh'   
+    - pattern: ([સ]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result: 's'    
+    - pattern: ([હ]=?)(?=[\u0abe\u0abf\u0ac0\u0ac1\u0ac2\u0ac3\u0ac5\u0ac7\u0ac8\u0ac9\u0acb\u0acc\u0acd])  
+      result: 'h'    
+
+
+  characters:
+
+    'અ': 'a'
+    'આ': 'ā'
+    'ઇ': 'i'
+    'ઈ': 'ī'
+    'ઉ': 'u'
+    'ઊ': 'ū'
+    'ઋ': 'ṛ'
+
+    'ઍ': 'e'
+    'એ': 'ȇ'
+    'ઐ': 'ai'
+
+    'ઑ': 'o'
+    'ઓ': 'ȏ'
+    'ઔ': 'au'
+
+    # II. Consonants (see Note 2)
+    # Gutturals
+    'ક': 'ka'
+    'ખ': 'kha'
+    'ગ': 'ga'
+    'ઘ': 'gha'
+    'ઙ': 'ṅa'
+
+    # Palatals
+    'ચ': 'ca'
+    'છ': 'cha'
+    'જ': 'ja'
+    'ઝ': 'jha'
+    'ઞ': 'ña'
+
+    # Cerebrals
+    'ટ':  'ṭa'
+    'ઠ':  'ṭha'
+    'ડ':  'ḍa'
+    'ઢ':  'ḍha'
+    'ણ':  'ṇa'
+
+    # Dentals
+    'ત': 'ta'
+    'થ': 'tha'
+    'દ': 'da'
+    'ધ': 'dha'
+    'ન': 'na'
+
+    # Labials
+    'પ': 'pa'
+    'ફ': 'pha'
+    'બ': 'ba'
+    'ભ': 'bha'
+    'મ': 'ma'
+
+    # Semivowels
+    'ય': 'ya'
+    'ર': 'ra'
+    'લ': 'la'
+    'ળ': 'ḷa'
+    'વ': 'va'
+
+    # Sibilants
+    'શ': 'śa'
+    'ષ': 'sha'
+    'સ': 'sa'
+
+
+    # Aspirate
+    'હ': 'ha'
+
+    # Bisarga
+    'ઃ': 'ḥ'
+
+    # Anusvāra
+    'ં': 'ṃ'
+
+    # Abagraha  (see Note 4)
+    'ઽ': '’' # (apostrophe)
+
+    # Medials # Needed for connecting constants
+
+    'ા': 'ā'
+    'િ': 'i'
+    'ી': 'ī'
+    'ુ': 'u'
+    'ૂ': 'ū'
+    'ૃ': 'ṛ'
+    'ૅ': 'e'
+    'ે': 'ȇ'
+    'ૈ': 'ai'
+    'ૉ': 'o'
+    'ો': 'ȏ'
+    'ૌ': 'au'
+
+    # digits
+
+    '૦': '0'
+    '૧': '1'
+    '૨': '2'
+    '૩': '3'
+    '૪': '4'
+    '૫': '5'
+    '૬': '6'
+    '૭': '7'
+    '૮': '8'
+    '૯': '9'
+
+
+
+    '્': ''
+    '઼': ''
+    '।': '.'
+    '\u09CD': '' # Used for joining
+    "‍": ''# Used for joining